X-ray of Tire Defects Detection via Modified Faster R-CNN
Authors:
- 1 November 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 257-260
Abstract
With the rapid development of deep learning model in computer vision area, it has outperformed most of traditional machine learning algorithms. Since tire factories pay much attention to defects detection of tires based on x-ray image, lots of tire x-ray image based defects detection methods are brought up. However, there are still challenges in detection accuracy. This paper put forward a novel deep learning model and modified Faster R-CNN to conduct x-ray defects detection. Some proper processing is done on x-ray image before extracting the features and detecting the defects and then adjusting the feature extractor, proposal generator and box classifier of Faster R-CNN respectively. Comprehensive experiments are carried out to testify that our proposed model is capable of achieving higher detection accuracy compared with other methods.Keywords
This publication has 12 references indexed in Scilit:
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Deep Residual Learning for Image Recognition2016
- Fully Convolutional Networks for Semantic SegmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Fast R-CNN2015
- The Pascal Visual Object Classes Challenge: A RetrospectiveInternational Journal of Computer Vision, 2014
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition2014
- Object Detection with Discriminatively Trained Part-Based ModelsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2009
- Speeded-Up Robust Features (SURF)Computer Vision and Image Understanding, 2008
- Object recognition from local scale-invariant features1999
- The CNN paradigmIEEE Transactions on Circuits and Systems I: Regular Papers, 1993