Colour Image Compression Using Block Truncation Coding and Genetic Algorithm

In this paper a colour image data compression method using block truncation coding (BTC) and genetic Algorithm (GA) is presented. This compression technique reduces the computational complexity and provides acceptable visual quality reconstructed images. BTC acts as the basic compression technique but it exhibits two disadvantages such as the false contour and blocking effect. Hence error diffused block truncation coding (EDBTC) is used to overcome the two issues. The proposed system is made applicable for colour images. Colour images comprise three planes: red, green, and blue. There are very high correlations between the images in these planes. This motivates the use of one common bit plane to represent all three colour bit planes in the proposed compression method for colour images. GA is used to make the common bit plane optimal. Experimental results show that the proposed method provides acceptable visual quality reconstructed images with high peak signal to noise ratio (PSNR) when compared to the conventional BTC and EDBTC methods.