{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,17]],"date-time":"2025-10-17T14:30:16Z","timestamp":1760711416034,"version":"3.41.2"},"reference-count":29,"publisher":"Emerald","issue":"2","license":[{"start":{"date-parts":[[2024,9,17]],"date-time":"2024-09-17T00:00:00Z","timestamp":1726531200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.emerald.com\/insight\/site-policies"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IR"],"published-print":{"date-parts":[[2025,3,4]]},"abstract":"<jats:sec><jats:title content-type=\"abstract-subheading\">Purpose<\/jats:title>\n<jats:p>This paper aims to propose a lightweight, high-accuracy object detection model designed to enhance seam tracking quality under strong arcs and splashes condition. Simultaneously, the model aims to reduce computational costs.<\/jats:p>\n<\/jats:sec>\n<jats:sec><jats:title content-type=\"abstract-subheading\">Design\/methodology\/approach<\/jats:title>\n<jats:p>The lightweight model is constructed based on Single Shot Multibox Detector (SSD). First, a neural architecture search method based on meta-learning and genetic algorithm is introduced to optimize pruning strategy, reducing human intervention and improving efficiency. Additionally, the Alternating Direction Method of Multipliers (ADMM) is used to perform structural pruning on SSD, effectively compressing the model with minimal loss of accuracy.<\/jats:p>\n<\/jats:sec>\n<jats:sec><jats:title content-type=\"abstract-subheading\">Findings<\/jats:title>\n<jats:p>Compared to state-of-the-art models, this method better balances feature extraction accuracy and inference speed. Furthermore, seam tracking experiments on this welding robot experimental platform demonstrate that the proposed method exhibits excellent accuracy and robustness in practical applications.<\/jats:p>\n<\/jats:sec>\n<jats:sec><jats:title content-type=\"abstract-subheading\">Originality\/value<\/jats:title>\n<jats:p>This paper presents an innovative approach that combines ADMM structural pruning and meta-learning-based neural architecture search to significantly enhance the efficiency and performance of the SSD network. This method reduces computational cost while ensuring high detection accuracy, providing a reliable solution for welding robot laser vision systems in practical applications.<\/jats:p>\n<\/jats:sec>","DOI":"10.1108\/ir-05-2024-0233","type":"journal-article","created":{"date-parts":[[2024,9,16]],"date-time":"2024-09-16T06:24:25Z","timestamp":1726467865000},"page":"195-203","source":"Crossref","is-referenced-by-count":1,"title":["Model optimization and acceleration method based on meta-learning and model pruning for laser vision weld tracking system"],"prefix":"10.1108","volume":"52","author":[{"given":"Yanbiao","family":"Zou","sequence":"first","affiliation":[]},{"given":"Jianhui","family":"Yang","sequence":"additional","affiliation":[]}],"member":"140","published-online":{"date-parts":[[2024,9,17]]},"reference":[{"year":"2020","key":"key2025022811244937100_ref001","article-title":"YOLOv4: optimal speed and accuracy of object detection"},{"issue":"1","key":"key2025022811244937100_ref002","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1561\/2200000016","article-title":"Distributed optimization and statistical learning via the alternating direction method of multipliers","volume":"3","year":"2010","journal-title":"Foundations and Trends\u00ae in Machine Learning"},{"year":"2020","key":"key2025022811244937100_ref003","article-title":"End-to-End object detection with transformers"},{"issue":"4","key":"key2025022811244937100_ref004","first-page":"48","article-title":"Model compression and hardware acceleration for neural networks: a comprehensive survey","volume":"108","year":"2020","journal-title":"Proceedings of the IEEE"},{"issue":"7\/8","key":"key2025022811244937100_ref005","first-page":"15","article-title":"A weld line detection robot based on structure light for automatic NDT","volume":"111","year":"2020","journal-title":"International Journal of Advanced Manufacturing Technology"},{"year":"2020","key":"key2025022811244937100_ref006","article-title":"An image is worth 16x16 words: transformers for image recognition at scale"},{"first-page":"6568","article-title":"CenterNet: keypoint triplets for object detection","year":"2019","key":"key2025022811244937100_ref007"},{"key":"key2025022811244937100_ref008","first-page":"13683494","article-title":"Bayesian optimization with clustering and rollback for CNN auto pruning","volume":"XXIII","year":"2022","journal-title":"Computer Vision, Eccv 2022, Pt"},{"issue":"9","key":"key2025022811244937100_ref009","first-page":"9","article-title":"Vision-based initial weld point positioning using the geometric relationship between two seams","volume":"66","year":"2013","journal-title":"International Journal of Advanced Manufacturing Technology"},{"year":"2015","key":"key2025022811244937100_ref010","article-title":"Deep compression: compressing deep neural networks with pruning, trained quantization and Huffman coding"},{"key":"key2025022811244937100_ref011","article-title":"Learning both weights and connections for efficient neural networks","volume":"289","year":"2015","journal-title":"Advances in Neural Information Processing Systems 