{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T22:07:04Z","timestamp":1778969224104,"version":"3.51.4"},"reference-count":38,"publisher":"National Library of Serbia","issue":"1","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"unspecified","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["ComSIS","COMPUT SCI INF SYST","COMPUT SCI INFORM SY","COMPUTER SCI INFORM","COMSIS J"],"published-print":{"date-parts":[[2026]]},"abstract":"<jats:p>Quality inspection of steel plate welding is critical in industrial manufacturing. However, weld proximity defects often present diverse morphologies, overlapping regions, and dense distributions, posing challenges to accurate industrial defect inspection. Therefore, we propose an industrial detector based on the EP-YOLOv7. First, an Efficient Multiscale Channel Attention (EMCA) is introduced to strengthen multi-scale feature perception and improve the model\u2019s focus on weld proximity defects. Second, the EMCA module is integrated into the Efficient Layer Aggregation Network to enhance feature fusion and defect representation. Finally, a Partial-Bottleneck Decoupling Predictor Head (P-BD Head) is designed to significantly improve localization accuracy and reduce missed detections of small targets. Experimental evaluations on both a self-built weld proximity defect dataset and a public generalization dataset show that EP-YOLOv7 achieves mAP of 85.2%\/56.2% and F1 scores of 80.3%\/43.3%. Meanwhile, the model size increases by only 0.6 MB (total 37.9 MB), demonstrating that the proposed approach delivers substantial accuracy gains while maintaining lightweight computational complexity, suitable for practical industrial applications.<\/jats:p>","DOI":"10.2298\/csis250922007z","type":"journal-article","created":{"date-parts":[[2026,1,20]],"date-time":"2026-01-20T10:44:04Z","timestamp":1768905844000},"page":"255-276","source":"Crossref","is-referenced-by-count":0,"title":["Weld proximity defect detection model for steel thin plates based on EP-YOLOv7"],"prefix":"10.2298","volume":"23","author":[{"given":"Runmei","family":"Zhang","sequence":"first","affiliation":[{"name":"School of Electronic and Information Engineering, Anhui Jianzhu University, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingwei","family":"Fan","sequence":"additional","affiliation":[{"name":"School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zihua","family":"Chen","sequence":"additional","affiliation":[{"name":"Provincial and Ministerial Key Laboratory, Chang\u2019an University, Xi\u2019an, China + School of Electronic and Information Engineering, Anhui Jianzhu University, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhong","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bin","family":"Yuan","sequence":"additional","affiliation":[{"name":"School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1078","reference":[{"key":"ref1","doi-asserted-by":"crossref","unstructured":"Matthew Rumsa, Michele John, and Wahidul Biswas. Global steel decarbonisation roadmaps: Near-zero by 2050. Environmental Impact Assessment Review, 112:107807, 2025.","DOI":"10.1016\/j.eiar.2025.107807"},{"key":"ref2","doi-asserted-by":"crossref","unstructured":"Xiaopeng Wang, Uwe Zscherpel, Paolo Tripicchio, Salvatore D\u2019Avella, Baoxin Zhang, Juntao Wu, Zhimin Liang, Shaoxin Zhou, and Xinghua Yu. A comprehensive review of welding defect recognition from x-ray images. Journal of Manufacturing Processes, 140:161-180, 2025.","DOI":"10.1016\/j.jmapro.2025.02.039"},{"key":"ref3","doi-asserted-by":"crossref","unstructured":"Changlu Xu, Linsheng Li, Jiwei Li, and Chuanbo Wen. Surface defects