This project attempts to create a system which would bring in added ease to the visually impaired, through our nagivation, obstacle-detection, obstacle distance identification and speech-driven system to seamlessly integrate applications like Ola, Uber, etc. This was built during the PW-Hacks Jan-2023.
- Visually impaired means that one can hardly fathom to cross the busy day-to-day life laden with obstacles, nay, even navigating through a micro-environment like home or work-place is a challenge.
- We attempt to solve this through a two way model. That is we work with both the individual and his/her guardian to create a safe navigating experience for the person.
- Hosting this application in any hand-held device like a mobile for ease of access
- Flask (Python)
- HTML
- CSS
- Kivy (Python)
- Android Studio
Install the pre-requisite modules namely Flask and Kivy and other modules as applicable Running the Detection Code independently
python object classification/yolo_opencv.py
python person distance/distance.py
python speech recognition/speech.py
python main.py
python kivy3.py
- Requires KivyMD as well
- The Web-application calls the voice-assisted system and is recieptive to user commands
- There is always an alert for obstacle and when the obstacle arrives beyond a certain safe distance from an individual a beep sound is raised
- The user can easily book an Ola by just an instructions
- We faced issues in figuring out the weights in the model and in the design of the mobile application linked to the cloud, but were able to find that use of COCO dataset and Firebase which enabled seamless deployment
- Integrating Face Recognition
- Including Map API for Navigation
- https://www.loom.com/share/81663c4ab2684b14ac4adac4f243cd56
- https://www.loom.com/share/0705a9717f444843a6f9e510d2e1a250
- https://www.loom.com/share/aa7c4922a2644034938e23dc5b33d81a