基于NSST和CLAHE的零件表面缺陷图像增强算法研究

Research on image enhancement algorithm of parts surface defects based on NSST and CLAHE

为了改善传统剪切波变换在零件表面缺陷图像中边缘不明显和去除噪声效果不理想的问题,本文提出了基于非下采样剪切波变换(NSST)和限制对比度的自适应直方图均衡化(CLAHE)的图像增强算法,在后续的Canny算子边缘检测中效果较好。首先,将缺陷图像进行NSST变换,获得相应的高频图像和低频图像;其次,将高频图像进行CLAHE变换,NSST逆变换的图像经过Canny算子应用于检测缺陷图像边缘。结果表明:该算法在面对零件的拉伤、倒偏角和碰伤缺陷时,均获得了较高的峰值信噪比和信息熵;在增强边缘的同时,能够更好的去除噪声,证明了其有效性和鲁棒性。

In order to improve the non-obvious edge and unsatisfactory noise removal effect of traditional shear-wave transform in parts surface defect image,an image enhancement algorithm based on non-subsampled shear-wave transform(NSST)and adaptive histogram equalization(CLAHE)with limited contrast is proposed in this paper,which has good effect in subsequent edge detection.Firstly,the defect image is transformed by NSST to obtain the corresponding high frequency image and low frequency image.Secondly,the high frequency image is processed by CLAHE transformation,and the image with NSST inverse transformation is applied to detect the defect image edge through Canny operator.The results show that the proposed algorithm achieves high peak signal-to-noise ratio and information entropy in the face of strain,deflection Angle and collision defects of parts.The proposed algorithm can enhance edge and remove noise better,which proves the effectiveness and robustness of the proposed algorithm.