The project is based on idea of detection of indian sign language. This is a computer vision based project which detects what the hand is showing using certain algorithm as described below.
- opencv - pip install opencv-python
- Seperate out the skin color from the input frame (captured through attached camera) using HSV or other color format with upper and lower bound for skin color as per range.
- Removal of noise from image via dialation and errosion, and applying filters to smooothen the image.
- Finding the contour of the mask(seperated skin color mask) and drawing convex-hull.
- Using the various properties for seperation various Hand_Gestures(In this project we will use Indian-sign-language as Hand_Gesture- 1,2,3..) like-
- Finding the extreme most points of contours and computing angle between bottommost and other extreme points and using the values in if loop.
- Finding the corner points of hull.
- Finding number of lines in contour of specific gesture.
- Area of contour for different gestures.
1) git clone https://github.com/ash-S26/Indian-Sign-language.git
2) python HAND_GESTURE_DETECTION.py
