For detecting various surface defects of strip steel with non-uniform illumination and noise,a defect detection algorithm based on mathematical morphology enhancement and image fusion method was proposed.Firstly,the multi-structural morphology quotient image enhancement(MMQIE) and multi-structural morphology edge enhancement(MMEE) were adopted to process defect image of the strip steel.Secondly,weighted fusion method was used to fuse the enhanced images,and the weight coefficients were determined by the image background entropy and the pixel average ratio of the two images processed by MMQIE and MMEE.Finally,binary processing was carried out for fused images.The experiment results show that the proposed algorithm can accurately extract the surface defects of strip steel with uneven illumination and strong noise,and can achieve good effects for other types of strip steel defect images.In addition,the algorithm has strong anti-noise capability and high stability.
参考文献
相似文献
引证文献
引用本文
王凡,彭国华,谢昊伶.基于形态学增强和图像融合的板带钢缺陷检测[J].激光与红外,2018,48(1):124~128 WANG Fan, PENG Guo-hua, XIE Hao-ling. Strip steel defect detection based on morphological enhancement and image fusion[J]. LASER & INFRARED,2018,48(1):124~128