detection and identification of lithium battery pole piece based on multi-feature fusion and pso-svm. Ieee Access, 9:85232-85239, 2021.","DOI":"10.1109\/ACCESS.2021.3067641"},{"key":"ref4","doi-asserted-by":"crossref","unstructured":"Ruichen Ma, Jinglong Chen, Yong Feng, Zitong Zhou, and Jingsong Xie. Ela-yolo: An efficient method with linear attention for steel surface defect detection during manufacturing. Advanced Engineering Informatics, 65:103377, 2025.","DOI":"10.1016\/j.aei.2025.103377"},{"key":"ref5","doi-asserted-by":"crossref","unstructured":"Dorra Zaibi, Maroua Salhi, Khaoula Tbarki, and Riadh Ksantini. An incremental software defect detection model based on support vector machine. Engineering Computations, 42(1):76- 95, 2025.","DOI":"10.1108\/EC-11-2023-0799"},{"key":"ref6","doi-asserted-by":"crossref","unstructured":"Yirga Kene Molla and Emebet Abeje Mitiku. Cnn-hog based hybrid feature mining for classification of coffee bean varieties using image processing. Multimedia Tools and Applications, 84(2):749-764, 2025.","DOI":"10.1007\/s11042-024-18952-z"},{"key":"ref7","doi-asserted-by":"crossref","unstructured":"Qianwen Ye, Yiwei Dong, Xiaoxin Zhang, Duo Zhang, and Siyi Wang. Robustness defect detection: improving the performance of surface defect detection in interference environment. Optics and Lasers in Engineering, 175:108035, 2024.","DOI":"10.1016\/j.optlaseng.2024.108035"},{"key":"ref8","doi-asserted-by":"crossref","unstructured":"Xianghe Zou, Chongyang Wu, Hongen Liu, Zhangwei Yu, and Xianyan Kuang. An accurate object detection of wood defects using an improved faster r-cnn model. Wood Material Science & Engineering, 20(2):413-419, 2025.","DOI":"10.1080\/17480272.2024.2352605"},{"key":"ref9","unstructured":"Zhige He and Yuanqing He. As-faster-rcnn: An improved object detection algorithm for airport scene based on faster r-cnn. IEEE Access, 2025."},{"key":"ref10","doi-asserted-by":"crossref","unstructured":"XinWeng, Qihua Ma, Qilin Li, andWenchaoWang. Improved mask r-cnn algorithm: Multi-ore detection and positioning based multi-sensor fusion in complex field environment. Measurement, 246:116602, 2025.","DOI":"10.1016\/j.measurement.2024.116602"},{"key":"ref11","doi-asserted-by":"crossref","unstructured":"Ling Tan, Hui Wu, Zifeng Xu, and Jingming Xia. Multi-object garbage image detection algorithm based on sp-ssd. Expert Systems with Applications, 263:125773, 2025.","DOI":"10.1016\/j.eswa.2024.125773"},{"key":"ref12","doi-asserted-by":"crossref","unstructured":"Beilong Chen, Mingjun Wei, Jianuo Liu, Hui Li, Chenxu Dai, Jinyun Liu, and Zhanlin Ji. Efsyolo: a lightweight network based on steel strip surface defect detection. Measurement Science and Technology, 35(11):116003, 2024.","DOI":"10.1088\/1361-6501\/ad66fe"},{"key":"ref13","doi-asserted-by":"crossref","unstructured":"Wenqi Cui, Zhenye Li, Anning Duanmu, Sheng Xue, Yiren Guo, Chao Ni, Tingting Zhu, and Yajun Zhang. Ccg-yolov7: A wood defect detection model for small targets using improved yolov7. IEEE Access, 12:10575-10585, 2024.","DOI":"10.1109\/ACCESS.2024.3352445"},{"key":"ref14","doi-asserted-by":"crossref","unstructured":"Keji Mao, Runhui Jin, Lingkang Ying, Xingda Yao, Guanglin Dai, and Kai Fang. Sc-yolo: Provide application-level recognition and perception capabilities for smart city industrial cyberphysical system. IEEE Systems Journal, 17(4):5118-5129, 2023.","DOI":"10.1109\/JSYST.2023.3292113"},{"key":"ref15","doi-asserted-by":"crossref","unstructured":"Yongbin Chen, Jingran Wang, and Guitang Wang. Intelligent welding defect detection model on improved r-cnn. IETE Journal of Research, 69(12):9235-9244, 