基于轻量级网络的钢铁表面缺陷分类

doi:10.11772/j.issn.1001-9081.2020081244

计算机应用 ›› 2021, Vol. 41 ›› Issue (6): 1836-1841.DOI: 10.11772/j.issn.1001-9081.2020081244

所属专题：前沿与综合应用

基于轻量级网络的钢铁表面缺陷分类

史杨潇¹, 章军¹, 陈鹏¹, 王兵²

1. 安徽大学电气工程与自动化学院, 合肥 230601;
2. 安徽工业大学电气信息学院, 安徽马鞍山 243002

收稿日期:2020-08-18 修回日期:2020-10-27 出版日期:2021-06-10 发布日期:2020-12-09
通讯作者: 史杨潇
作者简介:史杨潇(1997-),男,安徽合肥人,硕士研究生,主要研究方向:深度学习、缺陷检测;章军(1971-),男,安徽合肥人,教授,博士,主要研究方向:深度学习在视频分析中的应用、化学信息学;陈鹏(1978-),男,安徽肥东人,教授,博士,主要研究方向:机器学习、生物信息学;王兵(1976-),男,安徽怀宁人,副教授,博士,主要研究方向:模式识别、生物信息学。
基金资助:
国家自然科学基金资助项目（61872004）。

Classification of steel surface defects based on lightweight network

SHI Yangxiao¹, ZHANG Jun¹, CHEN Peng¹, WANG Bing²

1. School of Electrical Engineering and Automation, Anhui University, Hefei Anhui 230601, China;
2. School of Electrical and Information, Anhui University of Technology, Ma'anshan Anhui 243002, China

Received:2020-08-18 Revised:2020-10-27 Online:2021-06-10 Published:2020-12-09
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61872004).

摘要/Abstract

摘要： 缺陷分类是钢铁表面缺陷检测的重要内容。在卷积神经网络（CNN）取得良好效果的同时，网络日益增长的参数量耗费了大量计算成本，为缺陷分类任务在个人计算机或低算力设备上的部署带来了巨大的挑战。针对上述问题，提出了一种新颖的轻量级网络模型Mix-Fusion。首先，通过组卷积和通道洗牌两种操作，在保持精度的同时有效降低计算成本；其次，利用一个狭窄的特征映射对组间信息进行融合编码，并将生成的特征与原始网络结合，从而有效解决了“稀疏连接”卷积阻碍组间信息交换的问题；最后，用一种新型的混合卷积（MixConv）替代了传统的深度卷积（DWConv），以进一步提高模型的性能。在NEU-CLS数据集上的实验结果表明，Mix-Fusion网络在缺陷分类任务中的浮点运算次数和分类准确率分别为43.4 MFLOPs和98.61%。相较于ShuffleNetV2和MobileNetV2网络，Mix-Fusion网络不仅降低了模型参数，压缩了模型大小，同时还得到了更好的分类精度。

关键词: 表面缺陷检测, 缺陷分类, 模型加速, 深度学习, 轻量级网络

Abstract: Defect classification is an important part of steel surface defect detection. When the Convolutional Neural Network (CNN) has achieved good results, the increasing number of network parameters consumes a lot of computing cost, which brings great challenges to the deployment of defect classification tasks on personal computers or low computing power devices. Focusing on the above problem, a novel lightweight network model named Mix-Fusion was proposed. Firstly, two operations of group convolution and channel-shuffle were used to reduce the computational cost while maintaining the accuracy. Secondly, a narrow feature mapping was used to fuse and encode the information between the groups, and the generated features were combined with the original network, so as to effectively solve the problem that "sparse connection" convolution hindered the information exchange between the groups. Finally, a new type of Mixed depthwise Convolution (MixConv) was used to replace the traditional DepthWise Convolution (DWConv) to further improve the performance of the model. Experimental results on NEU-CLS dataset show that, the number of floating-point operations and classification accuracy of Mix-Fusion network in defect classification task is 43.4 Million FLoating-point Operations Per second (MFLOPs) and 98.61% respectively. Compared to the networks of ShuffleNetV2 and MobileNetV2, the proposed Mix-Fusion network reduces the model parameters and compresses the model size effectively, as well as obtains the better classification accuracy.

