A Method for Extracting Planar Image Features based on Convolution Neural Network

Shengjun Yang; Zhonglan Wu; Xiaohong Li; Yun Hong

doi:10.62051/s439k645

Authors

Shengjun Yang
Zhonglan Wu
Xiaohong Li
Yun Hong

DOI:

https://doi.org/10.62051/s439k645

Keywords:

Process Measurement and Control; Minor Defects; Target Detection; SSD.

Abstract

This paper is supported by the 2022YFG0070 science and technology plan of Sichuan Province. Image feature extraction technology plays a very important role in the measurement and control process of industrial product processing, but for planar micro defects in the background, the current general single shot multibox detector (SSD) has some disadvantages, such as easy loss of feature information, low detection accuracy, and insufficient number of detection feature maps. In view of the above problems, combined with the characteristics and requirements of image feature extraction in the process of measurement and control processing, this paper proposes and designs BSSD algorithm. The algorithm uses ResNet34 to extract more micro defect information to solve the problem of feature extraction; Seven multi-scale feature maps were selected to increase the number of feature maps for detecting micro defects; A backtracking layer is set up to fuse the abstract information of the high-level network into the shallow network before the multi-scale feature map is input into the classification network to enhance the expression ability of the abstract features. The experimental data show that the accuracy is comparable to DSSD, and the speed is similar to FSSD, which shows a significant advantage in the accuracy of small target detection.

Downloads

Download data is not yet available.

References

Howard A, Menglong ZhuBo Chen, D. Kalenichenko, W. Wang, Tobias Weyand, M. Andreetto and H. Adam. Efficient Convolutional Neural Networks for Mobile Vision Applications[J]. ArXiv abs/ 1704. 04861, 2017: n. pag.

Deng C, Wang M, Liu L, et al. Extended Feature Pyramid Network for Small Object Detection[J]. 2020.

Shelhamer, Evan, J. Long and Trevor Darrell. Fully Convolutional Networks for Semantic Segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017: 640-651.

Ren, Shaoqing, Kaiming He, Ross B. Girshick and J. Sun. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence 2015, (39): 1137-1149.

Tao, Xian & Zhang, Dapeng & Ma, Wenzhi & Liu, Xilong & Xu, De. Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks[J]. Applied Sciences. 2018, 8(9):1575.

Mei S, Wang Y, Wen G. Automatic Fabric Defect Detection with a Multi-Scale Convolutional Denoising Autoencoder Network Model. Sensors (Basel), 2018, 18(4):1064.

Zhao Z, Bo L, Rong D, et al. A Surface Defect Detection Method Based on Positive Samples[C]// Pacific Rim International Conference on Artificial Intelligence. Springer, Cham, 2018.

Deng C, Wang ML iu L, et al. Extended Feature Pyramid Network for Small Object Detection[J]. 2020.

Han M, Qingxiang Wu, Xiongjun Zeng. Gear visual micro defect detection based on deep learning [J]. computer system applications,2020,29(03).

Yongxiong Yu, Liuqing Du, Cuilan Zeng, et al. Automatic location method of casting micro defects based on deep learning feature matching [J]. Journal of instrumentation, 2016, 37(06).

Bingyin Ren, Zhiyong Li, Dai Yong. Detection method of slight scratch on mobile phone screen [j]. Journal of Harbin Institute of technology, 2021, 53 (01).

Xu X, Zheng H, Guo Z, et al. SDD-CNN: Small Data-Driven Convolution Neural Networks for Subtle Roller Defect Inspection[J]. Applied Sciences, 2019, 9(7).

Girshick, Ross B. Fast R-CNN[J]. IEEE International Conference on Computer Vision (ICCV),2015: 1440-1448.

Ren S, He K, Girshick R B, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. neural information processing systems, 2015, 22(5): 91~99.

Girshick R. B, Donahue J, Darrell T, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: IEEE Computer Society, 2014, 580-587.

He K, Zhang X, Ren S, et al. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition. European Conference on Computer Vision. Zurich, Switzerland: ECCV, 2014. 346-361.

Liu W, Anguelov D, Erhan D, et al. SSD: Single Shot MultiBox Detector[C]. European Conference on Computer Vision. Zurich, Switzerland: ECCV, 2016, 21-37.

Fu C, Liu W, Ranga A, et al. DSSD: Deconvolutional Single Shot Detector. Neural Networks, 2017, 17(1):455-468.

Li Z, Zhou F. FSSD: Feature Fusion Single Shot Multibox Detector. Neural Networks[J], 2017,2(9):52~69.

Liu S, Huang D, Wang Y, et al. Receptive Field Block Net for Accurate and Fast Object Detection[C]. European Conference on Computer Vision. Zurich, Switzerland: ECCV, 2018, 404-419.

Redmon J, Divvala S. K, Girshick R. B, et al. You Only Look Once: Unified, RealTime Object Detection[J]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA, 2016, 779-788.

Liu W, Anguelov D, Erhan D, et al. SSD: Single Shot MultiBox Detector[C]. European Conference on Computer Vision. Zurich, Switzerland: ECCV, 2016, 21-37.

Redmon J, Farhadi A. YOLO9000: Better, Faster, Stronger[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: IEEE Computer Society, 2017, 6517-6525.

Redmon J, Farhadi A. YOLOv3: An Incremental Improvement. Artificial Intelligence and Statistics[J], 2018, 88(252):687-690.

Liu Wei, Zhu Hong and Han Gui-Jin, "NMS-based blurred image sub-pixel registration," 2011 International Conference on Image Analysis and Signal Processing, Hubei, 2011, pp. 98-101, doi: 10.1109/IASP.2011.6109006.