Improvement Of Wavelet Denoising Algorithm And Its Application In Noise Reduction Of Wet Steam CCD Image

In the process of acquisition,transmission and conversion of CCD images,the CCD images often contain noise due to reasons such as the mechanical movement of the equipment,the internal circuit of the system and the materials of the equipment,which will reduce the image quality,increase the experimental error,and subsequently affect further processing of the images.The wavelet threshold denoising method is a main technology for image denoising nowadays.However,traditional soft and hard threshold functions and traditional threshold estimation rules are unable to meet the need of image denoising due to their own limitations.For instance,the traditional hard threshold function is discontinuous at the threshold,which may bring forward a Pseudo-Gibbs phenomenon after image reconstruction;the traditional soft threshold function always has a constant deviation between the estimated wavelet coefficient and the real wavelet coefficient,which will lead to a low accuracy after image reconstruction.The traditional fixed threshold estimation rule selects a same threshold for each decomposition layer,which causes a overly strict suppression of the signal.In response to the above limitations,the work done in this paper is as follows:(1)Based on the traditional threshold function,an improved threshold function is proposed.The new threshold function combines the advantages of traditional soft and hard thresholds.It not only has continuity at the threshold and higher order conductivity in the threshold region,but also has estimated wavelet coefficients for approximating the actual wavelet coefficients.(2)Based on the traditional fixed threshold estimation rules,an adaptive threshold estimation rule is proposed.The new threshold criterion is different from the unified threshold and has adaptive characteristics in the selection of threshold.The threshold value of each decomposition layer can be adaptively changed according to the size of the wavelet coefficient,which is more consistent with the actual denoising law.(3)Using the existing experimental instruments,a test bench according to the backscattering experimental measurement model is built up,and then different experimental conditions parameters for experimental measurement are set.CCD experimental images of different scattered light intensity distribution are collected and finally used for the wavelet proposed in this paper.The adaptive threshold denoising algorithm is put in use for the denoising of the collected CCD experimental images.The experimental results show that compared with the traditional denoising algorithm,the proposed wavelet adaptive threshold denoising algorithm not only has better subjective visuals,but also has good effect on the objective evaluation index coefficient.