Hand Gesture Recognition System

Hand Gesture Recognition System Documentation:

1) For comprehensive algorithm understanding, analyze the following flowchart images illustrating the processing pipeline:

- AlgorithmOverall.jpg (Main system architecture diagram)

- MKRoDAlgorithm.jpg (Detailed feature extraction methodology)

2) The system prototype utilizes MATLAB image processing toolbox, initially developed in MATLAB 7.0 and optimized in MATLAB 7.10.0. Key functions include imread() for image input, rgb2gray() for conversion, and regionprops() for hand feature analysis.

3) Critical preprocessing requirement: Input images must contain pure black backgrounds (RGB [0,0,0]) with clearly visible palm and finger contours. This contrast enhancement ensures reliable segmentation for ASL gesture classification.

4) Implementation uses a curated ASL dataset containing seven distinct gestures: "b, c, h, i, o, l, y". The recognition algorithm employs template matching with normalized correlation coefficients for pattern comparison.

----------------------------------------------------------------

Execution Instructions: Run the main script 'execHGR.m' within the 'Project' directory. This master file coordinates all submodules including image preprocessing, feature extraction, and classification stages.

The system provides a foundation for vision-based gesture recognition with expandable architecture for additional ASL gestures.