The Application Of Wavelet Transform In Image Processing

In image signal after wavelet transformation, the state of image information and noise information is different. Processed in the corresponding rules, most noise can be removed. In addition, the wavelet analysis theory into the whole process of image processing can also enhance the processing performance of other function modules, and can optimize the whole image signal processingThreshold denoising is chosen to be discussed in this paper. We analyze the selection of parameters in detail, including wavelet function selection, decomposition level selection, and the selection of threshold function. An in-depth analysis of the threshold selection puts forward an improved method called "point threshold", combined with "partial block threshold". It makes full use of the information of high frequency coefficients after image decomposition. Combined with the adaptive method, each coefficient has its own threshold value. These threshold values have strong pertinence. At the same time, in order to make full use of the amount of transform calculation, we do white balance on decomposition image and do color interpolation on Bayer image combined with the idea of wavelet transform. Finally, through simulation experiments, compared with other commonly used noise reduction, white balance and color interpolation, to demonstrate the effectiveness and feasibility of the application of these methods.Simulation results show that, for the image subject to the same level of noise pollution, image after "point threshold" noise reduction has higher peak signal to noise ratio than other noise reduction method; for the same images subject to different degrees of noise pollution, "point threshold" noise reduction is also better than others. In the white balance, wavelet transform reduces the amount of data to be processed. Compared with the commonly used method, the stability is improved, and the effect is more real. In the color interpolation, wavelet transform provides a new processing method, and it is proved by experiments that the image obtained from using this method has high resolution ratio. In addition, the wavelet transform is used in the white balance and color interpolation, largely mean the computation volume of wavelet decomposition and reconstruction process in the whole image processing, and can change or simplifying the existing image processing process, and provide new ideas for the future exploration and research.