一种新型小波阈值函数下的毫米波图像去噪  

Research on Millimeter Wave Image Denoising Based on a New Wavelet Threshold Function

作　　者：吴超[1] 张胜[1] 陈建飞 WU Chao;ZHANG Sheng;CHEN Jian-fei(College of Electronic and Optical Engineering,Nanjing University of Posts and Telecommunications,Nanjing Jiangsu 210046,China)

机构地区：[1]南京邮电大学电子与光学工程学院,江苏南京210046

出　　处：《计算机仿真》2022年第9期239-242,313,共5页Computer Simulation

基　　金：国家自然科学基金项目(61601237)。

摘　　要：小波变换通过伸缩和平移等运算将图像进行了多尺度分析,可以有效地提取信息,因此可以将其运用到毫米波图像去噪领域当中。然而,小波阈值去噪方法中硬阈值函数和软阈值函数都有其自身的局限性。硬阈值函数的间断性会导致图像模糊,软阈值函数存在固定差值会影响重构精度。提出一种新的新阈值函数,通过加入了调节因子而更加具有灵活性,可以克服两种方法的缺点。仿真结果表明,新阈值函数下毫米波图像的峰值信噪比比硬、软阈值函数分别提升0.67db、1.06db,且新阈值函数下毫米波图像的主观视觉感官更好、去噪效果更佳。Wavelet transform can extract information effectively by multi-scale analysis of image through scaling and translation, so it can be used in the field of millimeter wave image denoising. However, both the hard thresholding function and the soft thresholding function have their own limitations. The discontinuity of hard thresholding function will lead to image blur, and the fixed difference of soft thresholding function will affect the reconstruction accuracy. Therefore, a new threshold function is proposed in this paper. The new threshold function is more flexible by adding the adjustment factor, which can overcome the shortcomings of the two methods. The simulation results show that the peak signal-to-noise ratio of millimeter wave image under the new threshold function is 0.67 db and 1.06 db higher than than of the hard and soft threchold functions respectively, and the subjective visual sense of millimeter wave image under the new threshold function is better, and the denoising effect is better.

关 键 词：小波变换 阈值函数 毫米波图像 图像去噪 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]