OV9623 Preliminary Specification CSP - Version 1 23 - Waching
OV9623 Preliminary Specification CSP - Version 1 23 - Waching
on ac
W
fid in
en g o
h
tia nly
lf
or
OV9623
datasheet
PRELIMINARY SPECIFICATION
1/2.7" color CMOS WXGA (1280 x 800) high dynamic range (HDR)
high definition (HD) image sensor
OmniVision Technologies, Inc. and all its affiliates disclaim all liability, including liability for infringement of any proprietary
rights, relating to the use of information in this document. No license, expressed or implied, by estoppel or otherwise, to
any intellectual property rights is granted herein.
The information contained in this document is considered proprietary to OmniVision Technologies, Inc. and all its
affiliates. This information may be distributed to individuals or organizations authorized by OmniVision Technologies, Inc.
to receive said information. Individuals and/or organizations are not allowed to re-distribute said information.
C
on ac
Trademark Information
OmniVision and the OmniVision logo are registered trademarks of OmniVision Technologies, Inc. OmniPixel3-HS is a
trademark of OmniVision Technologies, Inc.
W
fid in
All other trademarks used herein are the property of their respective owners.
en g o
h
tia nly
lf
or
color CMOS WXGA (1280x800) high dynamic range (HDR) high definition (HD) image sensor
datasheet (CSP)
PRELIMINARY SPECIFICATION
version 1.23
december 2019
00features
C
support for image sizes: WXGA (1280x800), auto white balance control
HD 720p (1280x720), WVGA (752x480),
aperture/gamma correction
VGA (640x480), 600x400, CIF (352x288),
on ac
parallel DVP interface external frame sync capability provide the best image
high sensitivity 50/60 Hz flicker cancellation quality, OmniVision
recommends an IR cut
automatic exposure/gain defective pixel correction filter
en g o
tia nly
C
on ac
W
fid in
en g o
h
tia nly
lf
or
00table of contents
7.15 combine [0x5400 ~ 0x542D, 0xC30C ~ 0xC4CB, 0x5A08 ~ 0x5A97, 0x5C18 ~ 0x5C6F] 7-86
7.16 normalize (NMLZ) control [0x5480 ~ 0x5A98, 0x5C71 ~ 0x5C78] 7-95
en g o
7.17 tone mapping (TMAP) [0x5500 ~ 0x5511, 0xC4E4 ~ 0xC4F9, 0x5A9C ~ 0x5CFB] 7-96
7.18 frame counter (FC) control [0x4200 ~ 0x4203] 7-105
h
7.20 VFIFO control [0x4600 ~ 0x4603, 0x4605 ~ 0x4613, 0x4620 ~ 0x4639] 7-107
7.21 digital video port (DVP) control [0x4700 ~ 0x470D] 7-109
lf
tia nly
lf
or
00list of figures
C
on ac
W
fid in
en g o
h
tia nly
lf
or
00list of tables
1 signal descriptions
table 1-1 lists the signal descriptions and their corresponding pin numbers for the OV9623 image sensor. The package
information is shown in section 9.
tia nly
top view
pin pin
number signal name type description
A2 AVDD power 3.3V power
pin pin
number signal name type description
A4 SVDD power 3.3V power
pin pin
number signal name type description
C7 DOGND ground I/O ground
pin pin
number signal name type description
G2 PWDN input input (active high with pull down resistor)
J2 RESETB input reset input (active low with internal pull up resistor)
h
pin pin
number signal name type description
L4 DOGND ground I/O ground
pin pin
number signal name type description
N7 DVDD power 1.5V power
tia nly
tia nly
lf
or
C
on ac
W
fid in
en g o
h
tia nly
lf
or
The OV9623 color image sensors are low voltage, high performance, 1/2.7-inch, CMOS image sensors that provide the
full functionality of a single chip 1 megapixel (1280x800) camera using OmniPixel3-HS™ technology in a small footprint
package. They provide full-frame, sub-sampled and windowed images in various formats via the control of the Serial
Camera Control Bus (SCCB) interface.
The OV9623 has an image array capable of operating at up to 30 frames per second (fps) in full resolution with complete
user control over image quality, formatting and output data transfer. All required image processing functions, including
exposure control, white balance, defective pixel canceling, etc., are programmable through the SCCB interface. In
addition, OmniVision image sensors use proprietary sensor technology to improve image quality by reducing or
C
eliminating common lighting/electrical sources of image contamination, such as fixed pattern noise, smearing, etc., to
produce a clean, fully stable image.
on ac
2.2 architecture
W
fid in
The OV9623 sensor core generates streaming pixel data at a constant frame rate, indicated by HREF, VSYNC, and
PCLK.
The timing generator outputs clocks to access the rows of the imaging array, precharging and sampling the rows of the
en g o
array sequentially. In the time between precharging and sampling a row, the charge in the pixels decrease with exposure
to incident light. This is the exposure time in rolling shutter architecture.
h
The exposure time is controlled by adjusting the time interval between precharging and sampling. After the data of the
tia nly
pixels in the row has been sampled, it is processed through analog circuitry to correct the offset and multiply the data
with corresponding gain. Following analog processing is the ADC.
lf
or
OV9623
image sensor core image sensor image output
processor interface
column
sample/hold
digital gain
calibration
black level
row select
image
FIFO
DVP
ISP
array AMP ADC D[9:0]
C
gain
control
on ac
PLL timing generator and system control logic SCCB slave interface
en g o
XVCLK
PWDN
RESETB
TMB
GPIO[3:0]
TM
VSYNC
HREF
PCLK
FSIN
SID[2:0]
SIOC
SIOD
h
tia nly
lf
or
AVDD
C15 0.1μF-0603
D5 3 4 D4
D5 D4
C1 0.1μF-0603
D7 5 6 D6
D7 D6
DVDD
C18 0.1μF-0603
D9 7 8 D8
D9 D8
12.17.2019
C2 0.1μF-0603
9 10 PWDN
NC PWDN
B14
B13
B12
B11
B10
B9
B8
B7
B6
B5
B4
B3
B2
B1
A13
A12
A11
A10
A9
A8
A7
A6
A5
A4
A3
A2
C19 0.1μF-0603
11 12 SIOD
figure 2-2
NC SIOD
C3 0.1μF-0603
VF4
HREF 13 14 SIOC
AVDD
AVDD
AVDD
AVDD
SVDD
SVDD
SVDD
HREF SIOC
SGND
SGND
SGND
AGND
AGND
AGND
C20 0.1μF-0603 N13
DEVDD
DOVDD
DOGND
DOGND
DOGND
DOGND
DOGND
DOGND
DOGND
DOGND
DOGND
DOGND
VSYNC 15 16 DVDD
VSYNC GND
C1 N12 HREF
PCLK 17 18 AGND HREF
PCLK GND
T1
C2 N11
19 20 XCLK AVDD DOGND
PWR XCLK
J1CON32A
C3 N10 D0
21 22 AGND D0
PWR GND
C21 0.1μF-0603 C4 N9 D2
L3
FSIN
D1 23 24 D0 DOGND D2
D1 D0
C5 N8
25 26 DOGND DOVDD
NC NC
T2
C22 0.1μF-0603 C6 N7
27 28 DOGND DVDD
NC NC
C7 N6 D5
29 30 DOGND D5
NC NC
C8 N5
GPIO3
31 32 DOGND DOVDD
GND GND
C9 N4
DOGND DOGND
T3
C10 N3 D8
DOGND D8
10μH-L1008
C23 0.1μF-0603 C11 N2
DOGND DVDD
C12 M14 SID1
VF GPIO1/SID1
C24 0.1μF-0603 C13 M13
C9 10μF/16V VH DGND
PWR
+ C14 M12 VSYNC
AGND VSYNC
C25 0.1μF-0603 D1 M11 PCLK
DOGND PCLK
D2 M10
DOGND DOVDD
W
C26 0.1μF-0603 D3 M9 D1
DOGND D1
D4 M8 D3
DOGND D3
C36 0.1μF-0603 C30 0.1μF-0603 DVDD D12 M7
DOGND DGND
C
PWR
U1
D13 M6 D4
DOGND
CSP
D4
1
2
C27 0.1μF-0603 D14 M5 D6
OV9623
AVDD D6
3
2
1
E1 M4 D7
VIN
GND
DOGND D7
OV9623 reference schematic
IN
EN
C28 0.1μF-0603 E2 M3 D9
U2
GND
DOGND D9
E3 M2
U4
OUT
DOGND DOGND
3
C29 0.1μF-0603 E4 M1 SIOD
PRELIMINARY SPECIFICATION
FB
OUT
PWR XC62FP3302-SOT89
C31 10μF/6V-EIA-A DOGND SDA
4
5
ADP123-TSOT23-5
E12 L14
DOGND DOVDD
on ac
E13 L13
h
C32 0.1μF-0603 EVDD DOGND
R8
R7
E14 L12 SID0
AVDD
PVDD GPIO0/SID0
R6 0-0603 F1 L11
TM DOGND
F2 L10
DOGND DOGND
F3 L9
DOGND DOGND
F4 L8
10K-0603
20K-0603
DOGND DOGND
F12 L7
fid in
C33 0.1μF-0603 DOGND DOGND
C37 10μF/6V-EIA-A
XCLK F13 L6
XVCLK DOGND
DVDD
F14 L5
1
2
EGND DOGND
G1 L4
AGND DOGND
VIN
GND
PWDN G2 L3
PWDN DGND
U3
G3 L2 SIOC
SID2
SID1
SID0
DOGND SCL
OUT
L1
3
DOVDD
en g o
PWR XC62FP3302-SOT89
R3
R4
R5
C34 10μF/6V-EIA-A
C4 0.1μF-0603
DOGND
DOGND
DGND
DOVDD
AVDD
DOGND
DOGND
DOGND
DOGND
DOGND
DVDD
TMB
RESETB
DOGND
DOGND
DOGND
GPIO2/SID2
GPIO3
DVDD
DGND
FSIN
DOGND
DOGND
DVDD
DGND
C35 0.1μF-0603
J1
J2
J3
J4
K1
K2
K3
K4
G4
H1
H2
H3
H4
C5 0.1μF-0603
J12
J13
K12
K13
K14
G12
G13
G14
H12
H13
H14
DOVDD
C6 0.1μF-0603
10K-0603
10K-0603
10K-0603
SID2
J14 GPIO3
FSIN
tia nly
C7 0.1μF-0603
C10 0.1μF-0603
C16 0.1μF-0603
C8 0.1μF-0603
C11 0.1μF-0603
DOVDD
DVDD
AVDD
R2
R1
C12 0.1μF-0603
lf
C13 0.1μF-0603
0-0603
15K-0603
C14 0.1μF-0603
or
2-3
OV9623 color CMOS WXGA (1280x800) high dynamic range (HDR) high definition (HD) image sensor
The OV9623 supports 8/10-bit YUV, up to 18-bit combined RAW and separated 10-bit RAW. For further information on
the registers affecting windowing, cropping, and skipping (subsampling), see section 4.3.
The OV9623 I/O pad direction and driving capability can be easily adjusted. table 2-2 lists the driving capability and
direction control registers of the I/O pads.
h
tia nly
table 2-2 driving capability and direction control for I/O pads (sheet 1 of 2)
lf
output drive
0x3011 RW 01: 2x
capability control
10: 3x
11: 4x
0x3008[1:0],
D[9:0] output value RW D[9:0] output value
0x3009[7:0]
table 2-2 driving capability and direction control for I/O pads (sheet 2 of 2)
1: output
The OV9623 has an on-chip PLL which generates the maximum 96 MHz system clock from a 6~27 MHz input clock. A
programmable clock divider is provided to generate different frequencies for the system. PLL adjustment should be
applied while the sensor is in software standby mode to prevent image corruption or unstable performance. Real-time
adjustment is not allowed.
XVCLK 0
/1, /1.5, /2, /3, /4, /5, /6, /7 register 0x3005[5:0] /2(1+register 0x3006[2:0])
tia nly
lf
or
The Serial Camera Control Bus (SCCB) interface controls the image sensor operation. Refer to the OmniVision
Technologies Serial Camera Control Bus (SCCB) Specification for detailed usage of the serial control port.
message type: 16-bit sub-address, 8-bit data, and 7-bit slave address
index[15:8] index[7:0]
en g o
The sub-address in the sensor automatically increases by one after each read/write operation.
In a single read from random locations, the master does a dummy write operation to desired sub-address, issues a
repeated start condition and then addresses the camera again with a read operation. After acknowledging its slave
address, the camera starts to output data onto the SIOD line as shown in figure 2-5. The master terminates the read
operation by setting a negative acknowledge and stop condition.
index value M
If the host addresses the camera with read operation directly without the dummy write operation, the camera responds
by setting the data from last used sub-address to the SIOD line as shown in figure 2-6. The master terminates the read
C
slave slave
S 1 A data A P S 1 A data A P
en g o
address address
h
The sequential read from a random location is illustrated in figure 2-7. The master does a dummy write to the desired
tia nly
sub-address, issues a repeated start condition after acknowledge from slave and addresses the slave again with read
operation. If a master issues an acknowledge after receiving data, it acts as a signal to the slave that the read operation
shall continue from the next sub-address. When master has read the last data byte, it issues a negative acknowledge
lf
The sequential read from current location is similar to a sequential read from a random location. The only exception is
that there is no dummy write operation as shown in figure 2-8. The master terminates the read operation by setting a
negative acknowledge and stop condition.
slave
S 1 A data A data A data A P
address
C
L bytes of data
on ac
The write operation to a random location is illustrated in figure 2-9. The master issues a write operation to the slave, sets
the sub-address and data correspondingly after the slave has acknowledged. The write operation is terminated with a
stop condition from the master.
W
fid in
tia nly
index value M
or
The sequential write is illustrated in figure 2-10. The slave automatically increments the sub-address after each data
byte. The sequential write operation is terminated with stop condition from the master.
tF tHIGH tR
SIOC
tLOW
tSU:DAT
tHD:STA tSU:STO
SIOD (IN)
tAA
C
SIOD (OUT)
on ac
tDH
W
fid in
2.7 standby
To initiate hardware standby mode, the PWDN pin must be tied to high. When this occurs, the OV9623 internal device
clock is halted and all internal counters are reset and registers are maintained.
TDVDD
DOVDD TAVDD
C
TPWDN
DVDD
on ac
AVDD
W
fid in
PWDN
en g o
RESETB
h
tia nly
TRST
lf
SIOC
TSCCB
...
or
SIOD
...
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Turn on video stream after power
0x0100 STREAM MODE 0x00 RW up, always set to "1"
0: Not used
1: Stream on
0x0103 SOFTWARE RESET 0x00 RW Software Reset will Auto Clear by Itself to 0x00
C
on ac
W
fid in
en g o
h
tia nly
lf
or
The color filters are arranged in a Bayer pattern. The primary color BG/GR array is arranged in line-alternating fashion.
Of the 828 rows, 814 rows are active rows and can be output. The other rows are used for black level calibration and
interpolation.
