Innovations Embedded

Optical Image Stabilization (OIS)

White Paper
 Optical Image Stabilization (OIS) White Paper

Introduction

Whether capturing still images or recording for pixel count density and resolution quality
moving video, image stabilization will always have increased drastically over the last decade
be a major factor in reproducing a near perfect and will continue to rise. The market shift to
digital replica. A lack thereof will result in image compact mobile devices with high megapixel
distortion through pixel blurring and the creation capturing ability has created a demand for
of unwanted artifacts. While media capturing advanced stabilization techniques. Two
devices such as digital cameras, digital methods, electronic image stabilization (EIS)
camcorders, mobile phones, and tablets have and optical image stabilization (OIS), are the
OIS Target
decreased in physical size, their requirements mostMarket
common implementations.

High

SLR Camera

Mirrorless Camera
Resolution

Point & Shoot Camera

10Mpixel

Smartphone
Low
Now Future
Timeline

Figure 1. Optical Image Stabilization Target Market

2
 Optical Image Stabilization (OIS) White Paper

Image Stabilization Principles

Image stabilization is used to reduce blurring frames to reduce the sense of motion. Though
associated with motion and/or shaking of the advantage with this method is the ability to
the camera during the time the image sensor create inexpensive and compact solutions, the
is exposed to the capturing environment. resulting image quality will always be reduced
However, it does not prevent motion blur due to image scaling and image signal post-
caused by movement of the target subject or processing artifacts, and more power will be
extreme movements of the camera itself, only required for taking additional image captures
the relatively small shaking of the camera lens and for the resulting image processing. EIS
by the user – within a few optical degrees. This systems also suffer when at full electronic
camera-user movement can be characterized zoom (long field-of-view) and under low-light
by its pan and tilt components, where the conditions.
angular movements are known as yaw and
pitch, respectively. Camera roll cannot be
compensated since 'rolling' the lens doesn't Yaw

actually change/compensate for the roll motion,

Pitch
and therefore does not have any effect on the
image itself, relative to the image sensor.
Roll

EIS is a digital image compensation technique

which uses complex algorithms to compare
frame contrast and pixel location for each
changing frame. Pixels on the image border
provide the buffer needed for motion
Roll
compensation. An EIS algorithm calculates the Pitch
subtle differences between each frame and
Yaw
then the results are used to interpolate new
Figure 2. Axes of Motion

3
 Optical Image Stabilization (OIS) White Paper

OIS vs. EIS

200% Zoom

Figure 3. OIS and EIS Image Quality Comparison

OIS Behavior

OIS is a mechanical technique used in imaging of the entire module, including the fixed lens
devices to stabilize the recording image by and image sensor. Module tilt allows for a
controlling the optical path to the image sensor. greater range of movement compensation
The two main methods of OIS in compact by the OIS system, with the largest tradeoff
camera modules are implemented by either being increased module height. Minimal image
moving the position of the lens (lens shift) or distortion is also achieved with module tilt due
the module itself (module tilt). to the fixed focal length between the lens and
image sensor. Overall, in comparison to EIS,
Camera movements by the user can cause
OIS systems reduce image blurring without
misalignment of the optical path between the
significantly sacrificing image quality, especially
focusing lens and center of the image sensor.
for low-light and long-range image capture.
In an OIS system using the lens shift method,
However, due to the addition of actuators and
only the lens within the camera module is
the need for power driving sources compared
controlled and used to realign the optical path
to no additional hardware with EIS, OIS
to the center of the image sensor. In contrast,
modules tend to be larger and as a result are
the module tilt method controls the movement
more expensive to implement.

4
 Optical Image Stabilization (OIS) White Paper

Lens Shift Method Module Tilt Method

Lens within the module is moved Entire camera module is moved

Movement Direction Movement Direction

Lens Lens
Magnet Photoreflector

Hall Sensor

Image Sensor
Image Sensor

Figure 4. Main Methods of OIS Compensation

OIS Module Components

An OIS system relies on a complete module for the user's natural movement. With lens
of sensing, compensation, and control shift, Hall sensors are used to detect real-
components to accurately correct for unwanted time X/Y locations of the lens after taking into
camera movement. This movement or vibration consideration actuator mechanical variances
is characterized in the X/Y-plane, with yaw/ and the influence of gravity. The controller uses
pan and pitch/tilt movements detected by a separate internal servo system that combines
different types of isolated sensors. The lens shift the lens positioning data of the Hall sensors
method uses Hall sensors for lens movement with the target lens position calculation from the
detection while the module tilt method uses gyroscope to calculate the exact driving power
photoreflectors to detect module movement. needed for the actuator to reposition the lens.
Both methods require a gyroscope in order With module tilt, the process is similar but the
to detect human movement. ROHM’s OIS module’s location is measured and repositioned
controllers use gyroscope data within a lens instead of just the lens. With both methods, the
target positioning circuit to predict where the new lens position realigns the optical path to
lens needs to return in order to compensate the center of the image sensor.

