{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T09:08:58Z","timestamp":1771319338682,"version":"3.50.1"},"reference-count":49,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T00:00:00Z","timestamp":1771286400000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Scientific Research Fund of Zhejiang Provincial Education Department, China","award":["Y202456182"],"award-info":[{"award-number":["Y202456182"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>X-ray security inspection plays a vital role in public safety by enabling the automated detection of hazardous and prohibited items. However, real-world inspection scenarios often involve dense object stacking and severe occlusion, leading to frequent missed or false detections and limiting the effectiveness of existing detection models. To address these challenges, we propose a novel object detection framework based on You Only Look Once (YOLO)v8, termed YOLOv8-MS-DWConv. The model integrates a newly designed Multi-Scale Depthwise Separable Convolution (MS-DWConv) module to enhance feature extraction for small and occluded objects while preserving computational efficiency. Furthermore, we introduce a new localization loss function, Pixels-Intersection-over-Union (PIoU), as a replacement for the conventional complete IoU (CIoU) loss, which significantly improves bounding box regression performance, particularly in scenarios involving rotated or overlapping targets. By incorporating both MS-DWConv and PIoU into the YOLOv8 architecture, the proposed model achieves a lightweight yet robust detection capability tailored for complex X-ray security applications. Extensive experiments on three public benchmark datasets\u2014CLCXray, HIXray, and SIXray\u2014demonstrate the superior performance of our approach. Specifically, mAP@0.5:0.95 improved by 2.9% on CLCXray, by 1.2% on HIXray (from 50.0% to 51.2%), and by 4.5% on SIXray (from 61.5% to 66.0%). These results validate the model\u2019s enhanced robustness and generalization capability, particularly under heavy occlusion and cluttered conditions.<\/jats:p>","DOI":"10.7717\/peerj-cs.3593","type":"journal-article","created":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T08:16:47Z","timestamp":1771316207000},"page":"e3593","source":"Crossref","is-referenced-by-count":0,"title":["YOLOv8-MS-DWConv: enhancing hazardous goods detection in X-ray security imaging with multi-scale depthwise separable convolution"],"prefix":"10.7717","volume":"12","author":[{"given":"Qi","family":"Diao","sequence":"first","affiliation":[{"name":"Faculty of Artificial Intelligence, Zhejiang Dongfang Polytechnic, Wenzhou, Zhejiang, China"},{"name":"Faculty of Computing, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weng Howe","family":"Chan","sequence":"additional","affiliation":[{"name":"Faculty of Computing, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2004-3289","authenticated-orcid":true,"given":"Azlan Mohd","family":"Zain","sequence":"additional","affiliation":[{"name":"Faculty of Computing, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-5905-7047","authenticated-orcid":true,"given":"YiFan","family":"Chen","sequence":"additional","affiliation":[{"name":"Faculty of Computing, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia"},{"name":"Jiaxing Key Laboratory of Industrial Internet Security, Jiaxing Vocational and Technical College, Jiaxing, Zhejiang, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"4443","published-online":{"date-parts":[[2026,2,17]]},"reference":[{"issue":"3","key":"10.7717\/peerj-cs.3593\/ref-1","doi-asserted-by":"publisher","first-page":"523","DOI":"10.1117\/1.1760571","article-title":"Improving the detection of low-density weapons in X-ray luggage scans using image enhancement and novel scene-decluttering techniques","volume":"13","author":"Abidi","year":"2004","journal-title":"Journal of Electronic Imaging"},{"key":"10.7717\/peerj-cs.3593\/ref-2","first-page":"35","article-title":"Screener evaluation of pseudo-colored single energy X-ray luggage images","author":"Abidi","year":"2005"},{"issue":"7","key":"10.7717\/peerj-cs.3593\/ref-3","doi-asserted-by":"publisher","first-page":"1045","DOI":"10.1007\/s00138-015-0706-x","article-title":"Multi-view