The sensor array design is based on a read-out system with line-by-line transfer and an electronic shutter with a
synchronous pixel read-out scheme.
columns
1306
1307
1308
1309
1310
1311
on ac 0
1
2
3
4
5
0 B Gb B Gb B Gb B Gb B Gb B Gb
1 Gr R Gr R Gr R Gr R Gr R Gr R
2 B Gb B Gb B Gb B Gb B Gb B Gb
W
fid in
3 Gr R Gr R Gr R Gr R Gr R Gr R
dummy
10 B Gb B Gb B Gb B Gb B Gb B Gb
11 Gr R Gr R Gr R Gr R Gr R Gr R
12 B Gb B Gb B Gb B Gb B Gb B Gb
en g o
13 Gr R Gr R Gr R Gr R Gr R Gr R
rows
14 B Gb B Gb B Gb B Gb B Gb B Gb
h
15 Gr R Gr R Gr R Gr R Gr R Gr R
16 B Gb B Gb B Gb B Gb B Gb B Gb
tia nly
17 Gr R Gr R Gr R Gr R Gr R Gr R
active
pixel
824 B Gb B Gb B Gb B Gb B Gb B Gb
825 Gr R Gr R Gr R Gr R Gr R Gr R
lf
826 B Gb B Gb B Gb B Gb B Gb B Gb
827 Gr R Gr R Gr R Gr R Gr R Gr R
or
active
pixel
C
on ac
W
fid in
en g o
h
tia nly
lf
or
The OV9623 provides mirror mode, which reverses the sensor data read-out order horizontally, and flip mode which
reverses it vertically (see figure 4-1).
F F
F F
C
on ac
a. when mirror mode is on, register 0x6900[0] needs to be set to 0x01 in order to ensure correct color output
For testing purposes, the OV9623 offers one type of analog test pattern and three types of digital test patterns. The
analog test pattern is a color bar overlaid on an image, which can be enabled by register 0x370A[2]. The digital test
patterns include color bar, square and random data. The OV9623 also offers two digital effects for the test patterns:
transparent effect and rolling bar effect. The digital test pattern function is enabled by register 0x503D[7] and the pattern
is selected by register 0x503E[1:0].
The digital test pattern passes through the pipeline. To get a consistent output pattern, register 0x5000 must be set to
0x78 to turn off some of the ISP blocks and register 0x3042 must be set to 0xF9 after the software reset and before
enabling the test pattern.
0x503D[5:4]=2'b00 0x503D[5:4]=2'b01
lf
or
4.2.2 square
There are two types of square test patterns: color square and black-white square.
There are two types of random data test patterns: frame-changing and frame-fixed random data.
tia nly
lf
or
default
address register name value R/W description
Bit[2]: Analog color bar enable
0x370A SENSOR REG0A 0x00 RW 0: Disable
1: Enable
H_crop_start H_crop_end
en g o
H_win_off
sensor array vertical output size
tia nly
H_output_size
vertical full size: 814
V_win_off
V_output_size
lf
or
V_crop_end
VTS is adjusted by registers (0x6E42[7:0], 0x6E43[7:0]). The reference initialization settings must be used for these two
registers to be valid.
default
address value R/W description
Bit[4:3]: Horizontal crop mode select
00: Full size
01: Horizontal crop to 768
0x3621 0x03 RW
10: Horizontal crop to 656
11: Not used
Bit[2:0]: Analog delay option
default
address value R/W description
Bit[7:4]: Output format select
0x3: YUV mode
0xF: RAW mode
Others: Not allowed
0x4300 0xF8 RW Bit[3:0]: pix_order_ctrl
1000: YUYV
1001: YVYU
1010: UYVY
1011: VYUY
C
In the OV9623, the exposure/gain control is designed to adjust the weighted frame average to a user defined range. The
weight of each pixel includes three parts: position weight, combination weight and luminance weight. Instead of using
the whole frame, the statistic window can be defined manually with the left-top corner {0x5601[2:0], 0x5602},
W
fid in
{0x5603[1:0], 0x5604}, width {0x5605[2:0], 0x5606} and height {0x5607[1:0], 0x5608}. The pixels outside of the window
will not be included in the weighted average.
There are three target modes: AA, AB and ABC mode. This is defined by the target mode register 0xC450[1:0].
en g o
The AEC/AGC algorithms support HDR mode and non-HDR mode. Register 0xC454[0] must be set to 1 for non-HDR
h
In non-HDR mode, the sensor only uses one sub-pixel. In HDR mode, the AEC/AGC needs to determine the exposure
and the gain for the two sub-pixels. It supports auto ratio mode, fixed ratio mode and geometric proportion mode.
lf
Auto ratio mode means the long exposure/short exposure ratio changes automatically according to the scene. It supports
all modes (AA,AB and ABC). To enable auto ratio mode, the fixed ratio mode register 0xC456[0] and the geometric
proportion mode register 0xC457[0] must be set to 0.
or
Fixed ratio mode means the long exposure/short exposure ratio is fixed regardless of the scene. It supports all modes
(AA, AB and ABC). To enable fixed ratio mode, register 0xC456[0] should be set to 1. The fixed ratio can be set by
register 0xC490. In this mode, the geometric proportion mode register 0xC457[0] should be set to 0.
Geometric proportion mode works only in AB or ABC modes. This means that the relationship between the stable range
of AB or ABC frame is fixed. The fixed relationship can be adjusted by registers 0xC492 and 0xC493. In geometric
proportion mode, the register 0xC457[0] must be set to be 1.
by whether the pixel is within the ROI region which is defined with the registers 0x5619~0x5628. The weight of pixels,
which are in ROI region is defined with register 0x562A (for long exposure channel) or 0x562C (for short exposure
channel). The weight of other pixels is defined with register 0x562B (for long exposure channel) and 0x562D (for short
exposure channel). The long ROI shift (0x5629[5:3]) and short ROI shift (0x5629[2:0]) control the precision of the long
ROI weight and the short ROI weight, separately. The weight given to a window is relative to the other windows. Thus,
a window weighted "4" has four times the weight as one weighted "1". The default is that all windows are weighted "1",
and thus, all are weighted evenly.
{0x5605[2:0], 0x5606}
0 1
C
L: {0x5611[2:0], 0x5612}
S: {0x5613[2:0], 0x5614} ROI
S: {0x5617[2:0], 0x5618}
L: {0x5615[2:0], 0x5616}
4 5 6
W
fid in L: {0x5609[2:0], 0x560A}
S: {0x560B[2:0], 0x560C}
7 8 9
L: {0x5619[2:0], 0x561A}
S: {0x561B[2:0], 0x561C}
L: {0x5621[2:0], 0x5622}
S: {0x5623[2:0], 0x5624}
en g o
10 11 12
h
tia nly
2 3
L: {0x5625[2:0], 0x5626}
{0x5607[2:0], 0x5608}
lf
S: {0x5627[2:0], 0x5628}
or
default
address register name value R/W description
Bit[7:2]: Not used
Bit[1:0]: Sampling
0x5600 AEC CTRL00 0x01 RW 0x: 2
10: 4
11: 8
default
address register name value R/W description
Bit[7:0]: Statwinleft[7:0]
0x5602 AEC CTRL02 0x00 RW Horizontal start point of outer 4-zone
statistic window
Bit[7:0]: Statwintop[7:0]
0x5604 AEC CTRL04 0x00 RW
Vertical start point for statistic image
C
statistic window
Bit[7:0]: Statwinright[7:0]
0x5606 AEC CTRL06 0x00 RW Horizontal end point of outer 4-zone
W
fid in
statistic window
window
Bit[7:0]: Statwinbottom
h
0x5608 AEC CTRL08 0x00 RW Bit[7:0]: Vertical end point of outer 4-zone statistic
tia nly
window
Bit[7:0]: winleft_l
or
0x560A AEC CTRL0A 0x64 RW Bit[7:0]: Horizontal start point of inner 9-zone
window long exposure sub-pixel
Bit[7:0]: winleft_s[7:0]
0x560C AEC CTRL0C 0x64 RW Horizontal start point of inner 9-zone
window short exposure sub-pixel
default
address register name value R/W description
Bit[7:0]: wintop_l[7:0]
0x560E AEC CTRL0E 0x4B RW Vertical start point of inner 9-zone window
long exposure sub-pixel
Bit[7:0]: wintop_s[7:0]
0x5610 AEC CTRL10 0x4B RW Vertical start point of inner 9-zone window
short exposure sub-pixel
C
Bit[7:0]: winwidth_l[7:0]
W
fid in
Bit[7:0]: winwidth_s[7:0]
tia nly
Bit[1:0]: winheight_l[9:8]
0x5615 AEC CTRL15 0x00 RW
Vertical width of inner 9-zone window long
exposure sub-pixel
or
Bit[7:0]: winheight_l[7:0]
0x5616 AEC CTRL16 0x96 RW Vertical width of inner 9-zone window long
exposure sub-pixel
Bit[7:0]: winheight_s[7:0]
0x5618 AEC CTRL18 0x96 RW Vertical width of inner 9-zone window long
exposure sub-pixel
default
address register name value R/W description
Bit[7:0]: roi_left_l[7:0]
0x561A AEC CTRL1A 0x00 RW Horizontal start point for ROI for long
exposure sub-pixel
Bit[7:0]: roi_left_s[7:0]
0x561C AEC CTRL1C 0x00 RW Horizontal start point for ROI for short
exposure sub-pixel
C
Bit[7:0]: roi_top_l[7:0]
W
fid in
0x561E AEC CTRL1E 0x00 RW Vertical start point for ROI for long exposure
sub-pixel
Bit[7:0]: roi_top_s[7:0]
tia nly
0x5620 AEC CTRL20 0x00 RW Vertical start point for ROI for short
exposure sub-pixel
Bit[2:0]: roi_right_l[10:8]
0x5621 AEC CTRL21 0x00 RW
Horizontal end point for ROI for long
exposure sub-pixel
or
Bit[7:0]: roi_right_l[7:0]
0x5622 AEC CTRL22 0x00 RW Horizontal end point for ROI for long
exposure sub-pixel
Bit[7:0]: roi_right_s[7:0]
0x5624 AEC CTRL24 0x00 RW Horizontal end point for ROI for short
exposure sub-pixel
default
address register name value R/W description
Bit[7:0]: roi_bottom_l[7:0]
0x5626 AEC CTRL26 0x00 RW Vertical end point for ROI for long exposure
sub-pixel
Bit[7:0]: roi_bottom_s[7:0]
0x5628 AEC CTRL28 0x00 RW Vertical end point for ROI for short
exposure sub-pixel
C
Bit[2:0]: r_roishift_s
image
en g o
0 L: 0x562F 1
L: 0x562E
S: 0x563B
window S: 0x563C
h
tia nly
ROI
S: 0x562D
S: 0x562C
L: 0x562A
L: 0x562B
out ROI:
in ROI:
4 5 6
lf
L: 0x5630 L: 0x5631
2 S: 0x563D S: 0x563E 3
default
address register name value R/W description
Bit[4:3]: Horizontal crop mode select
00: Full size
01: Horizontal crop to 768
0x3621 ANA_ARRAY1 0x03 RW
10: Horizontal crop to 656
11: Not used
Bit[2:0]: Analog delay option
default
address register name value R/W description
0x5644 AEC CTRL44 0x01 RW Weights9 for Short Exposure Sub-pixel
weighted average of current frame. Long exposure can change freely in the whole range. Short exposure, however, is
limited by the new estimated dynamic range and long exposure.
on ac
The AEC/AGC adjustment step is calculated by the distance between current weighted average and the target. When
the current weighted average is far from the stable range, the exposure will adjust by big steps to quickly bring the image
to stable range. When the current weighted average is close to the stable range, the exposure will adjust by small steps
W
fid in
to avoid oscillating.
In the OV9623, the exposure time and gain changes every two frames. At the end of first frame, the new exposure time
en g o
and gain will be estimated and the exposure time registers will be updated afterward. The gain registers will be updated
at the end of the second frame. So, the third frame will be the result of new exposure and gain.
h
tia nly
lf
or
stable_out range
stable_in range
raw target
C
on ac
W
fid in
en g o
h
tia nly
default
address register name value R/W description
lf
default
address register name value R/W description
0x566D AEC CTRL6D 0x00 RW Bit[7:0]: his_addr
default
address register name value R/W description
Bit[7:1]: Not used
0x56EA AEC CTRLEA 0x00 RW
Bit[0]: snrgain_s_m[8]
Bit[7:0]: manual_expo11[15:8]
on ac
Bit[7:0]: manual_expo11[7:0]
Bit[7:0]: manual_expo12[15:8]
tia nly
Bit[7:0]: manual_expo12[7:0]
Bit[7:0]: manual_expo21[15:8]
Bit[7:0]: manual_expo21[7:0]
default
address register name value R/W description
Bit[7:0]: manual_expo22[15:8]
Bit[7:0]: manual_expo22[7:0]
Bit[7:0]: manual_expo31[15:8]
on ac
Bit[7:0]: manual_expo31[7:0]
W
fid in
Bit[7:0]: manual_expo32[15:0]
h
Bit[7:0]: manual_expo32[7:0]
lf
Bit[7:0]: manual_gain11[15:8]
Bit[7:0]: manual_gain11[7:0]
Bit[7:0]: manual_gain12[15:8]
default
address register name value R/W description
Bit[7:0]: manual_gain12[7:0]
Bit[7:0]: manual_gain21[15:8]
Bit[7:0]: manual_gain21[7:0]
on ac
Bit[7:0]: manual_gain22[15:8]
W
fid in
Bit[7:0]: manual_gain22[7:0]
h
Bit[7:0]: manual_gain31[15:8]
lf
Bit[7:0]: manual_gain31[7:0]
Bit[7:0]: manual_gain32[15:8]
Bit[7:0]: manual_gain32[7:0]
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: manual_en
0: Disable
1: Enable
0xC308 S_MANUAL_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is
enabled. This register value must be initialized
by user and must not be removed from start up
sequence. Default value is random.
Bit[2]: targetc_manual_en
0: Disable
on ac
1: Enable
Bit[1]: targetb_manual_en
0: Disable
1: Enable
0xC309 S_MANUAL_MODE – RW
W
fid in
Bit[0]: targeta_manual_en
0: Disable
1: Enable
en g o
Bit[1:0]: manual_done
00: Write protected
01: Write valid once
0xC30A S_MANUAL_DONE – RW 10: Write valid always
lf
default
address register name value R/W description
Bit[7:0]: L/S sensitivity ratio[7:0]
0: Disable
1: Enable
0xC456 FIXED_RATIO_EN – RW
When register value is 0x00, it means function is
en g o
1: Enable
0xC457 GP_MODE_EN – RW
When register value is 0x00, it means function is
or
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Only insert frame when in night
mode
0: Disable
1: Enable
0xC459 NIGHT_MODE_CTRL – RW
When register value is 0x00, it means function is
disabled; all other values mean function is
enabled. This register value must be initialized
by user and must not be removed from start up
sequence. Default value is random.
C
0: Disable
1: Enable
0xC45A FRACTAL_EXP_EN – RW
When register value is 0x00, it means function is
W
fid in
0xC45B NONLINEAR_GAIN_EN – RW
This register value must be initialized by user
tia nly
0xC45C MANU_GAMMA_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is
enabled. This register value must be initialized
by user and must not be removed from start up
sequence. Default value is random.
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Banding filter
0: Disable
1: Enable
0xC45F BAND_FILTER_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is
enabled. This register value must be initialized
by user and must not be removed from start up
sequence. Default value is random.
