Noise Reduction in Compressed Images Using Improved Fuzzy Transform Technique

Abstract

In the area of digital image compression, computer algorithms are used to perform processing of images and compression. It deals with developing a digital system that perform operations on digital image. It has many advantages using in digital camera, film, Satellite, X-ray and many more applications. Image compression is a technique used to save the storage space normally used to compress images and videos. Number of compression algorithms are used like run length encoding, huffman coding, discrete cosine transform, vector quantization, fuzzy transform. This gives a brief idea on improved fuzzy technique to reduce noise in compressing image. There are so many techniques for compression but in this only present the techniques of improved fuzzy method to reduce noise and compressed the image by using edge detection. The main idea behind applying this is we have to preserve the well significant edges as Jpeg is the popular standard but at low bit rate Jpeg exhibits blocking artifacts means noisy effects that affect the visual image quality so to produce high visual quality image at low bit rate ,the algorithm is efficient and simple. The proposed algorithm consists of three steps. First, image is preprocessed using competitive fuzzy edge detection. Second, based on edge information image is compressed and decompressed using improved fuzzy transform. Third, reconstructed image is post processed using hybrid median filter for artifact reduction. Analysis proves the superiority of proposed algorithm. The results of different number of coefficients are compared with the value of PSNR, MSE of algorithm. After comparison of techniques it is found to be efficient for visualization.