一种基于小波阈值去噪的改进方法在气液两相流图像中的应用

摘要

利用高速摄影机拍摄了气液两相流的图像,它们往往会伴随一些噪声,因此,去噪处理成为图像预处理的一个重要环节。通过对小波变换及阈值去噪的分析,在Donoho等人提出的传统阈值去噪方法基础上,提出了一种新的阈值函数去噪方法。该方法与传统的软硬阈值函数相比,能够有效保持图像边缘信息,并能避免硬阈值函数的不连续性。实验结果表明,新的方法优于传统的阈值去噪方法,而且信噪比较高。%Used high2speed video camera filming the flow image of gas-liquid two-phase flow,they always have some noise,so denoising processing become the important part of the image preprocessing.Based on wavelet transform and threshold denoising analysis,A new threshold function denoising method is presented based on the traditional denoising method put forward by Donoho in this paper.Compared with traditional soft threshold and hard threshold,the new algorithm not only has the capability of protecting edge information of image signal,but also avoids the discontinuity of hard threshold function.Experin results indicate that the new method over the traditional threshold denoising method,and has higher SNR.