1: Enable
0xC460 BAND_FILTER_SHORT – RW
When register value is 0x00, it means function is
disabled; all other values mean function is
W
fid in
1: Enable
tia nly
0xC461 LESS_1BAND_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is
enabled. This register value must be initialized
by user and must not be removed from start up
lf
default
address register name value R/W description
Bit[7:0]: Log target 1[7:0]
default
address register name value R/W description
Target of Raw Data for Short 2
default
address register name value R/W description
Slow AEC Adjustment Step for Short Exposure
default
address register name value R/W description
Bit[7:0]: Max gain for long[7:0]
0xC486 MIN_GAIN_SHORT_1 – RW
This register value must be initialized by user
and must not be removed from start up
sequence. Default value is random.
default
address register name value R/W description
Bit[7:0]: Max exposure for long[7:0]
default
address register name value R/W description
C/A Ratio in Gp Mode
default
address register name value R/W description
Bit[7:0]: Max gamma list 2[15:8]
default
address register name value R/W description
Bit[7:0]: Dynamic range list 3[7:0]
default
address register name value R/W description
Bit[7:0]: Sensor clock ratio[7:0]
Bit[7:0]: VTS[15:8]
Bit[7:0]: VTS[7:0]
on ac
0x5A00~
W
fid in
tia nly
lf
or
In order to maximize the ADC range, and thus, the SNR, the OV9623 compensates for the black level of active pixels by
using optically shielded pixels. There are coarse and fine BLC calibrations that, when used together, can compensate
for very large offsets with a high degree of accuracy.
Coarse BLC and fine BLC are initiated by any of the following conditions:
BLC can also be manually triggered by setting the register 0x4003[7] from 0 to 1.
en g o
default
h
0: Disable
1: Enable
or
default
address register name value R/W description
Bit[7]: BLC manual trigger
BLC will update manual_frame_num
frames continuously. Refer to register
BLC CTRL03[5:0] when this register
changes from 0 to 1
Bit[6]: BLC freeze
0x4003 BLC CTRL03 0x08 RW 0: BLC running
1: BLC freeze
Bit[5:0]: manual_frame_num
Number of frames BLC will be
updated continuously when BLC is
C
exposure channel
tia nly
default
address register name value R/W description
Bit[7]: Debug control
Changing this value is not allowed
Bit[6]: BLC temperature trigger enable for
short exposure channel
0: Temperature change does not
trigger BLC
1: Temperature change triggers
BLC
Bit[5]: BLC exposure trigger enable for
short exposure channel
0: Exposure change does not
C
trigger BLC
1: Exposure change triggers BLC
Bit[4]: BLC gain trigger enable for short
on ac
exposure channel
0: Gain change does not trigger
BLC
1: Gain change triggers BLC
0x4055 BLC CTRL55 0xFF RW
W
fid in
BLC
tia nly
C
on ac
W
fid in
en g o
h
tia nly
lf
or
The DSP top level control registers allow enabling and disabling of individual DSP blocks. However, the user must be
very careful as each image format requires specific blocks. Provided reference settings should always be used as a
guideline.
default
address register name value R/W description
C
enable
Bit[2]: AWB statistic enable
Bit[1]: AWB gain enable
Bit[0]: Lens shading correction enable
en g o
default
address register name value R/W description
Bit[7]: Vertical subsampling enable
0: Disable
1: Enable
Bit[6]: Lens shading correction center
option
0: Manually set by register
1: Automatically set based on
image window
Bit[5]: Output row in drop mode of
subsampling
0: First row
C
1: Second row
Bit[4]: Output column in drop mode of
subsampling
on ac
0: First pair
0x5005 ISP RW05 0x08 RW 1: Second pair
Bit[3]: Average enable in non-drop
mode of subsampling
W
fid in
0: Sum
1: Average
Bit[2]: Green/Y channel subsampling
mode
en g o
0: Non-drop
1: Drop
Bit[1]: RB/UV channel subsampling
h
mode
tia nly
0: Non-drop
1: Drop
Bit[0]: Subsampling mode enable
0: Full resolution
lf
1: Subsampling
or
5.2 LENC
The lens correction (LENC) algorithm compensates for the illumination drop off in the corners due to the lens. Based on
the radius of each pixel to the lens, the algorithm calculates a gain for each pixel and then corrects each pixel with the
calculated gain to compensate for the light distribution due to the lens curvature.
The LENC register settings are calculated from a lens correction tool developed by OmniVision and run with the specific
lens used on the application.
100%
on ac
LENC coefficient
W
fid in
en g o
register 0x509C/16
h
tia nly
lf
LENC gain is fixed by default. Register 0x5080[5] turns on the gain adaptive LENC.
default
address register name value R/W description
Bit[7]: Not used
Bit[6]: Gain manual mode enable
0: Use auto gain
1: Use manual gain set by user
0x5080 LENC CTRL0 0x10 RW Bit[5]: Auto LENC switch enable
0: LENC gain is fixed
1: LENC gain adjusts according to sensor
gain
Bit[4:0]: Manual gain input
Bit[2:0]: long_red_x0[10:8]
Bit[7]: long_red_sign
h
Bit[7]: long_grn_sign
0x508F LENC CTRL15 0x00 RW
Bit[6:0]: long_grn_b1
default
address register name value R/W description
Bit[7:3]: Not used
0x5091 LENC CTRL17 0x00 RW
Bit[2:0]: long_blu_x0[10:8]
Bit[7]: long_blu_sign
0x5097 LENC CTRL23 0x0 RW
Bit[6:0]: long_blu_b1
Bit[7]: short_red_sign
0x50A7 LENC CTRL39 0x00 RW
Bit[6:0]: short_red_b1
default
address register name value R/W description
Bit[7:3]: Not used
0x50A9 LENC CTRL41 0x00 RW
Bit[2:0]: short_grn_x0[10:8]
Bit[7]: short_grn_sign
0x50AF LENC CTRL47 0x00 RW
Bit[6:0]: short_grn_b1
Bit[6:0]: short_blu_a1
Bit[7]: short_blu_sign
0x50B7 LENC CTRL55 0x00 RW
Bit[6:0]: short_blu_b1
The raw R, G, and B values of a white object detected by an image sensor vary with the spectrum of the light source.
The light source spectrum is usually described by "color temperature". The white balance process applies different gains
to each color channel to make the white object appear white in the image.
The OV9623 builds the AWB algorithm to automatically adjust the gain of each channel to achieve white balance. There
are two kinds of AWB: color Temperature (CT) based AWB and simple gray world AWB. CT AWB is based on the color
temperature of the scene, which is based on G/R and G/B ratios. Simple AWB calculates the gains based on scene
simple statistics of the final image.
For OV9623, AWB gets two sets of statistics separately from long and short channels. It can work in four modes:
separated mode, long channel mode, short channel mode and combination mode. In separated mode, the two channels
may apply different AWB gain. In other modes, they apply same AWB gain. Based on the two sets of statistics, the AWB
C
will estimate two sets of AWB gain at first. In long or short channel mode, the two channels apply one of the two sets. In
combination mode, a weighted average of the two sets of statistics is used to estimate the AWB gain. In separate mode,
on ac
the two channels will apply the two sets of AWB gain respectively.
W
fid in
default
address register name value R/W description
en g o
1: Enable
0x5000 ISP RW00 0xFF RW
tia nly
default
address register name value R/W description
0x5103 GAIN AWB CTRL3 0x80 RW Bit[7:0]: manual_gain_gb_long[7:0]
Bit[1:0]: manual_gain_r_short[9:8]
tia nly
all pixels equal to each other by adjusting the gain of each color channel.
5.3.2 CT AWB
CT AWB algorithm adjusts R, G and B gain based on the color temperature of the ambient light. It will make the R, G
and B channel average of gray pixels equal to each other by adjusting the gain of each color channel. To identify the gray
pixels over the color temperature range, the characteristics of a gray object must be calibrated first using the target lens.
default
address register name value R/W description
Bit[7:0]: AWB_M_RNG[7:0]
Tolerance of AWB_M_X and
AWB_M_Y in middle color temperature
range.
0x5586 AWB_M_RNG 0x10 RW
Too small of a tolerance results in
unstable AWB, while too great of a
tolerance results in inaccurate AWB.
Bit[7:0]: AWB_L_XRNG[7:0]
Tolerance of AWB_L_X in low color
C
Bit[7:0]: AWB_H_YRNG[7:0]
Tolerance of AWB_H_Y in low color
temperature range, where AWB_H_Y
en g o
Bit[7:0]: AWB_M_X[7:0]
X characteristics of gray object in
or
default
address register name value R/W description
Bit[7:0]: AWB_M_Y[7:0]
Y characteristics of gray object in
middle color temperature range.
unpredictable
Bit[7:0]: AWB_L_K
K characteristics of gray object in low
W
fid in
Bit[7:0]: AWB_H_K
K characteristics of gray object in high
color temperature range
lf
Bit[7:0]: AWB_H_LMT[7:0]
Lower limit of AWB_H_X, where
AWB_H_X is X characteristics of gray
object in high color temperature range.
0x558D AWB_H_LMT 0x00 RW
Smaller AWB_H_LMT covers greater
upper limit of color temperature;
however, it also results in less accurate
white balance
default
address register name value R/W description
Bit[7:0]: AWB_L_LMT[7:0]
Lower limit of AWB_L_Y, where
AWB_L_Y is Y characteristics of gray
object in low color temperature range.
0x558E AWB_L_LMT 0x00 RW
Smaller AWB_L_LMT covers smaller
lower limit of color temperature;
however, it also results in less accurate
white balance.
Bit[7:0]: AWB_DBG1
0x558F AWB_DBG1 0x20 RW Debug control register, not effective in
C
normal usage
on ac
Bit[7:0]: AWB_DBG2
0x5590 AWB_DBG2 0x20 RW Debug control register, not effective in
normal usage
Bit[7:0]: AWB_DATA_ULMT
W
fid in
Bit[7:0]: AWB_DATA_LLMT
Pixels with output value smaller than
0x5592 AWB_DATA_LLMT 0x00 RW
AWB_DATA_LLMT are excluded in
h
AWB statistics
tia nly
default
or
default
address register name value R/W description
Bit[7:0]: AWB_L_XRNG[7:0]
Tolerance of AWB_L_X in low color
temperature range, where AWB_L_X is
X characteristics of gray object in low
color temperature range.
0x55A0 AWB_L_XRNG 0x10 RW
Too small of a tolerance results in
unstable AWB, while too great of a
tolerance results in inaccurate AWB.
Typical value ranges from 0x08~0x18.
Bit[7:0]: AWB_H_YRNG[7:0]
C
Bit[7:0]: AWB_M_X[7:0]
X characteristics of gray object in
h
default
address register name value R/W description
Bit[7:0]: AWB_M_Y[7:0]
Y characteristics of gray object in
middle color temperature range.
unpredictable
Bit[7:0]: AWB_L_K
K characteristics of gray object in low
W
fid in
than 0x80
tia nly
Bit[7:0]: AWB_H_K
K characteristics of gray object in high
color temperature range
lf
temperature range.
Bit[7:0]: AWB_H_LMT[7:0]
Lower limit of AWB_H_X, where
AWB_H_X is X characteristics of gray
object in high color temperature range.
0x55A6 AWB_H_LMT 0x00 RW
Smaller AWB_H_LMT covers greater
upper limit of color temperature;
however, it also results in less accurate
white balance
default
address register name value R/W description
Bit[7:0]: AWB_L_LMT[7:0]
Lower limit of AWB_L_Y, where
AWB_L_Y is Y characteristics of gray
object in low color temperature range.
0x55A7 AWB_L_LMT 0x00 RW
Smaller AWB_L_LMT covers smaller
lower limit of color temperature;
however, it also results in less accurate
white balance.
Bit[7:0]: AWB_DBG1
0x55A8 AWB_DBG1 0x20 RW Debug control register, not effective in
C
normal usage
on ac
Bit[7:0]: AWB_DBG2
0x55A9 AWB_DBG2 0x20 RW Debug control register, not effective in
normal usage
Bit[7:0]: AWB_DATA_ULMT
W
fid in
Bit[7:0]: AWB_DATA_LLMT
Pixels with output value smaller than
0x55AB AWB_DATA_LLMT 0x00 RW
AWB_DATA_LLMT are excluded in
h
AWB statistics
tia nly
default
address register name value R/W description
Bit[7:4]: Not used
Bit[3:2]: Scale of AWB_L_K and AWB_H_K for long
exposure. It is usually set to 2’b01
00: 2x
01: 4x
0x5581 AWB CT CTRL1 0x5B RW
10: 8x
11: Not allowed
Bit[1:0]: AWB debug mode
Changing these registers is not
recommended.
default
address register name value R/W description
Bit[7:6]: Scale of AWB_L_K and AWB_H_K for
short exposure, it is usually set to 2’b01
00: 2x
01: 4x
10: 8x
11: Not allowed
Bit[4]: Fast adjustment enable in simple AWB
mode
0: Disable, AWB speed is slow
1: Enable, AWB adjustment is fast for
fast scene change
0x5583 AWB CT CTRL3 0x10 RW
C
image boundary
10: Exclude 1/8 of total rows and columns
at each image boundary
11: Exclude 1/4 of total rows and columns
W
fid in
default
or
The DNS block uses a low pass filter to remove white noise in each color channel and white noise between Gb and Gr.
Control parameters are separated for long and short exposures. A difference below the threshold is treated as noise and
will be smoothed. A difference above the threshold is treated as an edge and will be preserved. The low pass filter is
adaptive to the gain value.
