Research On Static Image Compression Algorithm Based On JPEG

With the development of human society and technology, people's demanding on processing and exchanging information are increasing. The future information is multimedia information based on image. As the image has a huge amount of data, it is difficult to store and transport, which requires image transmission the smaller and less space for storing image bit-rate in the premise of quality assurance , it requires the application of image compression coding technology.The paper briefly introduced the history of standardized JPEG first, and the basic principles,classification and evaluation standards of JPEG algorithm in detail.Focuses on process and method of compression of JPEG, describes theory and mathematical derivation the DCT transform coding, quantization, entropy coding and Huffman coding module, as well as the functional analysis of each module. Simulating JPEG algorithm with Matlab, and comparative analysis the results from different quantify levels.The success of JPEG is obvious to all, the majority picture format is JPEG in the Internet of the current. However, JPEG also has shortcomings, it will have a serious effect box during encoding in the low-frequency and the compression speed is slower. According that problem, this paper uses look-up table method to reduce multiplication in the DCT to improve compression speed. The experimental results show that the look-up table method can improve the rate of image compression about 48%.In addition, quantify interval of high-frequency is high and low-frequency is small which caused by using quantitative methods of standard JPEG algorithm, that will result different image can't achieve the best results, bringing adaptive quantization method instead of the standard quantitative method that can improve compression quality in the same compression ratio without affecting any JPEG functions.