Key words: surface defect detection, defect classification, model acceleration, deep learning, lightweight network

中图分类号:

TP391.4

史杨潇, 章军, 陈鹏, 王兵. 基于轻量级网络的钢铁表面缺陷分类[J]. 计算机应用, 2021, 41(6): 1836-1841.

SHI Yangxiao, ZHANG Jun, CHEN Peng, WANG Bing. Classification of steel surface defects based on lightweight network[J]. Journal of Computer Applications, 2021, 41(6): 1836-1841.

参考文献

[1] HE Y,SONG K,MENG Q,et al. An end-to-end steel surface defect detection approach via fusing multiple hierarchical features[J]. IEEE Transactions on Instrumentation and Measurement, 2020,69(4):1493-1504.
[2] DONG H,SONG K,HE Y,et al. PGA-Net:pyramid feature fusion and global context attention network for automated surface defect detection[J]. IEEE Transactions on Industrial Informatics, 2020,16(12):7448-7458.
[3] RAZAVIAN A S, AZIZPOUR H, SULLIVAN J, et al. CNN features off-the-shelf:an astounding baseline for recognition[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops. Piscataway:IEEE, 2014:512-519.
[4] AMATO G,FALCHI F,GENNARO C. Geometric consistency checks for kNN based image classification relying on local features[C]//Proceedings of the 2011 4th International Conference on Similarity Search and Applications. New York:ACM, 2011:81-88.
[5] LIN Y,LV F,ZHU S,et al. Large-scale image classification:fast feature extraction and SVM training[C]//Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2011:1689-1696.
[6] WANG J, YANG Y, MAO J, et al. CNN-RNN:a unified framework for multi-label image classification[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2016:2285-2294.
[7] SIMONYAN K,ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2020-07-20]. http://arxiv.org/pdf/1409.1556.pdf.
[8] HE K,ZHANG X,REN S,et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2016:770-778.
[9] HOWARD A G,ZHU M,CHEN B,et al. MobileNets:efficient convolutional neural networks for mobile vision applications[EB/OL].[2020-06-08]. https://arxiv.org/pdf/1704.04861.pdf.
[10] SANDLER M,HOWARD A,ZHU M,et al. MobileNetV2:inverted residuals and linear bottlenecks[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:4510-4520.
[11] ZHANG X,ZHOU X,LIN M,et al. ShuffleNet:an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:6848-6856.
[12] IANDOLA F N,HAN S,MOSKEWICZ M W,et al. SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size[EB/OL].[2020-07-07]. https://arxiv.org/pdf/1602.07360.pdf.
[13] HU J,SHEN L,SUN G. Squeeze-and-excitation networks[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:7132-7141.
[14] 姚明海, 陈志浩. 基于深度主动学习的磁片表面缺陷检测[J]. 计算机测量与控制, 2018, 26(9):29-33.(YAO M H,CHEN Z H. Deep active learning in detection of surface defects on magnetic sheet[J]. Computer Measurement and Control,2018,26(9):29-33.)
[15] 张琪, 王国栋, 赵希梅, 等. 基于轻量级卷积神经网络的肝部病理组织切片分类[J]. 青岛大学学报(自然科学版), 2018, 31(4):76-82. (ZHANG Q,WANG G D,ZHAO X M,et al. Classification of liver pathological tissue sections based on the lightweight convolutional neural network[J]. Journal of Qingdao University(Natural Science Edition),2018,31(4):76-82.)
[16] QIN Z,ZHANG Z,ZHANG S,et al. Merging and evolution:improving convolutional neural networks for mobile applications[EB/OL].[2020-07-12]. https://arxiv.org/pdf/1803.09127.pdf.
[17] TAN M, LE Q V. MixConv:mixed depthwise convolutional kernels[EB/OL].[2020-08-02]. https://arxiv.org/pdf/1907.09595.pdf.
[18] IOFFE S,SZEGEDY C. Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 201532nd International Conference on Machine Learning. New York:ACM, 2015:448-456.
[19] CHOLLET F. Xception:deep learning with depthwise separable convolutions[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:1800-1807.
[20] SZEGEDY C,VANHOUCKE V,IOFFE S,et al. Rethinking the inception architecture for computer vision[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2016:2818-2826.
[21] ZOPH B, VASUDEVAN V, SHLENS J, et al. Learning transferable architectures for scalable image recognition[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:8697-8710.
[22] SONG K,YAN Y. A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects[J]. Applied Surface Science,2013,285(Pt B):858-864.
[23] REN R,HUNG T,TAN K C,et al. A generic deep-learningbased approach for automated surface inspection[J]. IEEE Transactions on Cybernetics,2018,48(3):929-940.
[24] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2015:1-9.
[25] KRIZHEVSKY A,SUTSKEVER I,HINTON G E. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 2012 25th International Conference on Neural Information Processing Systems. Red Hook:Curran Associates Inc.,2012:1097-1105.

