This project demonstrates real-time hand gesture recognition using OpenCV. The model detects hand gestures and identifies specific gestures like "ONE", "TWO", "THREE", etc., based on convexity defects in the hand contour.
✨ Use Case: This system can empower specially-abled children to study online without needing to attend special schools, by enabling interaction through hand gestures.
- Real-time hand gesture recognition using a webcam.
- Detects and identifies gestures like ONE, TWO, THREE, and more.
- Visualizes contours, convex hulls, and defects for understanding gesture anatomy.
- Accessible and adaptable for educational or interactive applications.
-
Hand Detection:
- A sub-window (ROI) isolates the hand for detection.
- Converts the image to grayscale and applies Gaussian blur.
-
Thresholding:
- Thresholding is used to separate the hand region from the background.
-
Contour and Convex Hull:
- Contours are extracted from the thresholded image.
- Convex hull and convexity defects are computed to analyze the hand's shape.
-
Gesture Recognition:
- Defects are analyzed using the cosine rule to determine the angle between fingers.
- Gestures are recognized based on the number of defects detected.
-
Real-Time Feedback:
- Displays the detected gesture on the screen in real-time.