{0x522E, 0x522F}
0x5220
C
{0x522C, 0x522D}
long
on ac
0x521f
threshold
{0x522A, 0x522B}
0x521e
W
fid in
0x521D
0x521C
0x521B {0x5228,
en g o
0x521A 0x5229}
{0x5226,
0x5227}
h
{0x5224,
{0x5222, 0x5225}
tia nly
0x5223}
{0x5255, 0x5256}
0x5248
{0x5253, 0x5254}
short
0x5247
threshold
{0x5251, 0x5252}
0x5246
0x5245
0x5244
C
0x5243 {0x524F,
0x5242 0x5250}
on ac
{0x524D,
{0x524B, 0x524E}
{0x5249, 0x524C}
0x524A}
W
fid in
default
address register name value R/W description
lf
0x5213 DNS CTRL13 0x02 RW Bit[7:0]: noise_y for long exposure sub-pixel
0x5215 DNS CTRL15 0x02 RW Bit[7:0]: noise_u[7:0] for long exposure sub-pixel
0x5217 DNS CTRL17 0x02 RW Bit[7:0]: noise_v[7:0] for long exposure sub-pixel
default
address register name value R/W description
0x5218 DNS CTRL18 0x06 RW Bit[7:0]: dns_edgethre for long exposure sub-pixel
0x521A DNS CTRL20 0x02 RW Bit[7:0]: noise_y_list_0 for long exposure sub-pixel
0x521B DNS CTRL21 0x04 RW Bit[7:0]: noise_y_list_1 for long exposure sub-pixel
0x521C DNS CTRL22 0x08 RW Bit[7:0]: noise_y_list_2 for long exposure sub-pixel
0x521D DNS CTRL23 0x14 RW Bit[7:0]: noise_y_list_3 for long exposure sub-pixel
C
0x521E DNS CTRL24 0x1E RW Bit[7:0]: noise_y_list_4 for long exposure sub-pixel
on ac
0x521F DNS CTRL25 0x28 RW Bit[7:0]: noise_y_list_5 for long exposure sub-pixel
0x5220 DNS CTRL26 0x32 RW Bit[7:0]: noise_y_list_6_l for long exposure sub-pixel
0x5223 DNS CTRL29 0x02 RW Bit[7:0]: noise_uv_list_0[7:0] for long exposure sub-pixel
en g o
0x5225 DNS CTRL31 0x04 RW Bit[7:0]: noise_uv_list_1[7:0] for long exposure sub-pixel
tia nly
0x5227 DNS CTRL33 0x0C RW Bit[7:0]: noise_uv_list_2[7:0] for long exposure sub-pixel
lf
0x5229 DNS CTRL35 0x28 RW Bit[7:0]: noise_uv_list_3[7:0] for long exposure sub-pixel
0x522B DNS CTRL37 0x32 RW Bit[7:0]: noise_uv_list_4[7:0] for long exposure sub-pixel
0x522D DNS CTRL39 0x3C RW Bit[7:0]: noise_uv_list_5[7:0] for long exposure sub-pixel
0x522F DNS CTRL41 0x4C RW Bit[7:0]: noise_uv_list_6[7:0] for long exposure sub-pixel
default
address register name value R/W description
Bit[7:5]: Not used
0x5239 DNS CTRL51 0x08 RW
Bit[4:0]: noise_uv_a for short exposure sub-pixel
0x523B DNS CTRL53 0x02 RW Bit[7:0]: noise_y for short exposure sub-pixel
0x523D DNS CTRL55 0x02 RW Bit[7:0]: noise_u[7:0] for short exposure sub-pixel
C
0x523F DNS CTRL57 0x02 RW Bit[7:0]: noise_v[7:0] for short exposure sub-pixel
0x5240 DNS CTRL58 0x06 RW Bit[7:0]: dns_edgethre for short exposure sub-pixel
W
fid in
0x5242 DNS CTRL60 0x02 RW Bit[7:0]: noise_y_list_0 for short exposure sub-pixel
en g o
0x5243 DNS CTRL61 0x04 RW Bit[7:0]: noise_y_list_1 for short exposure sub-pixel
h
0x5244 DNS CTRL62 0x08 RW Bit[7:0]: noise_y_list_2 for short exposure sub-pixel
tia nly
0x5245 DNS CTRL63 0x14 RW Bit[7:0]: noise_y_list_3 for short exposure sub-pixel
0x5246 DNS CTRL64 0x1E RW Bit[7:0]: noise_y_list_4 for short exposure sub-pixel
lf
0x5247 DNS CTRL65 0x28 RW Bit[7:0]: noise_y_list_5 for short exposure sub-pixel
0x5248 DNS CTRL66 0x32 RW Bit[7:0]: noise_y_list_6 for short exposure sub-pixel
or
0x524A DNS CTRL68 0x02 RW Bit[7:0]: noise_uv_list_0[7:0] for short exposure sub-pixel
0x524C DNS CTRL70 0x04 RW Bit[7:0]: noise_uv_list_1[7:0] for short exposure sub-pixel
0x524E DNS CTRL72 0x0C RW Bit[7:0]: noise_uv_list_2[7:0] for short exposure sub-pixel
0x5250 DNS CTRL74 0x28 RW Bit[7:0]: noise_uv_list_3[7:0] for short exposure sub-pixel
default
address register name value R/W description
Bit[7:1]: Not used
0x5251 DNS CTRL75 0x00 RW
Bit[0]: noise_uv_list_4[8] for short exposure sub-pixel
0x5252 DNS CTRL76 0x32 RW Bit[7:0]: noise_uv_list_4[7:0] for short exposure sub-pixel
0x5254 DNS CTRL78 0x3C RW Bit[7:0]: noise_uv_list_5[7:0] for short exposure sub-pixel
0x5256 DNS CTRLl80 0x4C RW Bit[7:0]: noise_uv_list_6[7:0] for short exposure sub-pixel
on ac
CIP block interpolates raw R,G,B pixels to RGB space. It also contains the sharpen function.
LONG SHORT
h
sharpen amount
lf
Reg. {0x5280, 0x5281} Reg. {0x5282, 0x5283} Reg. {0x52c0, 0x52c1} Reg. {0x52c2, 0x52c3}
gain gain
default
address register name value R/W description
0x5000 ISP RW00 1'b1 RW Bit[6]: cip_en
default
address register name value R/W description
Bit[7:0]: min_gain[7:0] for long exposure
0x5281 CIP CTRL01 0x10 RW Min_gain is used in CIP_start module and is used
to judge in which range current sensor is in
tia nly
default
address register name value R/W description
Bit[5:0]: min_sharpen_tp[5:0] for long exposure
0x528E CIP CTRL0E 0x10 RW Min_sharpen_tp is used for sharpen_tp
computation in auto mode
default
address register name value R/W description
Bit[5:0]: man_sharpen_p[5:0] for long exposure
0x529E CIP CTRL1E 0x08 RW Sharpen_p is input only in manual mode and is
used for function of adaptive sharpen
tia nly
0x52C4 CIP CTRL44 0x00 RW Min_noise used for calculating int_noise in auto
mode
or
default
address register name value R/W description
Bit[7:0]: unsharpen_mask1 for short exposure
0x52C9 CIP CTRL49 0x30 RW UnSharpenMask0 used in some filters as
multipliers
tia nly
default
address register name value R/W description
Bit[5:0]: lthre[5:0] for short exposure
0x52D9 CIP CTRL59 0x08 RW
Threshold for low frequency signals
0x52DC CIP CTRL5C 0x00 RW Inv_noise is input only in manual mode and is
used as threshold in some filters
on ac
0x52DE CIP CTRL5E 0x08 RW Sharpen_p is input only in manual mode and is
used for function of adaptive sharpen
0x52DF CIP CTRL5F 0x08 RW Sharpen_m is input only in manual mode and is
used for function of adaptive sharpen
h
0x52E0 CIP CTRL60 0x06 RW Sharpen_tp is input only in manual mode and is
used for function of adaptive sharpen
The main purpose of color matrix (CMX) is color correction. There is generally no need to change these register values.
default
address register name value R/W description
W
fid in
Bit[7]: cmx_en
0x5000 ISP CTRL00 0x01 RW 0: Disable CMX
1: Enable CMX
en g o
default
address register name value R/W description
Bit[7:0]: Long color matrix 3[7:0]
default
address register name value R/W description
Bit[7:0]: Long color matrix 8[15:8]
default
address register name value R/W description
Bit[7:0]: Long color matrix 12[7:0]
default
address register name value R/W description
Bit[7:0]: Short color matrix 5[15:8]
default
address register name value R/W description
Bit[7:0]: Short color matrix 9[7:0]
The auto color saturation block can adjust the color saturation level based on the sensor gain. Thus, in low light, when
the gain setting is high, the color saturation can be reduced to effectively reduce spatial noise in the scene (i.e., resulting
image will lose some color saturation but will be less noisy). In bright conditions, when gain is low and noise is also
relatively low, the color saturation will be at a high level.
0x20
C
on ac
16xs
W
fid in
0xC317
en g o
MinSaturation
h
tia nly
0 0xC315 0xC316
gain
lf
default
address register name value R/W description
Bit[1:0]: Saturation adjust enable
00: Keep previous saturation
01: Auto adjust color
saturation
10: Keep minimum saturation
0xC314 SATURATION_ADJ_EN – RW
11: Keep maximum saturation
default
address register name value R/W description
Minimum Gain To Adjust Color Saturation
0xC317 SATURATION_MINTHRE – RW
This register value must be initialized by user
and must not be removed from start up
sequence. Default value is random.
W
fid in
en g o
h
tia nly
lf
or
5.8 combine
The purpose of the combine block is to combine the long exposure and short exposure channels into a single pixel. The
registers define the relative weight of each channel and the overlap.
default
address register name value R/W description
Bit[0]: Combine enable
0x5001 ISP RW01 1'b1 RW 0: Disable
1: Enable
C
Bit[3:0]: comb_thre_l2
0x5406 COMB CTRL6 0x0A RW
Threshold3 of long channel
Bit[3:0]: comb_uv_thre_s0
0x5407 COMB CTRL7 0x05 RW
UV threshold1 of short channel
default
address register name value R/W description
Bit[3:0]: comb_uv_thre_s1
0x5408 COMB CTRL8 0x08 RW
UV threshold2 of short channel
Bit[3:0]: comb_uv_thre_s2
0x5409 COMB CTRL9 0x0A RW
UV threshold3 of short channel
Bit[3:0]: comb_uv_thre_l0
0x540A COMB CTRL10 0x09 RW
UV threshold1 of long channel
Bit[3:0]: comb_uv_thre_l1
0x540B COMB CTRL11 0x0A RW
UV threshold2 of long channel
Bit[3:0]: comb_uv_thre_l2
C
default
address register name value R/W description
0x5423 COMB CTRL35 0x60 RW Bit[7:0]: comb_uv_weight12
0: Disable
1: Enable
or
default
address register name value R/W description
Bit[7:0]: Max curve gain[15:8]
tia nly
default
address register name value R/W description
Dark Boost Amount
5.9 normalize
Normalize module is designed to adjust the image contrast. Normalize supports auto and manual mode. In manual
mode, the normalize algorithm uses the manual input high level and low level. In auto mode, the algorithm will
automatically update the low level and high levels.
default
address register name value R/W description
Bit[1]: Normalize enable
0x5001 ISP_CTRL1 1'b1 R/W 0: Disable normalize
1: Enable normalize
C
16 ~ 127
Bit[7:0]: min_low_level
0x5482 NORM RW02 0xF8 RW -128 to -16, complementary
en g o
code
Bit[6:0]: ps_thres[14:8]
tia nly
5.10 tone_mapping
default
address register name value R/W description
Bit[2]: tone_mapping enable
0x5001 ISP CTRL01 1'b1 RW 0: Disable tone_mapping
1: Enable tone_mapping
000: 16
001: 32
0x5500 TOMP RW00 0x03 RW
on ac
010: 64
011: 128
100: 256
101: 512
W
fid in
00: 16
01: 32
h
Bit[1:0]: uv_dark_thre
00: 16
01: 32
10: 48
lf
11: 64
Contrast Curve 1
default
address register name value R/W description
Contrast Curve 2
Contrast Curve 3
Contrast Curve 4
on ac
Contrast Curve 5
W
fid in
Contrast Curve 6
h
Contrast Curve 7
lf
Contrast Curve 8
Contrast Curve 9
Contrast Curve 10
default
address register name value R/W description
Contrast Curve 11
Contrast Curve 12
Contrast Curve 13
on ac
Contrast Curve 14
W
fid in
Contrast Curve 15
h
default
address register name value R/W description
Bit[7:0]: Curve max dynamic range[7:0]
tia nly
lf
or
default
address register name value R/W description
Bit[2]: winc_en
0: Disable window
0x5002 ISP CTRL02 1'b1 RW cropping
1: Enable window
cropping
enable
0: Disable
1: Enable
on ac
1: Second pair
tia nly
1: Average
Bit[2]: Green/Y channel
subsampling mode
or
0: Non-drop
1: Drop
Bit[1]: RB/UV channel
subsampling mode
0: Non-drop
1: Drop
Bit[0]: Subsampling mode
enable
0: Full resolution
1: Subsampling
The DPC function detects the defect pixel/cluster by using a programmable threshold which can be set manually by
registers or automatically calculated based on analog gain from a programmable threshold gain curve.
The OV9623 supports up to four groups. Each group can have up to 128 registers. For group operation, you must first
record the group, then write the related register values, and last, set record end.
default
C
Bit[5:4]: Group ID
Bit[3:2]: Chip debug
h
C
on ac
W
fid in
en g o
h
tia nly
lf
or
Embedded line contains register values. The embedded lines are prefixed to the normal image data. The line length of
the embedded line is the same as the normal image line. Embedded line only contains register values that are followed
with a 0x369 tag (10-bit) or 0xDA tag (8-bit). The 8-bit register values are output D[9:2].
There are two embedded lines. Only the last embedded line contains valid register data for each frame. The first
embedded line is a dummy line used to make an even number of lines per frame. Please refer to the Embedded Line
Application Note for details, and how to customize the embedded line.
C
default
on ac
Bit[7:4]: s2h_width
en g o
(1)
VSYNC
(2) (3) (4)
(7) (5)
(7)
HREF
(6) (6)
PCLK
PCLK
B1S[9:0] B1L[9:0] G2S[9:0] G2L[9:0]
D[9:0]
Y1[9:0] U1[9:0] Y2[9:0] V2[9:0]
W
fid in
D[9:0]
G1S[9:0] G1L[9:0] R2S[9:0] R2L[9:0]
D[9:0]
PCLK PCLK
D[9:0] D[9:0]
tia nly
D[9:0] G1[9:0] R2[9:0] G3[9:0] R4[9:0] D[9:0] G1[9:0] G1[17:8] R2[9:0] R2[17:8]
lf
format
or
format
(1) frame period 520 lines 520 lines 520 lines 520 lines
(1) frame period 520 lines 520 lines 520 lines 520 lines
tia nly
format
a. These parameters change with register settings. They are different with different register settings.
tia nly
lf
or
tCKNVSR tCKNVSF
VSYNC
tCKNHRF tCKNHRR
HREF/HSYNC
tCLKF tCLKR
PCLK
tHOLD
C
tCKNHRR PCLK falling edge to HREF rising edge delay – -0.5 0.5 ns
lf
C
on ac
W
fid in
en g o
h
tia nly
lf
or
7 register tables
The following table provides a description of the device control registers contained in the OV9623. The 7-bit SCCB slave
device address is 0x30. The low 3 bits come from GPIO[2:0]/SID[2:0] which is controlled by 0x300C[0].
In order to guarantee reasonable performance, the initialization register sequence must be based on the register
settings provided by OmniVision. Contact your local AE for register settings that suit your application.