基于轻量级网络的钢铁表面缺陷分类

Classification of steel surface defects based on lightweight network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	谢德峰, 吉建民. 融入句法感知表示进行句法增强的语义解析[J]. 计算机应用, 2021, 41(9): 2489-2495.
[2]	代雨柔, 杨庆, 张凤荔, 周帆. 基于自监督学习的社交网络用户轨迹预测模型[J]. 计算机应用, 2021, 41(9): 2545-2551.
[3]	陈成瑞, 孙宁, 何世彪, 廖勇. 面向C-V2X通信的基于深度学习的联合信道估计与均衡算法[J]. 计算机应用, 2021, 41(9): 2687-2693.
[4]	郑志强, 胡鑫, 翁智, 王雨禾, 程曦. 基于改进DenseNet的牛眼图像特征提取方法[J]. 计算机应用, 2021, 41(9): 2780-2784.
[5]	赵宏, 孔东一. 图像特征注意力与自适应注意力融合的图像内容中文描述[J]. 计算机应用, 2021, 41(9): 2496-2503.
[6]	徐江浪, 李林燕, 万新军, 胡伏原. 结合目标检测的室内场景识别方法[J]. 计算机应用, 2021, 41(9): 2720-2725.
[7]	曹玉红, 徐海, 刘荪傲, 王紫霄, 李宏亮. 基于深度学习的医学影像分割研究综述[J]. 计算机应用, 2021, 41(8): 2273-2287.
[8]	秦斌斌, 彭良康, 卢向明, 钱江波. 司机分心驾驶检测研究进展[J]. 计算机应用, 2021, 41(8): 2330-2337.
[9]	何正海, 线岩团, 王蒙, 余正涛. 融合句法指导与字符注意力机制的案情阅读理解方法[J]. 计算机应用, 2021, 41(8): 2427-2431.
[10]	李亚芳, 梁烨, 冯韦玮, 祖宝开, 康玉健. 基于社区优化的深度网络嵌入方法[J]. 计算机应用, 2021, 41(7): 1956-1963.
[11]	高钦泉, 黄炳城, 刘文哲, 童同. 基于改进CenterNet的竹条表面缺陷检测方法[J]. 计算机应用, 2021, 41(7): 1933-1938.
[12]	王月, 江逸茗, 兰巨龙. 基于改进三元组网络和K近邻算法的入侵检测[J]. 计算机应用, 2021, 41(7): 1996-2002.
[13]	侯笑晗, 金国栋, 谭力宁, 薛远亮. 基于自适应和最优特征的合成孔径雷达舰船检测方法[J]. 计算机应用, 2021, 41(7): 2150-2155.
[14]	杜炎, 吕良福, 焦一辰. 基于模糊推理的模糊原型网络[J]. 计算机应用, 2021, 41(7): 1885-1890.
[15]	刘世泽, 朱奕达, 陈润泽, 罗海勇, 赵方, 孙艺, 王宝会. 基于残差时域注意力神经网络的交通模式识别算法[J]. 计算机应用, 2021, 41(6): 1557-1565.