2023.","DOI":"10.1080\/03772063.2022.2040387"},{"key":"ref16","doi-asserted-by":"crossref","unstructured":"Rijun Wang, Fulong Liang, Xiangwei Mou, Lintao Chen, Xinye Yu, Zhujing Peng, and Hongyang Chen. Development of an improved yolov7-based model for detecting defects on strip steel surfaces. Coatings, 13(3):536, 2023.","DOI":"10.3390\/coatings13030536"},{"key":"ref17","doi-asserted-by":"crossref","unstructured":"Quan Lu, Kehong Lin, and Linfei Yin. 3d attention-focused pure convolutional target detection algorithm for insulator defect detection. Expert Systems with Applications, 249:123720, 2024.","DOI":"10.1016\/j.eswa.2024.123720"},{"key":"ref18","doi-asserted-by":"crossref","unstructured":"Qingqi Zhang, XinboWang, HaoWang, and Jie Li. Vehicle small target detection algorithm for remote sensing images based on improved yolov7-tiny. In 2024 5th International Conference on Artificial Intelligence and Computer Engineering (ICAICE), pages 183-187. IEEE, 2024.","DOI":"10.1109\/ICAICE63571.2024.10864056"},{"key":"ref19","doi-asserted-by":"crossref","unstructured":"Jinlai Zhang, Wenjie Yang, Yumei Chen, Mingkang Ding, Huiling Huang, Bingkun Wang, Kai Gao, Shuhan Chen, and Ronghua Du. Fast object detection of anomaly photovoltaic (pv) cells using deep neural networks. Applied Energy, 372:123759, 2024.","DOI":"10.1016\/j.apenergy.2024.123759"},{"key":"ref20","doi-asserted-by":"crossref","unstructured":"Jierun Chen, Shiu-hong Kao, Hao He, Weipeng Zhuo, Song Wen, Chul-Ho Lee, and S-H Gary Chan. Run, don\u2019t walk: chasing higher flops for faster neural networks. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition, pages 12021-12031, 2023.","DOI":"10.1109\/CVPR52729.2023.01157"},{"key":"ref21","doi-asserted-by":"crossref","unstructured":"Zhijin Li, Jinfeng Yan, Jie Zhou, Xiaozhen Fan, and Jiahui Tang. An efficient smd-pcba detection based on yolov7 network model. Engineering Applications of Artificial Intelligence, 124:106492, 2023.","DOI":"10.1016\/j.engappai.2023.106492"},{"key":"ref22","doi-asserted-by":"crossref","unstructured":"Yi Zhang and Qingjian Ni. A novel weld-seam defect detection algorithm based on the s-yolo model. Axioms, 12(7):697, 2023.","DOI":"10.3390\/axioms12070697"},{"key":"ref23","unstructured":"Shunxiang Zhang, Meng Chen, Kuan-Ching Li, Hua Wen, and Liang Sun. Fire detection models based on attention mechanisms and multiscale features. Computer Science and Information Systems, (00):59-59, 2025."},{"key":"ref24","doi-asserted-by":"crossref","unstructured":"Zihua Chen, Runmei Zhang, Meng-Yen Hsieh, Alireza Souri, and Kuan-Ching Li. Average sigmoid-tanh attention and multi-filter partially decoupled mechanism via yolov7 for detecting weld proximity defects. Metallurgical and Materials Transactions B, pages 1-15, 2025.","DOI":"10.1007\/s11663-025-03595-8"},{"key":"ref25","doi-asserted-by":"crossref","unstructured":"Yushan Zhao, Kuan-Ching Li, Shunxiang Zhang, and Tongzhou Ye. Hyperbolic graph attention network fusing long-context for technical keyphrase extraction. Information Fusion, 120:103061, 2025.","DOI":"10.1016\/j.inffus.2025.103061"},{"key":"ref26","doi-asserted-by":"crossref","unstructured":"Jie Hu, Li Shen, and Gang Sun. Squeeze-and-excitation networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7132-7141, 2018.","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref27","doi-asserted-by":"crossref","unstructured":"Junpeng Xu, Xiangbo Zhu, Lei Shi, Jin Li, and Ziman Guo. Squeeze-and-excitation attention and bi-directional feature pyramid network for filter screens surface detection. Journal of Electronic Imaging, 33(4):043044-043044, 