default
address register name value R/W description
on ac
1: Stream on
Bit[7:2]: Reserved
0x3000 SC_CMMN_PAD_OEN0 0x00 RW
Bit[1:0]: io_y_oen[9:8]
h
Bit[7:4]: io_y_oen[7:0]
tia nly
Bit[7]: io_vsync_oen
Bit[6]: io_href_oen
lf
Bit[5]: io_pclk_oen
Bit[4]: Reserved
0x3002 SC_CMMN_PAD_OEN2 0x00 RW
Bit[3]: io_strobe_oen
or
Bit[2]: io_sda_oen
Bit[1]: io_gpio1_oen
Bit[0]: io_gpio0_oen
default
address register name value R/W description
Bit[7]: Bypass PCLK PLL
Bit[6:4]: PCLK PLL pre div
0x3006 SC_CMMN_PLL_CTRL3 0x00 RW
Bit[3]: PCLK PLL cp[2]
Bit[2:0]: PCLK PLL sdiv
Bit[7:2]: Reserved
0x3008 SC_CMMN_PAD_OUT0 0x00 RW
Bit[1:0]: io_y_o[9:8]
Bit[7:1]: SCCB ID
Bit[0]: SCCB ID select
0x300C SC_CMMN_SCCB_ID 0x60 RW
0: {sccb_id[7:4],gpio_i[3:1]}
W
fid in
1: sccb_id[7:1]
Bit[7]: io_vsync_o
Bit[6]: io_href_o
0x300D SC_CMMN_PAD_OUT2 0x00 RW
Bit[5]: io_pclk_o
en g o
Bit[4:0]: Reserved
Bit[7:2]: Reserved
h
Bit[7]: io_vsync_sel
Bit[6]: io_href_sel
lf
0x3016~
SC_CMMN_CTRL – RW Bit[7:0]: Debug mode
0x3019
default
address register name value R/W description
Bit[7:6]: Reserved
Bit[5]: sclk_ac
Bit[4]: sclk_tc
0x301A SC_CMMN_CLKRST0 0x70 RW
Bit[3:2]: Reserved
Bit[1]: rst_ac
Bit[0]: rst_tc
Bit[7]: sclk_blc
Bit[6]: sclk_isp
Bit[5]: Chip debug
Bit[4]: sclk_vfifo
0x301B SC_CMMN_CLKRST1 0xB4 RW
Bit[3]: rst_blc
C
Bit[2]: rst_isp
Bit[1]: Chip debug
on ac
Bit[0]: rst_vfifo
Bit[7]: pclk_dvp
Bit[6:5]: Chip debug
Bit[4]: sclk_otp
W
fid in
Bit[7]: sclk2x_isp
Bit[6:5]: Chip debug
0x301D SC_CMMN_CLKRST3 0xB4 RW Bit[4]: sclk_aec_pk
h
Bit[0]: rst_aec_pk
default
address register name value R/W description
Bit[7:6]: Chip debug
Bit[5]: bist_en
Bit[4]: Clock switch
0: Switch all clock to pad
0x3023 SC_CMMN_CORE_CTRL 0x00 RW
clock
1: Switch from pad clock to
all clock
Bit[3:0]: Chip debug
01: Short
10: Long, short
on ac
11: Combined
Bit[3]: Debug mode
Bit[2:1]: YUV mode
0x3024 SC_CMMN_CORE_CTRL 0x04 RW
00: HDR
W
fid in
01: Long
10: Short
11: Not allowed
Bit[0]: PCLK PLL disable
en g o
Bit[7]: gpio0_sel
Bit[6]: gpio0_dir
Bit[5]: gpio0_out
Bit[4]: Chip debug
0x302F SC_CMMN_GPIO01 0x88 RW
Bit[3]: gpio1_sel
Bit[2]: gpio1_dir
Bit[1]: gpio1_out
Bit[0]: Chip debug
default
address register name value R/W description
Bit[7]: gpio2_sel
Bit[6]: gpio2_dir
Bit[5]: gpio2_out
Bit[4]: Chip debug
0x3030 SC_CMMN_GPIO23 0x80 RW
Bit[3]: gpio3_sel
Bit[2]: gpio3_dir
Bit[1]: gpio3_out
Bit[0]: Chip debug
Bit[7]: gpio4_sel_fsin
Bit[6]: gpio4_dir_fsin
0x3031 SC_CMMN_GPIO45 0x00 RW
Bit[5]: gpio4_out
C
011: sclk/8
100: Pad clock
101: Pad clock/2
110: Pad clock/4
en g o
Bit[2:0]: p_pump_clk_sel
000: sclk
tia nly
001: sclk/2
010: sclk/4
011: sclk/8
100: Pad clock
lf
0x303D~
RSVD – – Reserved
0x303E
default
address register name value R/W description
Bit[7]: sclk_isp_fc
Bit[6]: sclk_fc
Bit[5]: Reserved
Bit[4]: sclk_fmt
0x3040 SC_CMMN_CLKRST5 0xF0 RW
Bit[3]: rst_isp_fc
Bit[2]: rst_fc
Bit[1]: Reserved
Bit[0]: rst_fmt
Bit[7]: sclk_wb
C
Bit[6]: sclk_dr
Bit[5]: sclk_mp
on ac
Bit[4]: sclk_ct
0x3042 SC_SOC_CLKRST7 0xF9 RW
Bit[3]: rst_wb
Bit[2]: rst_dr
Bit[1]: rst_mp
Bit[0]: rst_ct
W
fid in
0x3043~
RSVD – – Reserved
0x3044
en g o
default
address register name value R/W description
0x3600 ANA_ADC1 0x54 RW ADC Control 1
default
address register name value R/W description
0x3610 ANA_ANALOG1 0x2C RW Analog Control 1
0x3616~
C
default
address register name value R/W description
0x3700 SENSOR_REG00 0x22 RW Timing Control 1
default
address register name value R/W description
0x3705 SENSOR_REG05 0x61 RW Timing Control 6
SENSOR_RSTYZ_
en g o
SENSOR_RSTYZ_
h
0x3716~
SENSOR DEBUG 0x03 RW Chip Debug
0x374F
or
default
address register name value R/W description
Manual Horizontal Start Address of Array for
0x3800 TIMING_X_START_ADDR 0x00 RW
Readout High Byte
default
address register name value R/W description
Manual Horizontal Start Address of Array for
0x3801 TIMING_X_START_ADDR 0x00 RW
Readout Low Byte
0x3808 TIMING_X_OUTPUT_SIZE 0x05 RW DVP Horizontal Output Size (Pixel) High Byte
0x3809 TIMING_X_OUTPUT_SIZE 0x00 RW DVP Horizontal Output Size (Pixel) Low Byte
en g o
0x380A TIMING_Y_OUTPUT_SIZE 0x03 RW DVP Vertical Output Size (Pixel) High Byte
h
0x380B TIMING_Y_OUTPUT_SIZE 0x20 RW DVP Vertical Output Size (Pixel) Low Byte
tia nly
0x3811 TIMING_ISP_X_WIN 0x00 RW ISP Horizontal Windowing Start Address Low Byte
0x3812 TIMING_ISP_Y_WIN 0x00 RW ISP Vertical Windowing Start Address High Byte
0x3813 TIMING_ISP_Y_WIN 0x00 RW ISP Vertical Windowing Start Address Low Byte
default
address register name value R/W description
Bit[7:4]: DVP SOF delay control
Bit[3]: BLC vflip
0x3819 TIMING_CTRL19 0x00 RW
Bit[2]: Enable use of left black line to do BLC
Bit[1:0]: Debug control
Bit[6]: hdr_en
0x381D TIMING_CTRL1D 0x40 RW Bit[5:2]: Not used
on ac
0x3820~
TIMING_GRP – RW Chip Debug
0x3823
en g o
default
address register name value R/W description
0x3833 TIMING_TC_CS_RST 0x00 RW Horizontal Counter Reset Value Low Byte
0x3848~
TIMING_CTRL 0x00 RW Frame Counter For Debug
0x3849
C
on ac
default
en g o
default
address register name value R/W description
0x3D0F OTP_DATA_F 0x00 RW OTP Dump/load Data BufferF
Bit[6:4]: write_speed
0x3D11 OTP_SPEED 0x46 RW
Bit[3]: Not used
Bit[2:0]: read_speed
0x3D1F OTP_EF_STATUS – R
Bit[0]: otp_busy
default
address register name value R/W description
0x3D41 OTP_DATA_21 0x00 RW OTP Dump/load Data Buffer21
default
address register name value R/W description
0x3D5D OTP_DATA_3D 0x00 RW OTP Dump/load Data Buffer3D
default
address register name value R/W description
Bit[7:4]: Not used
Bit[3:1]: Chip debug
W
fid in
offsets
tia nly
Bit[7]: format_change_en
0: Change of format will not
trigger BLC
lf
Bit[7]: trig_man
BLC manual trigger signal
BLC will update
manual_frame_num frames
continuously from its rising edge
0x4003 BLC CTRL03 0x08 RW Bit[6]: freeze_en
When set, BLC will freeze
Bit[5:0]: manual_frame_num
Number indicates how many
frames BLC will be updated
continuously when trig_man is set
default
address register name value R/W description
Bit[7:5]: Not used
Bit[4:0]: line_num
0x4004 LINE NUM 0x08 RW
Line number specifies black lines
used in offsets calculation
lines
1: Output image does not
on ac
will be reversed
Bit[1]: blc_always_do
When set, BLC will always update
h
Bit[7:0]: long_blc_target
0x4008 LONG BLC TARGET 0x10 R/W
BLC target for long exposure data
default
address register name value R/W description
Bit[7:0]: short_blc_target
0x4009 SHORT BLC TARGET 0x10 R/W
BLC target for short exposure data
0x400A~
NOT USED – – Not Used
0x400B
Bit[7:0]: man_offset00[7:0]
C
0x400D MANUAL OFFSET 00 0x00 R/W Manual offset for B channel of long
exposure data
on ac
Bit[7:0]: man_offset01[7:0]
0x400F MANUAL OFFSET 01 0x00 RW Manual offset for Gb channel of
long exposure data
en g o
Bit[7:0]: man_offset02[7:0]
0x4011 MANUAL OFFSET 02 0x00 RW Manual offset for Gr channel of
long exposure data
lf
exposure data
Bit[7:0]: man_offset03[7:0]
0x4013 MANUAL OFFSET 03 0x00 RW Manual offset for R channel of long
exposure data
0x4014~
NOT USED – – Not Used
0x4033
Bit[7:0]: man_offset10[7:0]
0x4035 MANUAL OFFSET 10 0x00 RW Manual offset for B channel of short
exposure data
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: man_offset11[8]
0x4036 MANUAL OFFSET 11 0x00 RW
Manual offset for Gb channel of
short exposure data
Bit[7:0]: man_offset11[7:0]
0x4037 MANUAL OFFSET 11 0x00 RW Manual offset for Gb channel of
short exposure data
Bit[7:0]: man_offset12[7:0]
on ac
Bit[0]: man_offset13[8]
0x403A MANUAL OFFSET 13 0x00 RW
Manual offset for R channel of
short exposure data
Bit[7:0]: man_offset13[7:0]
en g o
0x403C~
tia nly
0x404E~
NOT USED – – Not Used
0x404F
or
default
address register name value R/W description
Bit[7:4]: BLC debug control
Bit[3]: short_option
0: Short exposure channel will
use BLC statistics of short
exposure channel
1: Short exposure channel will
use BLC statistics of long
exposure channel
Bit[2]: long_option
0: Long exposure channel will
use BLC statistics of long
0x4052 BLC CTRL52 0x00 RW
C
exposure channel
1: Long exposure channel will
use BLC statistics of short
on ac
exposure channel
Bit[1]: blc_mid_en
0: Keep black line data
1: Median for black line data
W
fid in
Bit[0]: one_channel
When set, used offsets will be
average of calculated offsets for B,
Gb, Gr and R channel
en g o
Bit[5]: short_exp_chg_en
Short channel exposure changing
enable signal
or
Bit[4]: short_gain_chg_en
Short channel gain changing
enable signal
Bit[3]: long_ana_frz_en
0x4055 BLC CTRL55 0xFF RW
Long channel analog freeze enable
signal
Bit[2]: long_tmp_chg_en
Long channel temperature
changing enable signal
Bit[1]: long_exp_chg_en
long channel exposure changing
enable signal
Bit[0]: long_gain_chg_en
Long channel gain changing
enable signal
default
address register name value R/W description
OFFSET TOP LIMIT
0x4056 0x07 RW BLC Control 1
MSB
0: Disable
1: Enable
0xC4B7 AUTO_BLC_EN – RW
When register value is 0x00, it means function
en g o
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Smooth offset statistics in first 16
frames
0: Disable
1: Enable
BLC_PRE_SMOOTH_
0xC4E3 – RW
EN
When register value is 0x00, it means function
is disabled; all other values mean function is
enabled. This register value will be
automatically initialized by sensor after
powering up. Default value is random.
C
enable
Increases stable and slow range
Decreases adjustment step
0: Disable
BLC_HT_OPTION1_
W
fid in
0xC4FA – RW 1: Enable
EN
When register value is 0x00, it means function
is disabled; all other values mean function is
en g o
0: Disable
BLC_HT_OPTION2_
0xC4FB – RW 1: Enable
EN
or
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: High temperature BLC option3
enable
Switches analog gain to digital gain
0: Disable
BLC_HT_OPTION3_ 1: Enable
0xC4FC – RW
EN
When register value is 0x00, it means function
is disabled; all other values mean function is
enabled. This register value must be initialized
by user and must not be removed from start up
C
BLC_HT_TEMP_EXP_ 1: Enable
0xC4FD – RW
EN
When register value is 0x00, it means function
is disabled; all other values mean function is
en g o
Limited by temperature
BLC_HT_TEMP_
0xC4FE – RW
MINEXP This register value must be initialized by user
and must not be removed from start up
lf
BLC_HT_
0xC4FF – RW This register value must be initialized by user
EXPMAXSTEP
and must not be removed from start up
sequence. Default value is random.
default
address register name value R/W description
High Temperature BLC Temperature Threshold
Long 2
BLC_HT_TEMP_TH_
0xC502 – RW
2L This register value must be initialized by user
and must not be removed from start up
sequence. Default value is random.
