Overview / Usage

Drishti, is a computer vision-based assistive navigation system designed address the challenges faced by the visually impaired. By leveraging Intel’s AI Analytics tool advanced computer vision algorithms and depth estimation techniques.

Using a mobile device's camera and our web app, Drishti captures live video input and analyzes the visual data to identify objects in the user's environment. The system then employs depth estimation techniques to precisely measure the distance between the user and detected objects. This information is processed in real-time to provide immediate audio feedback and warnings to the user.

With Drishti, the visually impaired can gain a deeper understanding of their surroundings, identify potential obstacles, and receive timely alerts to avoid hazards. By enhancing their spatial awareness and providing reliable guidance, Drishti promotes safer and more independent navigation for the visually impaired.

Methodology / Approach

First I have to detect object so I have used the YOLO algorithm which detect the object and in return tell the detected object according to time stamp. For these detect object I have made an api using flask, and then connect with the react app. In react app, the api gives the information of objects and which give the information about the object converting text to speech, which help visually impaired person to detect the object .

Technologies Used

I have used MERN Stack, OpenCV, Flask

Repository

https://github.com/Harsha1309/One_API_Intel_Hackathon