基于光度立体和图像显著性的皮革缺陷检测  

Leather Defect Detection Based on Photometric Stereo and Saliency Object Detection

作　　者：刘根[1] 蔡念[1,2] 肖盼 林健发 LIU Gen;CAI Nian;XIAO Pan;LIN Jianfa(School of Information Engineering, Guangdong University of Technology, Guangzhou 510006, China;Foshan Deeple Vision Technology Co., Ltd, Foshan, Guangdong 528200, China)

机构地区：[1]广东工业大学信息工程学院,广州510006 [2]佛山缔乐视觉科技有限公司,广东佛山528200

出　　处：《计算机工程与应用》2019年第8期215-219,共5页Computer Engineering and Applications

基　　金：国家自然科学基金(No.91648108);广东省科技计划项目(No.2015B010124001;No.2015B010104006;No.2015A030312008);广州市产学研协同创新重大专项项目(No.201508010001;No.201604016022)

摘　　要：为了提高皮革缺陷检测效率,提出一种基于光度立体视觉和图像显著性的皮革缺陷检测算法。搭建光度立体视觉平台,完成不同角度的皮革样本采集,利用光度立体视觉技术计算皮革样本的合成图和表面法向量图;对表面法向量图进行曲率滤波操作,用近似表面粗糙度特征自适应选择合成图或滤波图;利用显著性目标检测算法完成皮革缺陷检测与定位。实验结果表明,与现有皮革缺陷检测方法相比,该算法能很好地检测不同材质皮革的多种缺陷,且准确率高,速度快。To improve the efficiency of leather defect detection, a leather defect detection algorithm is proposed by combining photometric stereo with saliency object detection. First, a acquisition platform for photometric stereo is established to collect leather samples in different orientation. Then, photometric stereo is applied to obtain a synthetic image and a surface normal vector image. Next, the surface normal vector image is processed by curvature filtering and the synthetic image or the filtered image is selected by the approximate surface roughness features. Finally, defect detection is performed based on saliency object detection. Experimental results indicate that, compared to the existing methods, the proposed algorithm can detect several kinds of defects on different leathers made of different materials at a high accuracy and a high speed.

关 键 词：皮革缺陷检测 光度立体视觉 曲率滤波 显著性目标检测 图像分割 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]