0xC505 BLC_HT_EXP_TH_12 – RW
This register value must be initialized by user
and must not be removed from start up
h
default
address register name value R/W description
Bit[7:0]: High temperature BLC threshold 1
0x5B1C~
BLC_R – R Debug Information for BLC
0x5B49
default
address register name value R/W description
0x5D1F BLC_RO04 0x00 RW Bit[7:0]: long_offset_01[7:0]
Bit[2:0]: short_offset_11[10:8]
tia nly
0x5D2C~ BLC_RW17~
– – Debug Information for BLC
0x5D2F BLC_RW20
lf
Bit[7]: short_ana_freeze
Bit[6]: long_ana_freeze
or
Bit[5]: short_tmp_chg
Bit[4]: long_tmp_chg
0x5D30 BLC_RW21 – R
Bit[3]: short_exp_chg
Bit[2]: long_exp_chg
Bit[1]: short_gain_chg
Bit[0]: long_gain_chg
default
address register name value R/W description
Bit[7:6]: Not used
Bit[5]: Gain delay option
0: One frame latch
1: Delay one frame latch
0x3503 AEC_PK_MANUAL 0x00 RW
Bit[4]: Choose delay option
C
0: Delay disable
1: Delay enable
Bit[3:0]: Not used
on ac
Bit[1:0]: Sampling
0x5600 AEC CTRL00 0x01 RW 0x: 2
10: 4
11: 8
en g o
Bit[7:0]: Statwinleft[7:0]
0x5602 AEC CTRL02 0x00 RW
Horizontal start point for statistic image
Bit[7:0]: Statwintop[7:0]
0x5604 AEC CTRL04 0x04 RW
Vertical start point for statistic image
Bit[7:0]: Statwinright[7:0]
0x5606 AEC CTRL06 0x00 RW
Horizontal end point for statistic image
Bit[7:0]: Statwinbottom[7:0]
0x5608 AEC CTRL08 0x08 RW
Vertical end point for statistic image
default
address register name value R/W description
Bit[7:3]: Not used
Bit[2:0]: winleft_l[10:8]
0x5609 AEC CTRL09 0x00 RW
Horizontal start point to compute
weight for long exposure sub-pixel
Bit[7:0]: winleft_l[7:0]
0x560A AEC CTRL0A 0x64 RW Horizontal start point to calculate
weight for long exposure sub-pixel
Bit[7:0]: winleft_s[7:0]
on ac
Bit[1:0]: wintop_l[9:8]
0x560D AEC CTRL0D 0x00 RW
Vertical start point to calculate weight
for long exposure sub-pixel
Bit[7:0]: wintop_l[7:0]
en g o
Bit[1:0]: wintop_s[9:8]
0x560F AEC CTRL0F 0x00 RW
Vertical start point to calculate weight
for short exposure sub-pixel
lf
Bit[7:0]: wintop_s[7:0]
0x5610 AEC CTRL10 0x4B RW Vertical start point to calculate weight
for short exposure sub-pixel
or
Bit[7:0]: winwidth_l[7:0]
0x5612 AEC CTRL12 0xC8 RW Horizontal width to calculate weight for
long exposure sub-pixel
Bit[7:0]: winwidth_s[7:0]
0x5614 AEC CTRL14 0xC8 RW Horizontal width to calculate weight for
short exposure sub-pixel
default
address register name value R/W description
Bit[7:2]: Not used
Bit[1:0]: winheight_l[9:8]
0x5615 AEC CTRL15 0x00 RW
Vertical width to calculate weight for
long exposure sub-pixel
Bit[7:0]: winheight_l[7:0]
0x5616 AEC CTRL16 0x96 RW Vertical width to calculate weight for
long exposure sub-pixel
Bit[7:0]: winheight_s[7:0]
on ac
Bit[2:0]: roileft_l[10:8]
0x5619 AEC CTRL19 0x00 RW
Horizontal start point for ROI for long
exposure sub-pixel
Bit[7:0]: roileft_l[7:0]
en g o
0x561A AEC CTRL1A 0x00 RW Horizontal start point for ROI for long
exposure sub-pixel
h
Bit[2:0]: roileft_s[10:8]
0x561B AEC CTRL1B 0x00 RW
Horizontal start point for ROI for short
exposure sub-pixel
lf
Bit[7:0]: roileft_s[7:0]
0x561C AEC CTRL1C 0x00 RW Horizontal start point for ROI for short
exposure sub-pixel
or
Bit[7:0]: roitop_l[7:0]
0x561E AEC CTRL1E 0x00 RW Vertical start point for ROI for long
exposure sub-pixel
Bit[7:0]: roitop_s[7:0]
0x5620 AEC CTRL20 0x00 RW Vertical start point for ROI for short
exposure sub-pixel
default
address register name value R/W description
Bit[7:3]: Not used
Bit[2:0]: roiright_l[10:8]
0x5621 AEC CTRL21 0x00 RW
Horizontal end point for ROI for long
exposure sub-pixel
Bit[7:0]: roiright_l[7:0]
0x5622 AEC CTRL22 0x00 RW Horizontal end point for ROI for long
exposure sub-pixel
exposure sub-pixel
Bit[7:0]: roiright_s[7:0]
on ac
0x5624 AEC CTRL24 0x00 RW Horizontal end point for ROI for short
exposure sub-pixel
Bit[1:0]: roibottom_l[9:8]
0x5625 AEC CTRL25 0x00 RW
Vertical end point for ROI for long
exposure sub-pixel
Bit[7:0]: roibottom_l[7:0]
en g o
0x5626 AEC CTRL26 0x00 RW Vertical end point for ROI for long
exposure sub-pixel
h
Bit[1:0]: roibottom_s[9:8]
0x5627 AEC CTRL27 0x00 RW
Vertical end point for ROI for short
exposure sub-pixel
lf
Bit[7:0]: roibottom_s[7:0]
0x5628 AEC CTRL28 0x00 RW Vertical end point for ROI for short
exposure sub-pixel
or
default
address register name value R/W description
0x5631 AEC CTRL31 0x01 RW Weightl3 for Long Exposure Sub-pixel
0x564B AEC CTRL4B 0x20 RW Bit[7:0]: Maxwl[7:0] for long exposure sub-pixel
default
address register name value R/W description
Bit[7:2]: Not used
0x564C AEC CTRL4C 0x01 RW Bit[1:0]: Maxws[9:8] for short exposure
sub-pixel
default
address register name value R/W description
Bit[7]: Not used
0x565E AEC CTRL5E 0x12 RW Bit[6:0]: r_blackweight2_s[6:0] for short
exposure sub-pixel
Bit[7]: fix_whole
Bit[6]: fix_eof
Bit[5:4]: fix_select
0x566B AEC CTRL6B 0x00 RW 01: my_l
10: my_s
11: idat
Bit[3:0]: fix_value
default
address register name value R/W description
0x566D AEC CTRL6D 0x00 RW Bit[7:0]: r_his_addr
Bit[1]: aecagc_debug
Bit[0]: aecagc_man_en
on ac
default
address register name value R/W description
Bit[7:4]: Not used
0x56E4 AEC CTRLE4 0x00 RW
Bit[3:0]: r_exp_l_f[11:8]
tia nly
0xC2EE~
RSVD – – Reserved
0xC2EF
Bit[7:0]: manual_expo11[15:8]
lf
Bit[7:0]: manual_expo11[7:0]
Bit[7:0]: manual_expo12[15:8]
Bit[7:0]: manual_expo12[7:0]
default
address register name value R/W description
Bit[7:0]: manual_expo21[15:8]
Bit[7:0]: manual_expo21[7:0]
Bit[7:0]: manual_expo22[15:8]
on ac
Bit[7:0]: manual_expo22[7:0]
W
fid in
Bit[7:0]: manual_expo31[15:8]
h
Bit[7:0]: manual_expo31[7:0]
lf
Bit[7:0]: manual_expo32[15:0]
Bit[7:0]: manual_expo32[7:0]
Bit[7:0]: manual_gain11[15:8]
default
address register name value R/W description
Bit[7:0]: manual_gain11[7:0]
Bit[7:0]: manual_gain12[15:8]
Bit[7:0]: manual_gain12[7:0]
on ac
Bit[7:0]: manual_gain21[15:8]
W
fid in
Bit[7:0]: manual_gain21[7:0]
h
Bit[7:0]: manual_gain22[15:8]
lf
Bit[7:0]: manual_gain22[7:0]
Bit[7:0]: manual_gain31[15:8]
Bit[7:0]: manual_gain31[7:0]
default
address register name value R/W description
Bit[7:0]: manual_gain32[15:8]
Bit[7:0]: manual_gain32[7:0]
0: Disable
1: Enable
0xC308 S_MANUAL_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
W
fid in
1: Enable
tia nly
Bit[1]: targetb_manual_en
0: Disable
1: Enable
0xC309 S_MANUAL_MODE – RW
Bit[0]: targeta_manual_en
lf
0: Disable
1: Enable
or
default
address register name value R/W description
Bit[7:2]: Not used
Bit[1:0]: Target number
01: AA mode
10: AB mode
0xC450 TARGET_NUM – RW 11: ABC mode
0: Disable
tia nly
1: Enable
0xC454 NONHDR_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
lf
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Fixed ratio mode
0: Disable
1: Enable
0xC456 FIXED_RATIO_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence.
Default value is random.
1: Enable
0xC457 GP_MODE_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
W
fid in
1: Enable
tia nly
0xC458 NIGHT_MODE_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence.
lf
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Allow fractal exposure
0: Disable
1: Enable
0xC45A FRACTAL_EXP_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value will be automatically initialized
by sensor after powering up. Default value is
random.
Chip Debug
C
0xC45B CHIP DEBUG – – This register value must be initialized by user and
on ac
0: Disable
1: Enable
0xC45C MANU_GAMMA_EN – RW
When register value is 0x00, it means function is
en g o
10: 50Hz
0xC45E BAND_FILTER_FLAG – RW
11: Not valid
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Short banding filter
0: Disable
1: Enable
0xC460 BAND_FILTER_SHORT – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence.
Default value is random.
1: Enable
0xC461 LESS_1BAND_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
W
fid in
1: Enable
tia nly
0xC462 LESS_1BAND_SHORT – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence.
lf
default
address register name value R/W description
Bit[7:0]: Log target 2[7:0]
default
address register name value R/W description
Slow Range for Long Exposure
default
address register name value R/W description
Max Slow Adjustment Ratio
0xC47A~
CHIP DEBUG – – Chip Debug
0xC47B
0xC481 MAX _GAIN_LONG_2 – RW This register value must be initialized by user and
must not be removed from start up sequence.
Default value is random.
default
address register name value R/W description
Bit[7:0]: Max gain for short[7:0]
0xC485 MIN_ GAIN_LONG_2 – RW This register value must be initialized by user and
must not be removed from start up sequence.
Default value is random.
W
fid in
default
address register name value R/W description
Bit[7:0]: Max exposure for short[7:0]
0xC48F MIN_EXP_ SHORT_2 – RW This register value must be initialized by user and
tia nly
0xC494~
CHIP DEBUG – – Chip Debug
0xC497
default
address register name value R/W description
Bit[7:0]: Min gamma list 1[15:8]
default
address register name value R/W description
Bit[7:0]: Max gamma list 2[7:0]
default
address register name value R/W description
Bit[7:0]: Band filter value for 60Hz[15:8]
0xC4AE~
CHIP DEBUG – – Chip Debug
0xC4B0
h
default
address register name value R/W description
Bit[7:0]: Sensor clock ratio[7:0]
Bit[7:0]: VTS[15:8]
Bit[7:0]: VTS[7:0]
on ac
0xC51A~
CHIP DEBUG – – Chip Debug
0xC51B
0x5A00~
en g o
default
address register name value R/W description
Bit[7:1]: Not used
0x5C0E AEC_RW0E – R
Bit[0]: SNRgain short[8]
tia nly
default
address register name value R/W description
or
default
address register name value R/W description
Bit[7]: Data and its weight synchronization
enable
Bit[6]: Black/white mode enable
Bit[5]: Dark level filter enable
0x5001 ISP RW01 0xBF RW Bit[4]: Buffer control enable
Bit[3]: AEC enable
Bit[2]: Tone mapping enable
Bit[1]: Normalize enable
Bit[0]: Long-short combination enable
1: VSYNC
Bit[2:0]: EOF_sel
h
000: AEC_done
0x5003 ISP RW03 0x04 RW
001: Simple_awb_done
tia nly
010: Tone_mapping_done
011: Combine_done
100: AEC_done and
simple_AWB_done and
lf
tone_mapping_done and
combine_done
or
default
address register name value R/W description
Bit[7]: Vertical subsampling enable
0: Disable
1: Enable
Bit[6]: Lens shading correction center option
0: Manually set by register
1: Automatically set based on image
window
Bit[5]: Output row in drop mode of
subsampling
0: First row
1: Second row
C
1: Average
Bit[2]: Green/Y channel subsampling mode
0: Non-drop
1: Drop
en g o
0: Full resolution
1: Subsampling
Bit[7:6]: raw_mode_man
lf
Bit[5:4]: yuv_mode_man
Bit[3]: raw_mode_man_en
0x5006 ISP RW06 0x00 RW
Bit[2]: yuv_mode_man_en
or
Bit[1]: yuv_en_man
Bit[0]: yuv_en_man enable
default
address register name value R/W description
Bit[7:2]: Not used
0x500D ISP RW13 0x00 RW
Bit[1:0]: otp_y_offset[9:8]
Bit[7]: pre_isp_test_en_i
Bit[6]: Not used
C
Bit[5:4]: pre_isp_bar_style_i
0x503D ISP RW61 0x00 RW Bit[3]: Not used
Bit[2]: pre_isp_rolling_i
on ac
Bit[1]: pre_isp_isp_test_i
Bit[0]: pre_isp_rnd_same_i
Bit[7:4]: pre_isp_seed_i
W
fid in
Bit[3]: pre_isp_squ_bw_i
0x503E ISP RW62 0x00 RW
Bit[2]: pre_isp_trans_i
Bit[1:0]: pre_isp_test_sel_i
0x5040~
en g o
tia nly
default
address register name value R/W description
Bit[7]: Not used
Bit[6]: Gain manual mode enable
0: Use auto gain
1: Use manual gain set by user
0x5080 LENC CTRL0 0x10 RW Bit[5]: Auto LENC switch enable
0: LENC gain is fixed
1: LENC gain adjusts according to
sensor gain
Bit[4:0]: Manual gain input
default
address register name value R/W description
Bit[7:2]: Not used
0x5083 LENC CTRL3 0x00 RW
Bit[1:0]: long_red_y0[9:8]
Bit[7]: long_red_sign
0x5087 LENC CTRL7 0x00 RW
C
Bit[6:0]: long_red_b1
Bit[3:0]: long_red_b2
Bit[7]: long_grn_sign
0x508F LENC CTRL15 0x00 RW
Bit[6:0]: long_grn_b1
default
address register name value R/W description
Bit[7]: long_blu_sign
0x5097 LENC CTRL23 0x0 RW
Bit[6:0]: long_blu_b1
Bit[1:0]: short_red_y0[9:8]
tia nly
Bit[7]: short_red_sign
0x50A7 LENC CTRL39 0x00 RW
Bit[6:0]: short_red_b1
default
address register name value R/W description
Bit[7:4]: Not used
0x50AE LENC CTRL46 0x01 RW
Bit[3:0]: short_grn_a2
Bit[7]: short_grn_sign
0x50AF LENC CTRL47 0x00 RW
Bit[6:0]: short_grn_b1
Bit[1:0]: short_blu_y0[9:8]
Bit[7]: short_blu_sign
0x50B7 LENC CTRL55 0x00 RW
Bit[6:0]: short_blu_b1
h
default
address register name value R/W description
Bit[7:2]: Not used
0x5100 GAIN AWB CTRL0 0x00 RW
Bit[1:0]: manual_gain_b_long[9:8]
default
address register name value R/W description
0x5105 GAIN AWB CTRL5 0x80 RW Bit[7:0]: manual_gain_gr_long[7:0]
Bit[1:0]: manual_gain_gb_short[9:8]
default
address register name value R/W description
Bit[7:2]: Not used
0x511C GAIN AWB CTRL28 0x00 RW
Bit[1:0]: manual_offset_gr_short[9:8]
0: Auto mode
1: Manual mode
on ac
Bit[7:6]: step_local
Bit[5:4]: step_fast
Bit[3]: slop_8x_l
W
fid in
Bit[7]: slop_8x_s
h
Bit[6]: slop_4x_s
tia nly
0x5584~
DEBUG MODE – – Debug Mode
0x5585
or
Bit[7:0]: AWB_M_RNG[7:0]
Tolerance of AWB_M_X and AWB_M_Y
in middle color temperature range.
0x5586 AWB_M_RNG 0x10 RW
Too small of a tolerance results in
unstable AWB, while too great of a
tolerance results in inaccurate AWB.
default
address register name value R/W description
Bit[7:0]: AWB_L_XRNG[7:0]
Tolerance of AWB_L_X in low color
temperature range, where AWB_L_X is X
characteristics of gray object in low color
temperature range.
0x5587 AWB_L_XRNG 0x10 RW
Too small of a tolerance results in
unstable AWB, while too great of a
tolerance results in inaccurate AWB.
Typical value ranges from 0x08~0x18.
Bit[7:0]: AWB_H_YRNG[7:0]
C
Bit[7:0]: AWB_M_X[7:0]
X characteristics of gray object in middle
h
default
address register name value R/W description
Bit[7:0]: AWB_M_Y[7:0]
Y characteristics of gray object in middle
color temperature range.
Bit[7:0]: AWB_L_K
K characteristics of gray object in low
color temperature range
W
fid in
temperature range.
In general, AWB_L_K should be no less
than 0x80
h
Bit[7:0]: AWB_H_K
tia nly
Bit[7:0]: AWB_H_LMT[7:0]
Lower limit of AWB_H_X, where
AWB_H_X is X characteristics of gray
object in high color temperature range.
0x558D AWB_H_LMT 0x00 RW
Smaller AWB_H_LMT covers greater
upper limit of color temperature; however,
it also results in less accurate white
balance
default
address register name value R/W description
Bit[7:0]: AWB_L_LMT[7:0]
Lower limit of AWB_L_Y, where
AWB_L_Y is Y characteristics of gray
object in low color temperature range.
0x558E AWB_L_LMT 0x00 RW
Smaller AWB_L_LMT covers smaller
lower limit of color temperature; however,
it also results in less accurate white
balance.
