Initated the project:
- Clone the github repo in you local system.
- 2)setup firebase and initate realtime database and download the configuration file of that and place the file in your project folder.
- you are good to go.
VisionAid: A Computer Vision Solution for the Visually Impaired Overview VisionAid is an innovative project aimed at assisting visually impaired individuals in navigating their environment, recognizing people, and reading text. This project leverages computer vision technology to enhance the independence and quality of life of its users.
Team Details Team Name: NCoderX
Problem Statement Title: VisionAid: A Computer Vision Solution for the Visually Impaired
Visually impaired individuals face numerous challenges in navigating their environment, recognizing people, and reading text. VisionAid aims to provide an all-in-one solution to enhance their independence and quality of life.
Solution Overview The VisionAid prototype will function as follows:
Object Detection and Classification: Identify and classify objects in real-time. Face Recognition: Recognize and remember people’s faces with names. Path Guidance: Determine if a path is clear or obstructed. Text Reading: Read text aloud to the user. Ask Question: Provide real-time answers to user queries. Hardware Components Camera LiDAR or Ultrasonic Sensors Microphone and Speakers Processing Unit Software Components OpenCV YOLOv10 PyAudio Python JavaScript TensorFlow Use Cases
- Navigating a Crowded Environment User: Sarah, a visually impaired individual
Scenario: Sarah needs to navigate through a crowded shopping mall to reach a specific store.
Steps:
Initialization: Sarah puts on her VisionAid-equipped smart glasses. Object Detection: The device starts detecting and classifying objects in real-time, identifying people, obstacles, and store signs. Path Guidance: The path guidance system uses depth cameras to analyze the path ahead, providing audio feedback like "Clear path ahead" or "Obstacle detected, move slightly to the left." Face Recognition: Sarah meets a friend in the mall, and the face recognition system identifies the friend, providing an audio prompt "John is approaching." Text Reading: Sarah uses the text reading feature to capture and read aloud promotional signs at the store. Outcome: Sarah successfully navigates the mall, meets her friend, and reads important information, enhancing her independence and confidence.
- Ensuring Safe Pathways User: David, a visually impaired individual
Scenario: David needs to walk to a nearby park and ensure the path is clear and safe.
Steps:
Initialization: David activates the path guidance feature on his VisionAid device. Path Analysis: The device uses depth cameras to scan the path ahead. Obstacle Detection: It detects any obstacles such as uneven surfaces, steps, or objects in the way. Audio Guidance: The device provides real-time audio feedback such as "Clear path ahead" or "Caution, step up in 5 feet." Navigation: David follows the audio instructions to navigate safely to the park. Outcome: David navigates safely to the park, avoiding obstacles and ensuring a clear pathway.
Recognized face Identification:
Unrecognized face Identification:
Object Classification:
Raspiberry (for computer vision processing): Screenshot 2024-07-20 125134](https://github.com/user-attachments/assets/1d2a72a8-7707-40c0-b913-63a88291053a)
Prototype of our project VISIONAID:
For further inquiries or collaboration, please contact:
Sonu Kumar Pandit Email: [email protected] Institute: Parul Institute of Engineering & Technology (PIET)