One of the most challenging tasks for visually impaired persons is the ability to identify people in various social situations. Recognition through voice is commonly used to identify people. However, this may be difficult or unfeasible in some environments.
This project focuses on development of a face recognition system using an Android device and related server-cloud based support system. The system will allow people to enrol their faces into the recognition system by using the camera of the device or by uploading an image directly to the server. People will be identified using the Android application by
matching input faces with those stored in the server database. When a person is identified, a message will be given on a wireless earpiece to notify the user. The user will enrol the names of new persons using speech recognition system.
Methods and Development
This project will use Artificial Intelligence methods, particular neural network based face recognition for blind persons. Blind persons use special software such as Orca for using computer devices. The use of Open Computer Vision Library (OpenCV) and Integration with Android Devices will be in the initial stage.
Integration of OpenCV based Android face image processing algorithms.
The integration of the Orca Software will also be done an the testing and development of intelligence neural network based face recognition will be done later.
Software release and updates. Application of system in difficult lighting conditions.
Project Team Members
Principal Investigator: Dr. Rohitash Chandra and MSc Research Student: Shonal Chaudhary
Funding and Sponsorship: Please contact Dr. Rohitash Chandra if your organisation can fund this project partially.