Author : Yogitha S.K and Savitri Raju

The objective of this paper is to find the effects of quantization matrices (10%, 50%, 90%) on Discrete cosine transform and Fuzzy logic (without quantization) using different resolution (256X256, 512X512) of an JPEG image. After comparison of DCT and fuzzy logic separately, the effect of quantization matrices is tested by linking DCT and fuzzy logic using different resolution of an JPEG image. In recent years, many researchers have applied the fuzzy logic to develop new techniques for contrast improvement. Fuzzy logic is a well known rather simple approach with good visual results, but proposed fuzzy operation algorithm is default nonlinear. Here proposed algorithm is a default nonlinear thus not straight forward applicable on the JPEG bit stream, it is possible when the right combination is found. The processing is much faster, due to the reduced number of co-efficient, because the majority of the coefficient in the DCT domain is zero after quantization. Discrete cosine transform is a technique which converts signal to elementary frequency components and is mainly used for image compression. Image compression has become important as storage or transmission of images requires large amount of bandwidth. Performance measurement is carried out in terms of Mean Squared Error (MSE) and Peak Signal-to-Noise Ratio (PSNR). The proposed work is designed using MATLAB 7.10.

Keywords: DCT, 2D DCT, Fuzzy logic, Fuzzy Intensification Operator, Image Compression, Quantization, MSE, PSNR.

Article published in the Proceedings of National Conference on ‘Women in Science & Engineering’ (NCWSE 2013), SDMCET Dharwad