A new method in wheel hub surface defect detection: Object detection algorithm based on deep learning

In wheel hub surface defect detection, a unified image background is required. However, it is a challenging task because of the various categories of wheel hubs, and the complicated image background of the defect areas caused by the collection of the images with the defect areas in a narrow field of vision. Compared to the traditional method, the deep learning algorithm is more robust, which doesn't need the unified image background. We use Faster-RCNN with ResNet-101 as the object detection algorithm. And our related experiments show that our deep learning method is able to detect the scratches and points on the wheel hub in an image with a complicated background, as shown in Figure5. Furthermore, the model can detect defects on any part of the wheel hub of various types, and obtain the position and the class of the defective area. Particularly, the method achieves 86.3% mAP on our own data set.

[1]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[3]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[4]  Trevor Darrell,et al.  Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Jian Sun,et al.  Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Sergio Guadarrama,et al.  Speed/Accuracy Trade-Offs for Modern Convolutional Object Detectors , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[9]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).