SMART BLIND STICK USING

ARTIFICIAL INTELLIGENCE BASED

IMAGE PROCESSING

1
 INTRODUCTION
 Blind and visually impaired persons always depend upon others
for their locomotion.
 Eyes are prime sense organ in perceiving the outside environment.
 Travelling in such outside environment is a challenge because the
blind cannot depend upon his own eyes.

2
 OBJECTIVE

To assist visually impaired people with obstacle

free path-finding
To convert Image into the audio output(Speech)
through the use of Artificial intelligence .
Detect the image and convert it into audio file.

3
 ABSTRACT
 In this project , a system is implemented which navigates the
person along his path.
 The object in front of the visually impaired person can be
detected using image recognition with AI.
 The information about object type in front of the person, can
be communicated via headphones/earphones.
 This will help the person to walk smoothly and also keeps
the guardian of the person informed about their
whereabouts.
 SMS alert with GPS location send in case of emergency
time.

4
 EXISTING METHOD
 Smart blind stick is used which is embedded with some sensors
which will tell the atmospheric condition of the atmosphere .
White canes and guide dogs.
 Electronic canes using laser technology.
 Laser Cane.
 Telecast and Minitact are such laser canes.
 GPS based Navigation devices.

5
 DRAWBACK

 Only considering environmental factors.

 Didn’t have GPS and GSM use.
 Only detect obstacles.
 Large in size and has limited scanning area.

6
 PROPOSED SYSTEM
 To find difficulties in detecting obstacles during walking in the
street for blind and visually impaired people and aged person.
 The priority information is provided to the user through audio
feedback mechanisms.
 The object in front of the visually impaired person can be detected using
image recognition with AI.
 ARDUINO controller used for voice processing

7
 ADVANTAGES
 It detects wet floors to avoid slipping accidents.
 It detects floor-level and knee-level obstacles.

8
Block diagram

9
 COMPONENTS

 Controller
 Camera
 Power supply

10
 Novelty

• Image processing based guidance to avoid sensor faults

• Low cost solution

11
 Working methodology -Object Detection
• Object Detection Method Based on Haar-Like
Feature Classifier Cascade and AdaBoost Learning
originated by Paul Viola and Rainer Lienhart
Feature Representation and Calculation, Feature
Selection, Classifiers Training and Cascade
Training

12
 Object Detection
• Haar-like Features
The feature value will be calculated by the
weighted sum of the pixels cover by the white
rectangles subtracted by the weighted sum of the
pixels covered by the dark rectangles

13
 Object classification process
• Cascade TrainingF Max acceptable false alarm rate per layer d
Min acceptable hit rate per layer Ftarget
overall false alarm rate.
• P positive examples set N negative examples
set
• Fi, Di false alarm rate and hit rate of layers
1i
• i0 F01.0 D0 1.0
• Loop while FigtFtarget
• i i1 ni0 Fi Fi-1
• While Fi gtf x Fi-1
• Use P and N to train a classifier with ni
features
• Evaluate Fi and Di of current cascade
• Decrease threshold for the i-th classifier to
have a hit rate of at least d x Di-1
• N Empty
• If Fi gtFtarget, N false detections of current
cascade on negative examples
14
 Hardware & software modules
• Arduino controller
• ISD1700 –Voice processing IC
• Head set
• Arduino programmer
• AI-Front end application

15
ISD1700
• The Winbond® ISD1700 ChipCorder®
Series is a high quality, fully integrated,
single-chip multimessage voice record and
playback device ideally suited to a variety
of electronic systems.
• The message duration is user selectable in
ranges from 26 seconds to 120 seconds,
depending on the specific device.
• The sampling frequency of each device can
also be adjusted from 4 kHz to 12 kHz
with an external resistor, giving the user
greater flexibility in duration versus
recording quality for each application
Arduino
• Arduino is an open source computer hardware
and software company, project and user
community that designs and manufactures
microcontroller -based kits for building digital
devices and interactive objects that can sense
and control objects in the physical world.
• The project is based on microcontroller board
designs, manufactured by several vendors,
using various microcontrollers.
• These systems provide sets of digital and
analog I/O pins that can be interfaced to
various expansion boards ("shields") and other
circuits.
• The boards feature serial communications
interfaces, including USB on some models, for
loading programs from personal computers.
 Literature survey

S,no Title Author Method

1 A System-Prototype Representing 3D NikolaosBourb 3-D space sensation that is
Space via Alternative-Sensing for akis interpreted by our computer
Visually Impaired Navigation,2013 vision prototype system
and transferred to the user
via a vibration array

2 A navigation aid for blind M. Bousbia- a navigation aid for the

People,2011 Salah blind based on a
microcontroller with
synthetic speech output

18
 Literature survey
S,no Title Author Method
3 Learning rotation-invariant G. Cheng, P. rotation-invariant CNN
convolutional Zhou (RICNN) model for
neural networks for object detection advancing the performance
in VHR optical remote sensing of object detection
images,2016

4 Wearable obstacle avoidance D. Dakopoulos portable/wearable obstacle

electronic and N. G. detection/avoidance
travel aids for blind,2010 Bourbakis systems

19
 Literature survey
S,no Title Author Method

5 A novel obstacle detection method M. C. Kang, S. object at risk of collision

based on deformable grid for the H. Chae based on the degree of the
visually impaired,2015 deformation of the DG

6 Magnetic tensor sensor and K. M. Lee, M. geomagnetic field effects

wayfinding Li
method based on geomagnetic field
effects with applications for
visually impaired users,2016

20
 CONCLUSION

• In this project , a guidance navigation system for blind people was presented. The
approach of the system is based on the idea that a blind pedestrian can be assisted by
spoken instructions from a controller by integrating sensors like ultrasonic and AI
based object recognition . The scenarios for the usability testing of overall system
for blind to evaluate the usability were presented.

21
 REFERENCES
 C. Yan et al., “Efficient parallel framework for HEVC motion estimation on many-
core processors,” IEEE Trans. Circuits Syst. Video Technol., vol. 24, no. 12, pp. 2077–
2089, Dec. 2014.
 X. Yang, S. Yuan, and Y. Tian, “Assistive clothing pattern recognition for visually
impaired people,” IEEE Trans. Human-Mach. Syst., vol. 44, no. 2, pp. 234–243, Apr.
2014.
 C. Ye, S. Hong, X. Qian, and W. Wu, “Co-robotic cane: A new robotic navigation aid
for the visually impaired,” IEEE Syst., Man, Cybern. Mag., vol. 2, no. 2, pp. 33–42,
Apr. 2016.
 D. Zhang, J. Han, J. L, Y. S, and C. X, “Revealing event saliency in unconstrained
video collection,” IEEE Trans. Image Process., vol. 26, no. 4, pp. 1746–1758, Apr.
2017.
 D. Zhang, J. Han, C. Li, J. Wang, and X. Li, “Detection of co-salient objects by
looking deep and wide,” Int. J. Comput. Vis., vol. 120, no. 22, pp. 215–232, 2016.
 P. A. Zientara et al., “Third eye: A shopping assistant for the visually impaired,”
Computer, vol. 50, no. 2, pp. 16–24, Feb. 2017
22
Thanking you

23
24