5
 Optical Image Stabilization (OIS) White Paper

Static Alignment

Lens Image
for OIS sensor

Optimal
Light of optical
subject
path

Phone Movement

Without OIS System With OIS System

A A
B

The difference between A & B Lens movement is compensated by the OIS

causes image degradation controller and the optical path is corrected

Figure 5. Lens Shift OIS Principle

6
 Optical Image Stabilization (OIS) White Paper

Y-Axis
OIS
Actuator

MIPI / MIDI
Application Power-Save
Image X-Axis Lens X-Axis
Processor Sensor Hall
Sensor
(Autofocus OIS
Actuator
I2C Optional)

Y-Axis
Hall
Sensor

OIS SPI OIS

Gyroscope Controller

Figure 6. General OIS Block Diagram

OIS System Control

OIS control is designed to be very simple from internal processor and RAM for the calculations.
the customer’s standpoint, consisting of simply
ROHM's OIS Architecture
ON/OFF and enable/power-save modes. The
only other commands are either optional manual ROHM offers two OIS controller architectures,

control of the lens in the X/Y plane or changing including a fully programmable ARM Cortex-M0

the OIS performance based on ambient processor with custom programmable digital

conditions such as day, night, sports, picture, signal processing for 'gyroscope signal

video, or viewfinder. This allows for minimal I2C processing' and 'servo control', as well as

traffic from the host application processor to ROHM’s custom, fully programmable RMCU

the OIS controller and simplifies software driver processor with integrated programmable

development for the end customer. All of the 'gyroscope signal processing' and 'servo

actual OIS control algorithms are performed control'. All of the OIS work memory and control

autonomously on the controller itself, using the calculations are performed on the OIS controller

7
 Optical Image Stabilization (OIS) White Paper

Power Down Mode

Lens Position:
Floating

Initialize (OIS OFF)

OIS Mode Setting

Picture / Movie / Viewfinder
Day / Night / Sport

Lens Position:
Stable (fixed position)

OIS ON Manual Control

(Optional)

OIS driver auto-control Host processor

controlled

Mode selection available via I2C commands

Figure 7. Standard OIS Operation Modes

8
 Optical Image Stabilization (OIS) White Paper

itself without the need for the external host temperature drift while at the same time not
processor’s computational power or external removing intentional pan and tilt movement
memory for storing calculation variables. Both by the camera user. The controllers can be
offerings achieve the smallest chip size, lowest implemented into either lens shift, module tilt,
external component count, and smallest overall or other less commonly used control systems
mounting area on the market. such as lens tilt. Customizable OIS control
software is also included for automatic lens
With industry-leading experience in OIS
control, automatic pan-tilt detection, and
architecture, ROHM’s lineup of OIS controllers
access to different programmable capturing
offer full control of the X- and Y-axis voice coil
modes and calibration settings.
motor (VCM) drivers, Hall amplifier and current
drivers, photoreflector drivers, I2C interface, Image Stabilization Performance
PLL oscillators, 12-bit ADC, SPI master
Image stabilization is measured by suppression
interface for digital gyroscope (along with
ratio (SR) and is utilized to gauge OIS
support for analog gyroscopes), and many
performance. The SR is calculated using a
other features. The lineup also includes options
spatial test chart with a target pattern. Images
that support integrated drivers for autofocus,
of the target pattern are captured with OIS ON/
neutral density filter, or shutter functions.
OFF and with/without vibration. The images
Selectable PWM/BTL linear operation is
are then used to compute a ratio comparing the
supported for choosing either improved VCM-
amount of blur measured in the images, which
driving power efficiency or improved image
determines the final SR. The test is typically
quality. In addition, ROHM's controllers feature
used to provide a final guarantee that all of the
class-leading power consumption due to wafer
components in the OIS system are functioning
processing and optimized chip architecture.
properly.
Both of ROHM’s proprietary 'servo control'
and 'gyroscope signal processing' circuits use The pictures in Figure 8 are representations of

a unique digital filter design that dynamically motion blur of the target pattern. The DSTATIC

compensates for gyroscope and actuator image represents the ideal image where there

9
 Optical Image Stabilization (OIS) White Paper

The dashed blue sections on the

generic test pattern are used for
calculating suppression ratio
Measurements are done for both
the X- and Y-axis

