adu3010 / Color-Pattern-Recognition-Software

Automated MATLAB software for recognizing color patterns in images, inspired by Lego's Life of George game, with a success rate of up to 91%.

Color Pattern Recognition in MATLAB

Overview

This MATLAB project automates the recognition of color patterns in images, inspired by Lego's Life of George game. The software reads a supplied image, detects color patterns using image processing techniques, and outputs an array of color names. The implementation achieves up to 91% accuracy, with functions for image loading, circle detection, distortion correction, and color extraction.

Getting Started

1. Ensure you have MATLAB installed on your system.
2. Clone the repository to your local machine.

   git clone https://github.com/your-username/color-pattern-recognition.git

3. Open MATLAB and navigate to the project directory.

Usage

1. Run the findColours function with the filename of the image you want to process.

   colorArray = findColours('your_image.png');
   disp(colorArray);

2. View the processed image with corrected distortion by the correctImage function.

Functions

• loadImage(filename): Loads an image and converts it to a double-precision array.
• findCircles(image): Detects the coordinates of the four largest black circles in the image.
• correctImage(coordinates, image): Corrects image distortion using specified coordinates.
• getColours(image): Extracts color patterns from the image and returns a 4x4 matrix of color names.

Results

Check the provided table for success rates on sample images, including accuracy percentages and notes on challenges.

References

Explore the code's foundation based on techniques like connected component labeling, image thresholding, and perspective transformation. See references in the provided appendix.

Contributors

Poovarasan Rajendiran