This project enables users to access frequently used websites through hand gestures using a webcam. The application recognizes hand gestures for 2, 3, and 4 fingers, opening specified websites in new tabs of the default web browser. Additionally, a gesture for 5 fingers will close all tabs of the web browser.
Features 2 Finger Gesture: Opens a user-specified website. 3 Finger Gesture: Opens a different user-specified website. 4 Finger Gesture: Opens another user-specified website. 5 Finger Gesture: Closes all open tabs of the web browser. Real-time Gesture Recognition: Uses OpenCV for real-time hand gesture recognition. Installation To run this project, ensure you have the following installed:
Python 3.x OpenCV NumPy You can install the necessary packages using pip: pip install opencv-python numpy
Usage Clone the repository: git clone https://github.com/vibhusharan2407/GestureRecognition cd GestureRecognition
Run the Python script: python gesture_recognition.py
Follow the prompts to enter the URLs for the 2-finger, 3-finger, and 4-finger gestures. Show your hand within the rectangle area displayed on the webcam feed to perform the gestures.
How It Works Capture Video: The webcam captures video frames. Hand Detection: A sub-region of the frame is processed to detect the hand. Preprocessing: The sub-region is converted to grayscale and blurred for better contour detection. Thresholding: Otsu's binarization is applied to create a binary image. Contour Detection: Contours are detected in the binary image. Convex Hull and Defects: The convex hull of the hand contour is found, and convexity defects are used to count the number of fingers. Gesture Recognition: Based on the number of fingers detected, the corresponding action is taken (opening a URL or closing the browser).
Contributing Contributions are welcome! Please feel free to submit a Pull Request or open an Issue.
License This project is licensed under the MIT License. See the LICENSE file for details.
Acknowledgements OpenCV for providing an extensive library for computer vision tasks. NumPy for numerical computations.