[1]吴鹏,陈信华,马宇超,等.基于优化深度学习的电动桥铸件表面瑕疵识别方法[J].常州大学学报(自然科学版),2022,34(05):65-71.[doi:10.3969/j.issn.2095-0411.2022.05.009]
　WU Peng,CHEN Xinhua,MA Yuchao,et al.Research on Casting Surface Defects of Electric Bridge Identification Method Based on Optimal Deep Learning[J].Journal of Changzhou University(Natural Science Edition),2022,34(05):65-71.[doi:10.3969/j.issn.2095-0411.2022.05.009]
点击复制

基于优化深度学习的电动桥铸件表面瑕疵识别方法()
分享到：

常州大学学报(自然科学版)[ISSN:2095-0411/CN:32-1822/N]

卷:
第34卷
期数:
2022年05期
页码:
65-71
栏目:
机械制造及其自动化
出版日期:
2022-09-28

文章信息/Info

Title:
Research on Casting Surface Defects of Electric Bridge Identification Method Based on Optimal Deep Learning
文章编号:
2095-0411(2022)05-0065-07
作者:
吴鹏1; 2; 陈信华2; 马宇超1; 王鼎1; 陈帅1
(1.常州大学机械与轨道交通学院,江苏常州213164;2.溧阳市新力机械铸造有限公司,江苏常州213300)
Author(s):
WU Peng1; 2;  CHEN Xinhua2;  MA Yuchao1;  WANG Ding1;  CHEN Shuai1
(1.School of Mechanical Engineering and Rail Transit,Changzhou University, Changzhou 213164, China; 2.Liyang Xinli Machinery Casting Co., Ltd., Changzhou 213300, China)
关键词:
瑕疵检测;  深度学习;  特征提取;  迁移学习
Keywords:
defect detection;  deep learning;  feature extraction;  transfer learning
分类号:
TP 271
DOI:
10.3969/j.issn.2095-0411.2022.05.009
文献标志码:
A
摘要:
针对传统电动桥铸件瑕疵检测方法普遍存在效率低、检测精度低、人工成本高等问题,文章将优化深度学习方法应用于铸件表面瑕疵检测中,实现瑕疵自主精确检测识别。依据铸造厂待检测铸件表面特征,对铸件图像进行了前期图像预处理; 同时,基于优化网络模型结构,采用残差网络(Res-Net)与特征金字塔网络(FPN)构成的骨干结构,进行全图特征提取; 采用区域建议网络(RPN)生成大量特征建议区域,经非极大值抑制(NMS)处理后,分别输入全连接层与全卷积完成检测任务; 运用TensorFlow深度学习框架搭建模型,并采用迁移学习提高模型的泛化能力,实验结果显示,优化后的模型整体性能优于原始模型。
Abstract:
Traditional electric bridgecasting defect detection methods are of low efficiency, low detection accuracy, and high labor cost. Thus the optimal deep learning method is presented for the detection of the casting surface defect of electric bridges, and it can implement automatic accurate detection and identification of defects. Image preprocessing of the castings is done based on surface feature of castings. Feature extraction is done by Res-Net and FPN based on optimal model. A large number of feature recommended areas are generated by RPN. It is inputed into FC layer and FCN after NMS, then the defect detection is completed. The model is conducted by TensorFlow and the generalization ability of the model is improved by transfer learning. The experimental result shows that the whole performance of the optimized model is superior to the original model.

参考文献/References:

