Weak Supervised Surface Defect Detection Method Based on Selective Search and CAM

Due to the large scale variation of surface defects of different types of strip steel, there are limitations in using threshold segmentation to locate objects, we propose a surface defect detection algorithm combining selective search and class activation mapping (CAM) to improve objects localization. First, we use selective search to generate defect bounding box in the image, and predicts the classification and CAM of the defect in the image through the trained model. Then, in the defect detection, filter the bounding box with the classification information of the defect as priori knowledge. We only retain the bounding box that approximate the shape of the defect and map the filtered defect bounding box to the CAM of the corresponding defect. Finally, select the bounding box with the highest score as a detection result. Experiment results show that the proposed method can achieve a mean average precision of 91.1% on our dataset. And it can more accurately locate defects in the image. Compared with traditional CAM, our method has more excellent detection performance in surface defect detection applications of strip steel.

[1]  Wenyu Liu,et al.  Multiple Instance Detection Network with Online Instance Classifier Refinement , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Koen E. A. van de Sande,et al.  Selective Search for Object Recognition , 2013, International Journal of Computer Vision.

[3]  Yonghong Song,et al.  Feature Fusion for Weakly Supervised Object Localization , 2018, 2018 Chinese Automation Congress (CAC).

[4]  Yong Jae Lee,et al.  Hide-and-Seek: Forcing a Network to be Meticulous for Weakly-Supervised Object and Action Localization , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[5]  Li Liu,et al.  Defect Detection of Rail Surface with Deep Convolutional Neural Networks , 2018, 2018 13th World Congress on Intelligent Control and Automation (WCICA).

[6]  Yi Zhu,et al.  Soft Proposal Networks for Weakly Supervised Object Localization , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[7]  Roberto Cipolla,et al.  SegNet: A Deep Convolutional Encoder-Decoder Architecture for Robust Semantic Pixel-Wise Labelling , 2015, CVPR 2015.

[8]  Koen E. A. van de Sande,et al.  Segmentation as selective search for object recognition , 2011, 2011 International Conference on Computer Vision.

[9]  B. S. Manjunath,et al.  Weakly Supervised Localization Using Deep Feature Maps , 2016, ECCV.

[10]  Qingbo Wu,et al.  Class Activation Map Generation by Representative Class Selection and Multi-Layer Feature Fusion , 2019, ArXiv.

[11]  Bolei Zhou,et al.  Learning Deep Features for Discriminative Localization , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[13]  Bart De Schutter,et al.  Deep convolutional neural networks for detection of rail surface defects , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[14]  K. F. C. Yiu,et al.  Fabric defect detection using morphological filters , 2009, Image Vis. Comput..

[15]  Du-Ming Tsai,et al.  Wavelet-based defect detection in solar wafer images with inhomogeneous texture , 2012, Pattern Recognit..

[16]  Ye Feng.  Unsupervised defect detection for gold surface of flexible printed board , 2010 .

[17]  Chang Liu,et al.  C-MIL: Continuation Multiple Instance Learning for Weakly Supervised Object Detection , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Dong-Joong Kang,et al.  Localization of Welding Defects using a Weakly Supervised Neural Network , 2018, 2018 18th International Conference on Control, Automation and Systems (ICCAS).

[19]  Wenyu Liu,et al.  PCL: Proposal Cluster Learning for Weakly Supervised Object Detection , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.