2024.","DOI":"10.1117\/1.JEI.33.4.043044"},{"key":"ref28","doi-asserted-by":"crossref","unstructured":"Zhaochao Li, Linxuan Xiao, Meiling Shen, and Xiya Tang. A lightweight yolov8-based model with squeeze-and-excitation version 2 for crack detection of pipelines. Applied Soft Computing, page 113260, 2025.","DOI":"10.1016\/j.asoc.2025.113260"},{"key":"ref29","doi-asserted-by":"crossref","unstructured":"Sanghyun Woo, Jongchan Park, Joon-Young Lee, and In So Kweon. Cbam: Convolutional block attention module. In Proceedings of the European conference on computer vision (ECCV), pages 3-19, 2018.","DOI":"10.1007\/978-3-030-01234-2_1"},{"key":"ref30","doi-asserted-by":"crossref","unstructured":"Hajar Zoubir, Mustapha Rguig, Mohamed El Aroussi, Rachid Saadane, and Abdellah Chehri. Pixel-level concrete bridge crack detection using convolutional neural networks, gabor filters, and attention mechanisms. Engineering Structures, 314:118343, 2024.","DOI":"10.1016\/j.engstruct.2024.118343"},{"key":"ref31","doi-asserted-by":"crossref","unstructured":"Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko. End-to-end object detection with transformers. In European conference on computer vision, pages 213-229. Springer, 2020.","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"ref32","doi-asserted-by":"crossref","unstructured":"Liang Tian Jiao, Pan Wei Guo, Bao Hong, and Pan Feng. Vehicle wheel weld detection based on improved yolo v4 algorithm. , 46(2):271-279, 2022.","DOI":"10.18287\/2412-6179-CO-887"},{"key":"ref33","doi-asserted-by":"crossref","unstructured":"Moyun Liu, Youping Chen, Jingming Xie, Lei He, and Yang Zhang. Lf-yolo: A lighter and faster yolo for weld defect detection of x-ray image. IEEE sensors journal, 23(7):7430-7439, 2023.","DOI":"10.1109\/JSEN.2023.3247006"},{"key":"ref34","doi-asserted-by":"crossref","unstructured":"Zihua Chen, Yu Zheng, Ling-Huey Li, Tien-Hsiung Weng, Kuan-Ching Li, and Aneta Poniszewska-Maranda. An improved dilated-transposed convolution detector of weld proximity defects. IEEE Access, 2024.","DOI":"10.1109\/ACCESS.2024.3484589"},{"key":"ref35","doi-asserted-by":"crossref","unstructured":"Xiangqian Xu and Xing Li. Research on surface defect detection algorithm of pipeline weld based on yolov7. Scientific reports, 14(1):1881, 2024.","DOI":"10.1038\/s41598-024-52451-3"},{"key":"ref36","unstructured":"Yong Ren, Dong Liu, and Sanhong Gu. A lightweight defect classification method for latex gloves based on image enhancement. Computer Science and Information Systems, (00):7-7, 2025."},{"key":"ref37","doi-asserted-by":"crossref","unstructured":"Qilong Wang, Banggu Wu, Pengfei Zhu, Peihua Li, Wangmeng Zuo, and Qinghua Hu. Ecanet: Efficient channel attention for deep convolutional neural networks. In Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition, pages 11534-11542, 2020.","DOI":"10.1109\/CVPR42600.2020.01155"},{"key":"ref38","unstructured":"Zheng Ge, Songtao Liu, Feng Wang, Zeming Li, and Jian Sun. Yolox: Exceeding yolo series in 2021. arXiv preprint arXiv:2107.08430, 2021."}],"container-title":["Computer Science and Information Systems"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,5,16]],"date-time":"2026-05-16T21:34:18Z","timestamp":1778967258000},"score":1,"resource":{"primary":{"URL":"https:\/\/doiserbia.nb.rs\/Article.aspx?ID=1820-02142600007Z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":38,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026]]}},"URL":"https:\/\/doi.org\/10.2298\/csis250922007z","relation":{},"ISSN":["1820-0214","2406-1018"],"issn-type":[{"value":"1820-0214","type":"print"},{"value":"2406-1018","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]}}}