Optimization of Image Compression Method Based on Fuzzy Relational Equations by Overlap Level of Fuzzy Sets

Abstract

A design method of coders on YUV color space is proposed based on an overlap level of fuzzy sets, in order to optimize the image compression/reconstruction method based on fuzzy relational equations. In the YUV color representation of the original image, the Y plane contains more information than the U and V planes, in terms of human perception. Therefore, coders with different sizes (Y plane coders bigger than U and V planes) lead to more effective compression/reconstruction, where the appropriate coders for YUV planes can be constructed based on an overlap level of fuzzy sets. Through image compression/ reconstruction experiments using 100 typical images (extracted from Corel Gallery, Arizona Directory), it is confirmed that the peak signal to noise ratio of the proposed method increases at a rate of 7.1% ~ 13.2%,compared to the conventional method, when compression rates range from 0.0234 ~ 0.0938.