28 (Nips 2015)"},{"key":"key2025022811244937100_ref012","first-page":"1121118","article-title":"AMC: autoML for model compression and acceleration on mobile devices","volume":"VII","year":"2018","journal-title":"Computer Vision - Eccv 2018, Pt"},{"key":"key2025022811244937100_ref013","first-page":"10798","article-title":"DNN compression by ADMM-based joint pruning","volume":"239","year":"2022","journal-title":"Knowledge-Based Systems"},{"key":"key2025022811244937100_ref014","doi-asserted-by":"crossref","first-page":"2999","DOI":"10.1109\/ICCV.2017.324","article-title":"Focal loss for dense object detection","volume-title":"2017 IEEE International Conference On Computer Vision (ICCV)","year":"2017"},{"article-title":"AutoCompress: an automatic DNN structured pruning framework for Ultra-High compression rates","volume-title":"Thirty-Fourth AAAI Conference On Artificial Intelligence, The Thirty-Second Innovative Applications Of Artificial Intelligence Conference And The Tenth AAAI Symposium On Educational Advances In Artificial Intelligence","year":"2020","key":"key2025022811244937100_ref015"},{"article-title":"SSD: single shot multi box detector","volume-title":"Computer Vision - ECCV 2016, PT I","year":"2016","key":"key2025022811244937100_ref016"},{"key":"key2025022811244937100_ref017","doi-asserted-by":"crossref","first-page":"2755","DOI":"10.1109\/ICCV.2017.298","article-title":"Learning efficient convolutional networks through network slimming","volume-title":"2017 IEEE International Conference On Computer Vision (ICCV)","year":"2017"},{"first-page":"10","article-title":"MetaPruning: meta learning for automatic neural network channel pruning","year":"2019","key":"key2025022811244937100_ref018"},{"issue":"12","key":"key2025022811244937100_ref019","first-page":"37","article-title":"Efficient deep learning: a survey on making deep learning models smaller, faster, and better","volume":"55","year":"2023","journal-title":"ACM Computing Surveys"},{"year":"2020","key":"key2025022811244937100_ref020","article-title":"Designing network design spaces"},{"first-page":"14","article-title":"ADMM-NN: an Algorithm-Hardware Co-Design framework of DNNs using alternating direction method of multipliers","year":"2019","key":"key2025022811244937100_ref021"},{"key":"key2025022811244937100_ref022","first-page":"4510","article-title":"MobileNetV2: inverted residuals and linear bottlenecks","volume-title":"2018 IEEE\/CVF Conference On Computer Vision And Pattern Recognition (CVPR)","year":"2018"},{"issue":"14","key":"key2025022811244937100_ref023","first-page":"30","article-title":"GAN-Based image dehazing for intelligent weld shape classification and tracing using deep learning","volume":"12","year":"2022","journal-title":"Applied Sciences-Basel"},{"issue":"4","key":"key2025022811244937100_ref024","first-page":"11","article-title":"Automatic identification of Multi-Type weld seam based on vision sensor with Silhouette-Mapping","volume":"21","year":"2021","journal-title":"Ieee Sensors Journal"},{"issue":"3","key":"key2025022811244937100_ref025","first-page":"11","article-title":"Features for image processing of OCT images for seam tracking applications in laser welding","volume":"34","year":"2022","journal-title":"Journal of Laser Applications"},{"key":"key2025022811244937100_ref026","doi-asserted-by":"crossref","first-page":"209","DOI":"10.1016\/j.neucom.2018.05.073","article-title":"Non-convex weighted \u2113p nuclear norm based ADMM framework for image restoration","volume":"311","year":"2018","journal-title":"Neurocomputing"},{"key":"key2025022811244937100_ref027","first-page":"11212191","article-title":"A systematic DNN weight pruning framework using alternating direction method of multipliers","volume":"PT VIII","year":"2018","journal-title":"Computer Vision - Eccv 2018"},{"key":"key2025022811244937100_ref028","first-page":"9","article-title":"Laser vision seam tracking system based on image processing and continuous convolution operator tracker","volume":"105","year":"2018","journal-title":"Optics and Lasers in Engineering"},{"issue":"7","key":"key2025022811244937100_ref029","first-page":"14","article-title":"A robust detector for automated welding seam tracking system","volume":"143","year":"2021","journal-title":"Journal of Dynamic Systems Measurement and Control-Transactions of the Asme"}],"container-title":["Industrial Robot: the international journal of robotics research and application"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IR-05-2024-0233\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IR-05-2024-0233\/full\/html","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T21:39:13Z","timestamp":1753393153000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.emerald.com\/ir\/article\/52\/2\/195-203\/1239446"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,9,17]]},"references-count":29,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2024,9,17]]},"published-print":{"date-parts":[[2025,3,4]]}},"alternative-id":["10.1108\/IR-05-2024-0233"],"URL":"https:\/\/doi.org\/10.1108\/ir-05-2024-0233","relation":{},"ISSN":["0143-991X","1758-5791"],"issn-type":[{"type":"print","value":"0143-991X"},{"type":"electronic","value":"1758-5791"}],"subject":[],"published":{"date-parts":[[2024,9,17]]}}}