[1] 吕继东, 王艺洁, 夏正旺, 等. 基于改进的Mask R-CNN自然场景下苹果识别研究[J]. 常州大学学报(自然科学版), 2022, 34(1): 68-77.
[2] 徐建东. 基于最小二乘法拟合的Otsu快速图像分割方法[J]. 常州大学学报(自然科学版), 2021, 33(1): 70-76.
[3] 翁玉尚, 肖金球, 夏禹. 改进Mask R-CNN算法的带钢表面缺陷检测[J]. 计算机工程与应用, 2021, 57(19): 235-242.
[4] 李维刚, 叶欣, 赵云涛, 等. 基于改进YOLOv3算法的带钢表面缺陷检测[J]. 电子学报, 2020, 48(7): 1284-1292.
[5] AO X, ZHANG D, MA W, et al. Automatic metallic surface defect detection and recognition with convolutional neural networks[J].Applied Sciences-Basel, 2018, 8(9): 123-138.
[6] 程婧怡, 段先华, 朱伟. 改进YOLOv3的金属表面缺陷检测研究[J]. 计算机工程与应用, 2021, 57(19): 252-258.
[7] LE X Y, MEI J H, ZHANG H D, et al. A learning-based approach for surface defect detection using small image datasets[J]. Neurocomputing, 2020, 408: 112-120.
[8] 罗菁, 董婷婷, 宋丹, 等. 表面缺陷检测综述[J]. 计算机科学与探索, 2014(9): 1041-1048.
[9] 严琴, 赵全育. 高频噪声下的螺栓表面缺陷检测[J]. 测控技术, 2021, 40(5): 75-79.
[10] SAAD N H, AHMAD A E, SALEH H M, et al. Automatic semiconductor wafer image segmentation for defect detection using multilevel thresholding[J]. MATEC Web of Conferences, 2016, 78: 01103.
[11] GILBLAS R, SENTENAC T, ORTEU J J, et al. Detection of micro-cracks on highly specular reflectors: dimensioning a vision machine based on optical properties[J]. IEEE Sensors Journal, 2017, 17(23): 7901-7907.
[12] JIAN C X, GAO J, AO Y H. Automatic surface defect detection for mobile phone screen glass based on machine vision[J]. Applied Soft Computing, 2017, 52: 348-358.
[13] YU H L, LIANG Y L, LIANG H, et al. Recognition of wood surface defects with near infrared spectroscopy and machine vision[J]. Journal of Forestry Research, 2019, 30(6): 2379-2386.
[14] LI Z L, DONG M H, WEN S P, et al. CLU-CNNs: object detection for medical images[J]. Neurocomputing, 2019, 350: 53-59.
[15] 高雅, 朱秦岭, 王珑. 基于机器视觉的护套编织线缺陷检测系统设计[J]. 测控技术, 2017, 36(9): 31-34.
[16] HE K M, GKIOXARI G, DOLLÁR P, et al. Mask R-CNN[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2): 386-397.
[17] 刘晓杰, 罗印升, 张旻, 等. 基于机器视觉的零部件表面缺陷检测方法研究[J]. 现代电子技术, 2017, 40(24): 181-183.
[18] 钱基德, 陈斌, 钱基业, 等. 基于机器视觉的液晶屏Mura缺陷检测方法[J]. 计算机科学, 2018, 45(6): 296-300, 313.
[19] 宋辉, 李钊. 基于工业机器人的汽车轮毂表面缺陷的视觉检测系统设计[J]. 计算机测量与控制, 2018, 26(9): 13-16, 22.
[20] 李超, 孙俊. 基于机器视觉方法的焊缝缺陷检测及分类算法[J]. 计算机工程与应用, 2018, 54(6): 264-270.
[21] 马永福. 基于机器视觉的H型钢表面缺陷检测[J]. 冶金与材料, 2019, 39(6): 13-14.
(责任编辑:李艳,周安迪)

相似文献/References:

[1]宋志理,胡胜利,王峰.基于深度学习特征表示协同过滤算法[J].常州大学学报(自然科学版),2021,33(01):62.[doi:10.3969/j.issn.2095-0411.2021.01.010]
　SONG Zhili,HU Shengli,WANG Feng.Research on Cooperative Filtering Algorithm Based on Deep Learning Feature Representation[J].Journal of Changzhou University(Natural Science Edition),2021,33(05):62.[doi:10.3969/j.issn.2095-0411.2021.01.010]
[2]王洪元,徐志晨,陈海琴,等.基于金字塔分割和时空注意力的视频行人重识别[J].常州大学学报(自然科学版),2023,35(02):66.[doi:10.3969/j.issn.2095-0411.2023.02.008 ]
　WANG Hongyuan,XU Zhichen,CHEN Haiqin,et al.Video-based person re-identification based on pyramid segmentation and spatial-temporal attention[J].Journal of Changzhou University(Natural Science Edition),2023,35(05):66.[doi:10.3969/j.issn.2095-0411.2023.02.008 ]
[3]罗俊如,丁言瑞,徐明华,等.基于深度AUC最大化算法的井漏风险预测[J].常州大学学报(自然科学版),2024,36(03):34.[doi:10.3969/j.issn.2095-0411.2024.03.005]
　LUO Junru,DING Yanrui,XU Minghua,et al.Lost circulation prediction based on deep AUC maximization[J].Journal of Changzhou University(Natural Science Edition),2024,36(05):34.[doi:10.3969/j.issn.2095-0411.2024.03.005]

备注/Memo

备注/Memo:
收稿日期: 2022-05-31。
基金项目: 江苏省产学研合作资助项目(BYBY2021221); 2021年江苏省研究生科研与实践创新资助项目(SJCX21_1277); 溧阳市科技资助项目(培育创新项目)(XMSB20210001)。
作者简介: 吴鹏(1987—), 男, 江苏常州人, 博士, 研究员。通信联系人: 陈信华(1967—), E-mail: xlcxh2016@126.com
更新日期/Last Update: 1900-01-01