Bit[7:0]: AWB_DBG1
0x558F AWB_DBG1 0x20 RW Debug control register, not effective in
C
normal usage
on ac
Bit[7:0]: AWB_DBG2
0x5590 AWB_DBG2 0x20 RW Debug control register, not effective in
normal usage
Bit[7:0]: AWB_DATA_ULMT
W
fid in
Bit[7:0]: AWB_DATA_LLMT
Pixels with output value smaller than
0x5592 AWB_DATA_LLMT 0x00 RW
AWB_DATA_LLMT are excluded in AWB
h
statistics
tia nly
Bit[7]: awb_gain_m
Bit[6]: Not used
Bit[5]: awb_freeze
Bit[4]: Not used
lf
Bit[3:2]: awb_sim_sel
0x5596 AWB CT CTRL22 0x03 RW 00: awb_simple from after awb_gain
01: awb_simple from after raw_gma
or
0x5593~
DEBUG MODE – – Debug Registers
0x5595
default
address register name value R/W description
Bit[7]: awb_gain_m
Bit[6]: Not used
Bit[5]: awb_freeze
Bit[4]: Not used
Bit[3:2]: awb_sim_sel
0x5596 AWB CT CTRL22 0x03 RW 00: awb_simple from after awb_gain
01: awb_simple from after raw_gma
10: awb_simple from after HDR
11: awb_simple from after awb_gain
Bit[1]: fast_enable
Bit[0]: awb_bias_stat
C
Bit[7:0]: AWB_M_RNG[7:0]
Tolerance of AWB_M_X and AWB_M_Y
in middle color temperature range
W
fid in
Bit[7]: bsum_l_fix
Bit[6]: gsum_l_fix
h
Bit[5]: rsum_l_fix
tia nly
Bit[4]: bsum_s_fix
0x55AF AWB CT CTRL41 0x00 RW
Bit[3]: gsum_s_fix
Bit[2]: rsum_s_fix
Bit[1]: allcnt_l_fix
lf
Bit[0]: allcnt_s_fix
Bit[7:0]: AWB_L_XRNG[7:0]
Tolerance of AWB_L_X in low color
or
default
address register name value R/W description
Bit[7:0]: AWB_H_YRNG[7:0]
Tolerance of AWB_H_Y in low color
temperature range, where AWB_H_Y is Y
characteristics of gray object in high color
temperature range.
0x55A1 AWB_H_YRNG 0x10 RW
Too small of a tolerance results in
unstable AWB, while too great of a
tolerance results in inaccurate white
balance. Typical value ranges from
0x08~0x10.
C
Bit[7:0]: AWB_M_X[7:0]
X characteristics of gray object in middle
on ac
and unpredictable.
tia nly
Bit[7:0]: AWB_M_Y[7:0]
Y characteristics of gray object in middle
color temperature range.
lf
default
address register name value R/W description
Bit[7:0]: AWB_L_K
K characteristics of gray object in low
color temperature range
Bit[7:0]: AWB_H_K
C
temperature range.
Bit[7:0]: AWB_H_LMT[7:0]
Lower limit of AWB_H_X, where
AWB_H_X is X characteristics of gray
en g o
Bit[7:0]: AWB_L_LMT[7:0]
lf
Bit[7:0]: AWB_DBG1
0x55A8 AWB_DBG1 0x20 RW Debug control register, not effective in
normal usage
Bit[7:0]: AWB_DBG2
0x55A9 AWB_DBG2 0x20 RW Debug control register, not effective in
normal usage
Bit[7:0]: AWB_DATA_ULMT
Pixels with output value greater than
0x55AA AWB_DATA_ULMT 0xFF RW
AWB_DATA_ULMT are excluded in AWB
statistics
default
address register name value R/W description
Bit[7:0]: AWB_DATA_LLMT
Pixels with output value smaller than
0x55AB AWB_DATA_LLMT 0x00 RW
AWB_DATA_LLMT are excluded in AWB
statistics
0x55AC~
AWB CTRL – – Debug Registers
0x55AE
Bit[7]: bsum_l_fix
Bit[6]: gsum_l_fix
Bit[5]: rsum_l_fix
Bit[4]: bsum_s_fix
0x55AF AWB CT CTRL41
C
0x00 RW
Bit[3]: gsum_s_fix
Bit[2]: rsum_s_fix
on ac
Bit[1]: allcnt_l_fix
Bit[0]: allcnt_s_fix
Bit[1:0]: midtone_ythre_l1[9:8]
Bit[7:0]: midtone_ythre_l1[7:0]
h
Bit[7:0]: midtone_ythre_l2[7:0]
0x5703 AWB CTRL3 0x00 RW (midtone_ythre_l1+midtone_ythre_l2 ≤
0x3FF)
or
Bit[7:0]: midtone_ythre_h1[7:0]
0x5705 AWB CTRL5 0x68 RW (midtone_ythre_h1+midtone_ythre_h2 ≤
0x3FF)
Bit[7:0]: midtone_ythre_h2[7:0]
0x5707 AWB CTRL7 0x80 RW (midtone_ythre_h1+midtone_ythre_h2 ≤
0x3FF)
default
address register name value R/W description
Bit[7:2]: Not used
0x570A AWB CTRL10 0x00 RW
Bit[1:0]: gamma_ythre_h[9:8]
Bit[7:0]: shadow_ythre1[7:0]
0x5711 AWB CTRL17 0x40 RW (shadow_ythre1 + shadow_ythre2 ≤
W
fid in
0x3FF)
Bit[7:0]: shadow_ythre2[7:0]
0x5713 AWB CTRL19 0x80 RW (shadow_ythre1 + shadow_ythre2 ≤
h
0x3FF)
tia nly
Bit[7:0]: shadow_uv_thre1[7:0]
lf
Bit[7:0]: shadow_uv_thre2[7:0]
0x5717 AWB CTRL23 0x10 RW (shadow_uv_thre1 + shadow_uv_thre2 ≤
0x3FF)
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Color temperature based AWB
0: Disable
1: Enable
0xC4B8 CT_AWB_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence. Default
value is random.
01: Long
10: Short
0xC4B9 AWB_WORK_MODE – RW
11: Combine
W
fid in
1: Enable
tia nly
0xC4BA AWB_FEEDBACK_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence. Default
lf
value is random.
default
address register name value R/W description
Bit[7:0]: Min statistic num for CT AWB[7:0]
0xC4D2~
W
fid in
0x5AB0~
AWB_R – R Debug Information for AWB Gain Control
0x5B1B
en g o
default
address register name value R/W description
Bit[7:2]: Not used
0x5D08 WBG_R13 – R
Bit[1:0]: Short_AWB_gain_Gr[9:8]
Complementary code
Bit[7:0]: Long_AWBoffset_B[7:0]
0x5D0D WBG_R18 – R
on ac
Complementary code
Bit[7:0]: Long_AWBoffset_GB[7:0]
0x5D0F WBG_R20 – R
Complementary code
en g o
Bit[7:0]: Long_AWBoffset_GR[7:0]
tia nly
0x5D11 WBG_R22 – R
Complementary code
Complementary code
Bit[7:0]: Long_AWBoffset_R[7:0]
0x5D13 WBG_R24 – R
or
Complementary code
Bit[7:0]: Short_AWBoffset_B[7:0]
0x5D15 WBG_R26 – R
Complementary code
Bit[7:0]: Short_AWBoffset_Gb[7:0]
0x5D17 WBG_R28 – R
Complementary code
default
address register name value R/W description
Bit[7:0]: Short_AWBoffset_Gr[7:0]
0x5D19 WBG_R30 – R
Complementary code
Bit[7:0]: Short_AWBoffset_R[7:0]
0x5D1B WBG_R32 – R
Complementary code
default
W
fid in
0x5213 DNS CTRL13 0x02 RW Bit[7:0]: noise_y for long exposure sub-pixel
lf
0x5215 DNS CTRL15 0x02 RW Bit[7:0]: noise_u[7:0] for long exposure sub-pixel
0x5217 DNS CTRL17 0x02 RW Bit[7:0]: noise_v[7:0] for long exposure sub-pixel
0x5218 DNS CTRL18 0x06 RW Bit[7:0]: dns_edgethre for long exposure sub-pixel
0x521A DNS CTRL20 0x02 RW Bit[7:0]: noise_y_list_0 for long exposure sub-pixel
0x521B DNS CTRL21 0x04 RW Bit[7:0]: noise_y_list_1 for long exposure sub-pixel
0x521C DNS CTRL22 0x08 RW Bit[7:0]: noise_y_list_2 for long exposure sub-pixel
0x521D DNS CTRL23 0x14 RW Bit[7:0]: noise_y_list_3 for long exposure sub-pixel
default
address register name value R/W description
0x521E DNS CTRL24 0x1E RW Bit[7:0]: noise_y_list_4 for long exposure sub-pixel
0x521F DNS CTRL25 0x28 RW Bit[7:0]: noise_y_list_5 for long exposure sub-pixel
0x5220 DNS CTRL26 0x32 RW Bit[7:0]: noise_y_list_6_l for long exposure sub-pixel
0x5223 DNS CTRL29 0x02 RW Bit[7:0]: noise_uv_list_0[7:0] for long exposure sub-pixel
C
0x5225 DNS CTRL31 0x04 RW Bit[7:0]: noise_uv_list_1[7:0] for long exposure sub-pixel
0x5227 DNS CTRL33 0x0C RW Bit[7:0]: noise_uv_list_2[7:0] for long exposure sub-pixel
0x5229 DNS CTRL35 0x28 RW Bit[7:0]: noise_uv_list_3[7:0] for long exposure sub-pixel
h
0x522B DNS CTRL37 0x32 RW Bit[7:0]: noise_uv_list_4[7:0] for long exposure sub-pixel
lf
0x522D DNS CTRL39 0x3C RW Bit[7:0]: noise_uv_list_5[7:0] for long exposure sub-pixel
or
0x522F DNS CTRL41 0x4C RW Bit[7:0]: noise_uv_list_6[7:0] for long exposure sub-pixel
0x523B DNS CTRL53 0x02 RW Bit[7:0]: noise_y for short exposure sub-pixel
default
address register name value R/W description
0x523D DNS CTRL55 0x02 RW Bit[7:0]: noise_u[7:0] for short exposure sub-pixel
0x523F DNS CTRL57 0x02 RW Bit[7:0]: noise_v[7:0] for short exposure sub-pixel
0x5240 DNS CTRL58 0x06 RW Bit[7:0]: dns_edgethre for short exposure sub-pixel
0x5242 DNS CTRL60 0x02 RW Bit[7:0]: noise_y_list_0 for short exposure sub-pixel
on ac
0x5243 DNS CTRL61 0x04 RW Bit[7:0]: noise_y_list_1 for short exposure sub-pixel
0x5244 DNS CTRL62 0x08 RW Bit[7:0]: noise_y_list_2 for short exposure sub-pixel
0x5245 DNS CTRL63 0x14 RW Bit[7:0]: noise_y_list_3 for short exposure sub-pixel
W
fid in
0x5246 DNS CTRL64 0x1E RW Bit[7:0]: noise_y_list_4 for short exposure sub-pixel
0x5247 DNS CTRL65 0x28 RW Bit[7:0]: noise_y_list_5 for short exposure sub-pixel
en g o
0x5248 DNS CTRL66 0x32 RW Bit[7:0]: noise_y_list_6 for short exposure sub-pixel
h
default
address register name value R/W description
Bit[7:0]: noise_uv_list_4[7:0] for short exposure
0x5252 DNS CTRL76 0x32 RW
sub-pixel
default
address register name value R/W description
h
default
address register name value R/W description
Bit[7:2]: Not used
Bit[1:0]: noise_slope[9:8] for long exposure sub-pixel
0x5286 CIP CTRL06 0x01 RW
Slope value used for calculating int_noise in
auto mode
multipliers
in auto mode
default
address register name value R/W description
Bit[7:6]: Not used
Bit[5:0]: min_sharpen_tm[5:0] for long exposure
0x5290 CIP CTRL10 0x20 RW sub-pixel
Min_sharpen_tm is used for sharpen_tm
computation in auto mode
RW
Threshold used for function of adaptive sharpen
0x5295 CIP CTRL15 0x08 RW Bit[5:0]: hthre[5:0] for long exposure sub-pixel
Threshold for high frequency signals
h
0x5297 CIP CTRL17 0x06 RW Bit[3:0]: hfreq_coef[3:0] for long exposure sub-pixel
Coefficients for high frequency signals
0x5298 CIP CTRL18 0x00 RW Bit[1:0]: efreq_coef[1:0] for long exposure sub-pixel
Coefficients for E frequency signals
0x5299 CIP CTRL19 0x08 RW Bit[5:0]: lthre[5:0] for long exposure sub-pixel
Threshold for low frequency signals
default
address register name value R/W description
Bit[7:0]: man_inv_noise[7:0] for long exposure sub-pixel
0x529D CIP CTRL1D 0x55 RW Inv_noise is input only in manual mode and is
used as threshold in some filters
default
address register name value R/W description
Bit[7:2]: Not used
Bit[1:0]: noise_slope[9:8] for short exposure sub-pixel
0x52C6 CIP CTRL46 0x01 RW
Slope value used for calculating int_noise in
auto mode
multipliers
components
default
address register name value R/W description
Bit[7:6]: Not used
Bit[5:0]: min_sharpen_tm[5:0] for short exposure
0x52D0 CIP CTRL50 0x20 RW sub-pixel
Min_sharpen_tm is used for sharpen_tm
computation in auto mode
default
address register name value R/W description
Bit[7:0]: man_inv_noise[7:0] for short exposure
sub-pixel
0x52DD CIP CTRL5D 0x55 RW
In_noise is input only in manual mode and is
used as threshold in some filters
sub-pixel:
0x52E1 CIP CTRL61 0x08 RW
Sharpen_tm is input only in manual mode and
is used for function of adaptive sharpen
h
tia nly
default
address register name value R/W description
Bit[7:0]: Long color matrix 1[15:8]
default
address register name value R/W description
Bit[7:0]: Long color matrix 2[15:8]
default
address register name value R/W description
Bit[7:0]: Long color matrix 6[7:0]
default
address register name value R/W description
Bit[7:0]: Long color matrix 11[15:8]
default
address register name value R/W description
Bit[7:0]: Short color matrix 3[7:0]
default
address register name value R/W description
Bit[7:0]: Short color matrix 8[15:8]
default
address register name value R/W description
Bit[7:0]: Short color matrix 12[7:0]
default
address register name value R/W description
W
fid in
tia nly
default
address register name value R/W description
Bit[7:4]: Not used
Bit[3]: Dark boost enable
0: Dark boost disable
1: Dark boost enable
Bit[2]: combine_uv_weight enable
C
0: Compensate disable
1: Compensate enable
Bit[0]: Compensate error enable
0: Compensate error disable
W
fid in
default
address register name value R/W description
Bit[7:4]: Not used
0x5409 COMB CTRL9 0x0A RW Bit[3:0]: comb_uv_thre_s2
UV threshold3 of short channel
tia nly
default
address register name value R/W description
0x5421 COMB CTRL33 0x80 RW Bit[7:0]: comb_uv_weight10
default
address register name value R/W description
Bit[7:1]: Not used
Bit[0]: Cut black level
0: Disable
1: Enable
0xC4B4 CUT_BL_EN – RW
When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence.
Default value is random.
1: Enable
DARKBOOST_AUTO_
0xC4B5 – RW
EN When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
W
fid in
1: Enable
AUTO_LOW_LEVEL_
tia nly
0xC4B6 – RW
EN When register value is 0x00, it means function is
disabled; all other values mean function is enabled.
This register value must be initialized by user and
must not be removed from start up sequence.
lf
default
address register name value R/W description
Bit[7:0]: Manual gamma[7:0]
0xC4C1 DB_GAIN_THRE_12 – RW
This register value will be automatically initialized by
on ac
0xC4C4 DB_AMT – RW
This register value will be automatically initialized by
sensor after powering up. Default value is random.