No Movement Movement Blurring Reduced Blurring

Vibrate OIS ON

DSTATIC DOISoff DOISon

OIS OFF OIS OFF OIS ON
Vibration OFF Vibration ON Vibration ON

Figure 8. Generic Spatial Test Chart Reference Image

DOISoff — Dstatic
Suppression Ratio [dB] = 20log [ ]
DOISon — Dstatic

Equation 1. Suppression Ratio Calculation

is no vibration/motion of the image. An ideal the zoomed white distance due to the blurring
OIS system attempts to match the quality of a effect from the image sensor when an image
still image with no motion blur, and the DSTATIC is captured. The observed amount of blur is
image is the benchmark when calculating the what needs to be corrected, or suppressed, in
SR/performance of the OIS system. In this order to have the DOISoff image match the DSTATIC
example, the DSTATIC image has the shortest image as closely as possible. The DOISon image
zoomed white area distance due to no represents the actual benefit of the OIS system
movement/blurring of the captured image. The being tested. In this example, the DOISon image
DOISoff image represents the appearance of the is vibrating/moving while the image stabilization
image when the image is vibrating/moving, with system is enabled. The image’s blurring is
no image stabilization being used. As a result, suppressed due to the stabilization and the
the DOISoff image exhibits much more blurring distance of the zoomed white area is less when
compared to the other images, which increases compared to the DOISoff image. After all three

10
 Optical Image Stabilization (OIS) White Paper

images have been captured, the blurring effect accurate simulation tools. Most OIS controller
of each image is measured as a function of suppliers are able to sufficiently simulate the
pixel count by counting the amount of pixels ideal performance of golden OIS components
within the width of the zoomed white area and such as the actuator, however ROHM has
then using Equation 1 to calculate the final SR. developed specialized simulation tools that
This process is repeated for each of the desired allow for not only proper simulation of OIS
image shaking frequency performance targets components, but real-world OIS component
and for each axis. simulations as well. The actuator, being a
mechanical system, has the most variance
Simulating the OIS System
due to the manufacturing process as well as
Proper OIS operation requires simulating the external stimuli. As such the actuator is a very
performance of the entire system, taking into complex component to realistically simulate,
account the interaction of all of the OIS system making it very difficult to accurately simulate the
components. ROHM is able to offer market- entire system's SR. ROHM's simulation tools
leading OIS control algorithms due to extremely have been market proven to provide accurate,

Figure 9. Real-World OIS Performance vs. ROHM's Simulated OIS Performance

11
 Optical Image Stabilization (OIS) White Paper

real-world results, allowing for fast lead-time calibration process has been performed, the
turnaround in implementing custom firmware calibrated data is used by the OIS controller to
features required by customers. intelligently modify the control of the OIS system
and its components. ROHM's calibration
Manufacturing Supply Chain
procedure is defined and market-tested, and
The typical OIS camera supply chain consists ROHM provides all of the necessary calibration
of an actuator vendor, a gyroscope vendor, a software and support so that the customer
Hall sensor or photoreflector vendor, a module doesn’t need to develop any custom factory
integrator, and ROHM's OIS controller. In calibration procedures.
addition to the controller, ROHM manufacturers
Summary
Hall sensors and photoreflectors, allowing
ROHM to have the most comprehensive ROHM offers a control and calibration software
system knowledge for providing the best OIS package that minimizes the amount of resources
results. required by the customer for implementing
OIS within their product and ensures market-
Assembly Calibration Process
leading performance with the fastest product
OIS systems require a calibration process for time-to-market. Additionally, ROHM has
proper operation. All of the components within formed partnerships and established a working
the OIS system posses individual manufacturing history with all of the required sensor/actuator
variances and exert influences on each other manufacturers and module integrators, making
after the assembly process, in addition to it possible to provide a complete, fully integrated
mechanical misalignment variances created OIS support package that enables customers
by the assembly process itself. A properly to meet their OIS performance and production
functioning system requires the OIS controller schedule requirements for launching their OIS-
to know the subtle sensitivity variances of embedded camera products on the market.
the actuator, gyroscope, Hall sensors, and Contact a ROHM representative today to help
controller, as well as the mounting variances start your OIS design.
introduced by the assembly process. Once the

12
ROHM Semiconductor
6815 Flanders Drive, Suite 150
San Diego, CA 92121

www.rohm.com | 1.888.775.ROHM
NOTE: For the most current product information, contact a ROHM sales representative in your area.
ROHM assumes no responsibility for the use of any circuits described herein, conveys no license under any patent or other
right, and makes no representations that the circuits are free from patent infringement. Specifications subject to change
without notice for the purpose of improvement.
The products listed in this catalog are designed to be used with ordinary electronic equipment or devices (such as audio
visual equipment, office-automation equipment, communications devices, electrical appliances and electronic toys). Should
you intend to use these products with equipment or devices which require an extremely high level of reliability and the
malfunction of which would directly endanger human life (such as medical instruments, transportation equipment, aerospace
machinery, nuclear-reactor controllers, fuel controllers and other safety devices), please be sure to consult with our sales
representative in advance.
© 2013 ROHM Semiconductor USA, LLC. Although every effort has been made to ensure accuracy, ROHM accepts no
responsibility for errors or omissions. Specifications and product availability may be revised without notice. No part of this
document represents an offer or contract. Industry part numbers, where specified, are given as an approximate comparative
guide to circuit function only. Consult ROHM prior to use of components in safety, health or life-critical systems. All trademarks
acknowledged.

1.800.775.ROHM www.rohm.com CUS13004-Ver.1