Body area sensor network for gesture recognition of autism disorder children

Autism spectrum disorder (ASD) is neurological dis-order which effects the development of communication skills and social interactions in children. People with ASD have restricted interest and repetitive behaviors. ASD hurts the person?s ability to function properly in school, work, and other areas

Project Title

Body area sensor network for gesture recognition of autism disorder children

Project Area of Specialization

Wearables and Implantable

Project Summary

Autism spectrum disorder (ASD) is neurological dis-order which effects the development of communication skills and social interactions in children. People with ASD have restricted interest and repetitive behaviors. ASD hurts the person’s ability to function properly in school, work, and other areas of life.

In this project, we propose design and implemention of a gesture recognition system for deaf and dumb people that converts the sign language into verbals.  This project will facilitate and improve a communication method between the verbal and nonverbal people. Many researches have been conducted to develop sign to speech translator, most of these are based on image-processing based system. The efficiency of these techniques depends on the image quality, which is significantly affected by the light intensity and situation of the implemented camera (resolution, focusing, alignment). These factors can affect the image quality and thus reduces the sys capability to recognize the gesture. In this project we will develop the following modules 1). Body-worn sensors network platform 2). Communication link between body-worn sensors network and raspberry-pi system-on-chip based processing system. 3) Signal conditioning and features extraction from sensors data to improve processing time and accuracy. 4) Classification models performance comparison for better accuracy achievement. The proposed systems will be aid-on to improve the social life of ASD, make them a good citizen, and assist them in communicating with normal people.

Project Objectives

1. Multisensor fusion platform for ASD children complex sign language recognition
2. Communication link for transferring body sensors data to processing system
3. Performance comparison of time-domain and frequency domain signal information extracted for better accuracy and time processing
4. Collection of Dataset from ASD children
5. Performance comparison of recognition algorithm such as neural networks and deep convolutional network
6. Translating the sign language to normal human understanding language

Project Implementation Method

figure shows the process diagram of the propsed system which consist of two main parts. First is wearable part, which consist of two types of sensor, Motion sensors and flex sensors. The motion sensors are used to approximate the motion of hand and head an the flex sensor will sense the bend in the fingers. Sensors are connected to controller which will collect data and transmit it to processing part wirelessly. The second part of the hardware is Raspberry-Pi (the processing part). After receiving the data from wearable part, Raspberry-Pi extracts varoius features from the data, and uses the trained classifiers and labels to recognize the physical activity which is then dispalyed on the LCD.

Benefits of the Project

• ASD individuals are facing serious problems communicatig with normal people.
• Due to this communication barrier they cannot excel in work places, even in public places they face serious issues.

This project will help the deaf and dumb people in the following ways;

• Provide a medium for ASD children to communicate with normal people.
• Provide an independent life for non-verbal people.
• This project will enable ASD children to work like normal people.

Technical Details of Final Deliverable

• Motion sensor MPU6050
  • Three-axis accelerometer(X,Y,Z)
  • Three-axis gyroscope(Z,Y,Z)
• Flex Sensors (bend in term of resist)
• Controller that will collect data from sensor and transmitt it wirelessly to the processing part.
• Raspberry-Pi based processing system that will extract various features and analyze and recognze the gesture.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Medical

Other Industries

Education , IT

Core Technology

Wearables and Implantables

Other Technologies

Artificial Intelligence(AI), Internet of Things (IoT)

Sustainable Development Goals

Good Health and Well-Being for People, Reduced Inequality, Peace and Justice Strong Institutions

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com