0xC4C5 DB_AMT_MIN – RW
This register value will be automatically initialized by
sensor after powering up. Default value is random.
or
default
address register name value R/W description
Bit[7:0]: Max dark boost gamma[7:0]
0x5A08~
on ac
Bit[2:0]: pLocalGainBuf2[10:8]
0x5C1E COMB_RO07 – R
Bit[2:0]: pLocalGainBuf3[10:8]
default
address register name value R/W description
0x5C27 COMB_RO16 – R Bit[7:0]: pLocalGainBuf7[7:0]
0x5C2C COMB_RO21 – R
Bit[2:0]: pLocalGainBuf10[10:8]
on ac
0x5C32 COMB_RO27 – R
Bit[2:0]: pLocalGainBuf13[10:8]
tia nly
Bit[2:0]: pLocalGainBuf14[10:8]
default
address register name value R/W description
Bit[7:3]: Not used
0x5C3E COMB_RO39 – R
Bit[2:0]: pLocalGainBuf19[10:8]
Bit[7:0]: AWBLogRatio_R[7:0]
0x5C4B COMB_RO52 – R
Two’s complement
or
Bit[7:0]: AWBLogRatio_G[7:0]
0x5C4D COMB_RO54 – R
Two’s complement
Bit[7:0]: AWBLogRatio_B[7:0]
0x5C4F COMB_RO56 – R
Two’s complement
default
address register name value R/W description
Bit[7:1]: Not used
0x5C51 COMB_RO58 – R Bit[0]: LogBlackX[16]
Two’s complement
Bit[7:0]: LogBlackX[15:8]
0x5C52 COMB_RO59 – R
Two’s complement
Bit[7:0]: LogBlackX[7:0]
0x5C53 COMB_RO60 – R
Two’s complement
Bit[7:0]: LogBlackY[15:8]
0x5C56 COMB_RO63 – R
Two’s complement
Bit[7:0]: LogBlackY[7:0]
W
fid in
0x5C57 COMB_RO64 – R
Two’s complement
Bit[6:0]: GlobalLogHDRGain[14:8]
default
address register name value R/W description
0x5C67 COMB_RO80 – R Bit[7:0]: nDarkBoostYThre2[7:0]
0x5C68~
NOT USED – – Not Used
0x5C69
Bit[7:0]: nLogE[15:8]
0x5C6A COMB_RO83 – R
Two’s complement
Bit[7:0]: nLogE[7:0]
0x5C6B COMB_RO84 – R
Two’s complement
tia nly
default
address register name value R/W description
Bit[7:6]: Not used
lf
Bit[7:0]: min_low_level
0x5482 NORM RW02 0xF8 RW
-128 to -16, complementary code
0x5485~
NORM RO – R Debug Information for NMLZ Control
0x5A98
default
address register name value R/W description
0x5C73 NML_RW04 – R Bit[7:0]: nNormalizeGain[7:0]
Bit[7:0]: RW_CurLowLevel[7:0]
0x5C78 NML_RW09 – R
C
default
address register name value R/W description
h
Bit[2:0]: edge_thre
000: 16
001: 32
0x5500 TOMP RW00 0x03 RW
010: 64
lf
011: 128
100: 256
101: 512
or
default
address register name value R/W description
0x5503 TOMP RW03 0x40 RW Bit[7:0]: Chip debug
0x550F~
TOMP RO – R Debug Information for TMAP Control
0x5511
Contrast Curve 1
lf
Contrast Curve 2
Contrast Curve 3
default
address register name value R/W description
Contrast Curve 4
Contrast Curve 5
Contrast Curve 6
on ac
Contrast Curve 7
W
fid in
Contrast Curve 8
h
Contrast Curve 9
lf
Contrast Curve 10
Contrast Curve 11
Contrast Curve 12
default
address register name value R/W description
Contrast Curve 13
Contrast Curve 14
Contrast Curve 15
on ac
default
address register name value R/W description
Max Curve Alpha
0x5A9C~
TMP_R – R Debug Information for TMAP Control
0x5AAD
default
address register name value R/W description
Bit[7:2]: Not used
0x5C91 TMP_RO22 – R
Bit[1:0]: pCurveList5[17:16]
0x5C9D TMP_RO34 – R
Bit[1:0]: pCurveList8[17:16]
tia nly
0x5CA1 TMP_RO38 – R
Bit[1:0]: pCurveList9[17:16]
default
address register name value R/W description
0x5CAA TMP_RO47 – R Bit[7:0]: pCurveList11[15:8]
0x5CB5 TMP_RO58 – R
Bit[1:0]: pCurveList14[17:16]
h
default
address register name value R/W description
Bit[7:2]: Not used
0x5CC2 TMP_RO71 – R
Bit[1:0]: pCurveGainList3[9:8]
default
address register name value R/W description
Bit[7:2]: Not used
0x5CD8 TMP_RO93 – R
Bit[1:0]: pCurveGainList14[9:8]
default
address register name value R/W description
Bit[7:5]: Not used
0x5CF1 TMP_RO118 – R
Bit[4:0]: pCurveSegBList5[4:0]
0x5CF5 TMP_RO122 – R
Bit[4:0]: pCurveSegBList9[4:0]
on ac
0x5CF9 TMP_RO126 – R
Bit[4:0]: pCurveSegBList13[4:0]
0x5CFA TMP_RO127 – R
Bit[4:0]: pCurveSegBList14[4:0]
tia nly
default
address register name value R/W description
Bit[7:3]: Not used
Bit[2]: fcnt_eof_sel
0x4200 FC_R0 0x00 RW
Bit[1]: fcnt_mask_dis
Bit[0]: fcnt_reset
default
address register name value R/W description
Bit[7]: Not used
Bit[6]: rblue_mask_dis
Bit[5]: data_mask_dis
Bit[4]: valid_mask_dis
0x4203 FC_R3 0x00 RW
Bit[3]: href_mask_dis
Bit[2]: eof_mask_dis
Bit[1]: sof_mask_dis
Bit[0]: all_mask_dis
default
W
fid in
1000: YUYV
tia nly
1001: YVYU
1010: UYVY
1011: VYUY
default
address register name value R/W description
Bit[7:4]: Not used
Bit[3]: r_fi_pt
VFIFO bypass
VFIFO_AFIFO_SRAM
0x4600 0x04 RW Bit[2]: r_16bitin
_CTRL
16-bit data into AFIFO
Bit[1]: r_sram_pt
Bit[0]: r_sram_nofrst
C
VFIFO_FIRST1_
0x4602 0x00 RW Bit[7:0]: r_first_pos_high
POSITION
VFIFO_FIRST1_
0x4603 0x00 RW Bit[7:0]: r_first_pos_low
POSITION
W
fid in
VFIFO_LINE_LENGTH
0x4606 0x00 RW Bit[7:0]: Manual set line length[15:8]
_MAN
lf
VFIFO_LINE_LENGTH
0x4607 0x00 RW Bit[7:0]: Manual set line length[7:0]
_MAN
or
VFIFO_HSYNC_
0x460A 0x00 RW Bit[7:0]: r_hsync_st[15:8]
START_POSITION
VFIFO_HSYNC_
0x460B 0xBF RW Bit[7:0]: r_hsync_st[7:0]
START_POSITION
default
address register name value R/W description
Bit[7:4]: Not used
Bit[3]: r_sof_clr_ram default 1
Bit[2]: r_st_mod
VFIFO_EMBD_LINE_ 0: PCLK cycles trigger (default)
0x460E 0x08 RW
CTRL 1: Byte size trigger
Bit[1]: Debug mode
Bit[0]: r_embd_en
Embedded line mode enable
VFIFO_EMBD_LINE_
0x460F 0x01 RW Bit[7:0]: Embedded line amount
NUM
C
0x4610 EMB_ST_PCNT_H 0x00 RW High Byte of Embedded Line Pcnt Start Point
0x4611 EMB_ST_PCNT_L 0x01 RW Low Byte of Embedded Line Pcnt Start Point
on ac
0x4612 EMB_ST_LCNT_H 0x00 RW High Byte of Embedded Line Lcnt Start Point
0x4613 EMB_ST_LCNT_L 0x01 RW Low Byte of Embedded Line Lcnt Start Point
W
fid in
Bit[7]: r_roi_sync_byp
Bit[6]: r_fr_comp
ROI output 8-bit data
Front comp 2-bit 0 or back
en g o
Bit[5]: r_full_dat_mod
0x4620 ROI_CTRL0 0x0E RW
Bit[4]: Not used
h
Bit[3]: r_roi_en_3
Bit[2]: r_roi_en_2
tia nly
Bit[1]: r_roi_en_1
Bit[0]: r_roi_func_e
Bit[6:4]: r_sc_fsz_delay
0x4621 ROI_CTRL1 0x31 RW
Bit[3:0]: roi_sync_cod_stv
Sync code start
or
default
address register name value R/W description
0x462B ROI_XOFFS_3 0x9C RW Window 3 Xoffset Low Byte
Bit[1:0]: roi_vsize1[9:8]
Bit[0]: roi_hsize3[8]
tia nly
default
address register name value R/W description
Bit[7:4]: Not used
Bit[3]: CCIR v select
0x4700 DVP_MOD_SEL 0x04 RW Bit[2]: CCIR f value
Bit[1]: CCIR656 mode enable
Bit[0]: HSYNC mode enable
default
address register name value R/W description
0x4702 DVP_HSYVSY_NEG_WIDTH 0x00 RW VSYNC Length, Pixel Count High Byte
low enable
tia nly
lf
or
default
address register name value R/W description
Bit[7]: Debug control
Bit[6:4]: Data bit swap
000: Data[9:0] outputs
through pin D[9:0]
001: Data[0:9] outputs
through pin D[9:0]
010: {Data[2:9], Data[1:0]}
outputs through pin
D[9:0]
011: {Data[7:0], Data[9:8]}
outputs through pin
C
D[9:0]
100: {Data[9:8], Data[0:7]}
outputs through pin
on ac
D[9:0]
101: {Data[9], Data[0:8]}
outputs through pin
D[9:0]
W
fid in
pattern
1: Repeat each data one
time in walking one
pattern. e.g., 8-bit
lf
pattern changes to
0x00 -> 0x00 -> 0x01 ->
0x01 -> 0x02-> 0x02 ->
or
default
address register name value R/W description
10: 10-bit walking one
pattern for 10 data pins
0x000 -> 0x001 ->
0x002-> 0x004->
0x008 -> 0x010 ->
0x020 -> 0x040 ->
0x080 -> 0x100 ->
0x200 -> 0x3FF
11: Debug option
Bit[0]: Walking one test pattern
enable
C
0: Disable
1: Enable
on ac
dvp_href_o
0x470D DVP_ROI_HREF_SEL 0x00 RW
010: ROI window 1 out via
dvp_href_o
011: ROI window 2 out via
lf
dvp_href_o
100: ROI window 3 out via
dvp_href_o
or
default
address register name value R/W description
Bit[7:1]: Not used
0x6800 EMB_LINE_EN 0x00 RW
Bit[0]: emb_line enable
default
address register name value R/W description
Bit[7:2]: Not used
0x6802 EMB_LINE_TAG 0x01 RW
Bit[1:0]: emb_line tag[1:0]
Bit[7:4]: s2h_width
0x6803 EMB_LINE_SOF_CTRL 0x11 RW
Bit[3:0]: sof_width
default
h
default
address register name value R/W description
Group Write Command Register
0x6F07 PARI_ADDR_MIN 0x00 RW
Address, Must Be 0x00
0x6F0C~
PARI_MASTER_SEL – RW Debug Control
0x6F1F
on ac
default
address register name value R/W description
h
Macro-code Registers
tia nly
0xD000~
MACRO-CODE REGISTERS – – Can not be deleted. Sensor functions are
0xDFFF
combined with such settings
lf
or
8 operating specifications
8.1 absolute maximum ratings
VDD-A 4.5V
supply voltage (with respect to ground)
VDD-IO 4.5V
C
a. exceeding the absolute maximum ratings shown above invalidates all AC and DC electrical specifications and may
result in permanent damage to the device. Exposure to absolute maximum rated conditions for extended periods
W
fid in
tia nly
parameter range
lf
a. sensor functions in the operating range; however, some image quality changes may be noticed at the temperature
extremes
8.3 DC characteristics
supply
VDD-A supply voltage (analog) 3.14 3.3 3.47 V
IDD-A 60 85 mA
a
IDD-IO 30 mA
on ac
IDDS-PWDN-A 5 µA
IDDS-PWDN-IO 10 µA
W
fid in
c
VIH SIOC and SIOD 1.26 1.8 2.3 V
8.4 AC characteristics
ADC parameters
B analog bandwidth 48 MHz
tia nly
C
on ac
W
fid in
en g o
h
tia nly
lf
or
9 mechanical specifications
9.1 physical specifications
L2 A A
B B
C J2 C
D D
E E center of BGA (die) =
F F center of the package
B G
WX Y Z G
C
H ABCD H
J J
K K
on ac
L L
M M
N N
D
J1
W
fid in
A
S1
top view bottom view
(bumps down) (bumps up)
C2 glass die C4 note mark code:
W - OVT product version
X - year the part is assembled
en g o
tia nly
tia nly
lf
or
240.0
tL TL
220.0
note
Tmax
200.0 reflow
180.0 The OV9623 uses a
temperature (°C)
120.0
100.0
80.0
C
ramp up
60.0
40.0
on ac
20.0 T0 Tf
0.0
0
20
40
60
80
100
120
140
160
180
200
220
240
260
280
300
320
340
360
380
400
420
440
460
480
500
520
540
W
fid in
time (sec)
note
en g o
soaking heating from 150°C to 200°C 90 ~ 150 seconds packages use underfill
tia nly
as part of camera
ramp up B (tL to TP) heating from 217°C to 245°C temperature slope ≤ 3°C per second assembly process.
peak temperature maximum temperature in SMT 245°C +0/-5°C (duration max 30 sec)
lf
ramp down A (TP to TL) cooling from 245°C to 217°C temperature slope ≤ 3°C per second
or
ramp down B (TL to Tf) cooling from 217°C to room temperature temperature slope ≤ 2°C per second
C
on ac
W
fid in
en g o
h
tia nly
lf
or
10 optical specifications
10.1 sensor array center
5510.4 µm
array center
on ac
package center
W
fid in
OV9623
tia nly
top view
lf
note 2 as most optical assemblies invert and mirror the image, the chip is
typically mounted with pin A2 to A13 oriented down on the PCB.
10
8
CRA (degrees)
6
C
4
on ac
2
W
fid in
CRA
0
en g o
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
h
tia nly
lf
or
Stray infrared light can affect image quality in various ways, from color crosstalk to internal reflections from metal
surfaces. To reduce these artifacts, OmniVision recommends an IR-cut filter that maintains near 0% transmission from
700nm to 1200nm. The camera application, acceptable artifacts, and packaging type will dictate the exact IR-cut
specifications. For further assistance, contact your regional OmniVision FAE.
R
Gr
Gb
B
C
on ac
QE
W
fid in
en g o
h
tia nly
wavelength (nm)
lf
or
revision history
• initial release
• in section 4.3, added "VTS is adjusted by registers (0x6E42[7:0], 0x6E43[7:0]). The reference
on ac
initialization settings must be used for these two registers to be valid." to end of section
• in table 4-6, added register 0xC2ED
• in table 7-7, added register 0xC2ED
W
fid in
• in key specifications, changed operating temperature range to "-40°C to +105°C sensor ambient
temperature (operating sensor ambient temperatures above 60°C may result in degraded image
h
quality)"
tia nly
• in table 8-2, changed operating temperature range to "-40°C to +105°C sensor ambient
temperature"
lf
temperature"
• in table 8-2, changed operating temperature range to "-30°C to +85°C junction temperature"
• in chapter 5, updated section 5.12 completely with new content and removed section 5.14
• in chapter 7, removed table 7-10
tia nly
lf
or
tia nly
lf
website: www.ovt.com
Powered by TCPDF (www.tcpdf.org)