结合NSST和快速非局部均值滤波的刀具图像去噪

为消除基于图像处理的刀具磨损检测中的图像噪声，提出了结合非下采样Shearlet变换(Non-subsampled Shearlet Transform, NSST)和快速非局部均值(Fast Non-local Means, FNLM)滤波的图像去噪方法。首先，利用基于决策的非对称剪切中值(Decision Based Un-symmetric Trimmed Median, DBUTM)方法滤除图像中的椒盐噪声；然后，对图像进行NSST多尺度分解，得到一个低频子带和一系列高频子带；最后，分别使用FNLM滤波和各向异性扩散模型调整低频和高频子带系数，并由调整后的各子带系数重构出噪声滤除后的图像。实验结果表明，与基于小波的阈值收缩方法、基于Contourlet的全变差模型结合各向异性扩散方法、基于NSST和标准非局部均值滤波方法相比，本文方法在主观视觉去噪效果、峰值信噪比、结构相似度以及处理速度等4个方面性能更优。 

Cutting tool image denoising based on NSST and fast non-local means filter

To avoid the influence of the image noise in cutting tool wear condition monitoring system based on image processing, a cutting tool image denoising method based on non-subsampled shearlet transform (NSST) and fast non-local means(FNLM) filter is proposed. Firstly, the decision based unsymmetrical trimmed median(DBUTM) filter is applied to eliminate the pepper and salt noise in the original image. Then the image is decomposed by NSST into a low-frequency component and a series of high-frequency components. Finally, the FNLM filter and the anisotropic diffusion model are introduced to process the low-frequency component and high-frequency components, respectively, after which the denoised image is reconstructed with those modified coefficients of frequency subbands. The experimental results demonstrate that, compared with wavelet based threshold shrink method, contourlet based method combining total invariance model with anisotropic diffusion, shearlet based standard non-local means filtering method, the proposed method has a better performance in four aspects such as subjective visual denoising effect, peak signal to noise ratio, structural similarity and running speed.