Improving Performance Of EZW Algorithm Based On Image Denoising Preprocessing

Image on the input and the acquisition will be polluted by noise. Therefore, before the image compression, image denoising is more effective to improve the image compression quality. Based on the adaptive threshold denoising, the paper improved EZW algorithm in order to obtain higher coding gain.The paper firstly improve adaptive threshold denoising method. Based on in-depth analysising advantages and disadvantages of hard and soft threshold function, a kind of adjustable parameters adaptive threshold function is given by taking Gaussian noise as background. The experimental results show that denoising effect of the function is better than soft threshold function and hard threshold function. But, the edge of the denoised image appears Gibbs visual distortion. In order to inhibit this phenomenon, the paper gives the adaptive threshold denoising algorithm based on the dyadic wavelet transform. The algorithm refers to the scale adaptive threshold to overcome defects of loss of image detail caused by the Donoho universal threshold haven’t the scale adaptive. The experimental results show that compared with wavelet soft threshold or hard threshold, the image handled by the algorithm both in peak signal to noise ratio or the visual effects are significantly enhance and improve is an effective method for image denoising.The paper also improved EZW algorithm. Auxiliary scan of the original EZW algorithm need to build complex quantitative, the study find that only need to transform the coefficient, a common quantizer can be used to handle the data output of the auxiliary scanning of each layer,auxiliary scan output encoding can be got.Therefore, the paper gives the significant bit approximation quantization EZW auxiliary scanning algorithm. In addition, for the problem of the main scan encoding process of the original EZW still have encoding waste of resources, the paper gives the additional encoding symbols adaptive EZW scanning algorithm. The algorithm combine additional encoding symbols EZW main scanning algorithms with adaptive EZW main scanning algorithm. The main idea of the algorithm is that before each scan code, do statistics to the number of bits of the main scan program, select the least number of bits program as the main scanning program. Furthermore, by additional coding symbols method, number of bits are reduced, to achieve the purpose of improving the coding efficiency. The experimental results show that the new improving algorithm gived by the paper not only reduces the number of bits, but also effectively improve the compression efficiency and reconstruction quality of the image.