object detection in dual-energy X-ray images","volume":"26","author":"Ba\u015ftan","year":"2015","journal-title":"Machine Vision and Applications"},{"key":"10.7717\/peerj-cs.3593\/ref-4","first-page":"360","article-title":"Visual words on baggage X-ray images","author":"Ba\u015ftan","year":"2011"},{"key":"10.7717\/peerj-cs.3593\/ref-5","first-page":"195","article-title":"Piou loss: towards accurate oriented object detection in complex environments","author":"Chen","year":"2020"},{"issue":"4","key":"10.7717\/peerj-cs.3593\/ref-6","doi-asserted-by":"publisher","first-page":"541","DOI":"10.1007\/s11227-025-07036-w","article-title":"Multi-scale and dynamic snake convolution-based YOLOv9 for steel surface defect detection","volume":"81","author":"Chen","year":"2025","journal-title":"The Journal of Supercomputing"},{"key":"10.7717\/peerj-cs.3593\/ref-7","article-title":"A combinational approach to the fusion, de-noising and enhancement of dual-energy X-ray luggage images","author":"Chen","year":"2005"},{"key":"10.7717\/peerj-cs.3593\/ref-8","first-page":"1251","article-title":"Xception: deep learning with depthwise separable convolutions","author":"Chollet","year":"2017"},{"key":"10.7717\/peerj-cs.3593\/ref-9","first-page":"46","article-title":"Use of threat image projection (TIP) to enhance security performance","author":"Cutler","year":"2009"},{"key":"10.7717\/peerj-cs.3593\/ref-10","first-page":"144","article-title":"Object detection in multi-view X-ray images","author":"Franzel","year":"2012"},{"issue":"2","key":"10.7717\/peerj-cs.3593\/ref-11","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1007\/s11554-025-01665-3","article-title":"An improved YOLOv8\u00a0model for prohibited item detection with deformable convolution and dynamic head","volume":"22","author":"Guan","year":"2025","journal-title":"Journal of Real-Time Image Processing"},{"issue":"1","key":"10.7717\/peerj-cs.3593\/ref-12","doi-asserted-by":"publisher","first-page":"100","DOI":"10.2307\/2346830","article-title":"Algorithm as 136: a k-means clustering algorithm","volume":"28","author":"Hartigan","year":"1979","journal-title":"Journal of the Royal Statistical Society: Series C (Applied Statistics)"},{"key":"10.7717\/peerj-cs.3593\/ref-13","doi-asserted-by":"publisher","first-page":"42816","DOI":"10.1109\/access.2024.3378568","article-title":"YOLOv1 to v8: unveiling each variant\u2014a comprehensive review of YOLO","volume":"12","author":"Hussain","year":"2024","journal-title":"IEEE Access"},{"issue":"20","key":"10.7717\/peerj-cs.3593\/ref-14","doi-asserted-by":"publisher","first-page":"8555","DOI":"10.3390\/s23208555","article-title":"EM-YOLO: an X-ray prohibited-item-detection method based on edge and material information fusion","volume":"23","author":"Jing","year":"2023","journal-title":"Sensors"},{"issue":"1","key":"10.7717\/peerj-cs.3593\/ref-15","doi-asserted-by":"publisher","first-page":"204","DOI":"10.1186\/s13638-020-01826-x","article-title":"Object detection in real time based on improved single shot multi-box detector algorithm","volume":"2020","author":"Kumar","year":"2020","journal-title":"EURASIP Journal on Wireless Communications and Networking"},{"issue":"8","key":"10.7717\/peerj-cs.3593\/ref-16","doi-asserted-by":"publisher","first-page":"1265","DOI":"10.1109\/tpami.2005.151","article-title":"A sparse texture representation using local affine regions","volume":"27","author":"Lazebnik","year":"2005","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"issue":"14","key":"10.7717\/peerj-cs.3593\/ref-17","doi-asserted-by":"publisher","first-page":"15953","DOI":"10.1007\/s11227-022-04502-7","article-title":"A system for quantifying facial symmetry from 3D contour maps based on transfer learning and fast R-CNN","volume":"78","author":"Lin","year":"2022","journal-title":"The Journal of Supercomputing"},{"issue":"6","key":"10.7717\/peerj-cs.3593\/ref-18","doi-asserted-by":"publisher","first-page":"646","DOI":"10.1007\/s42514-024-00200-3","article-title":"Optimizing depthwise separable convolution on DCU","volume":"6","author":"Liu","year":"2024b","journal-title":"CCF Transactions on High Performance Computing"},{"key":"10.7717\/peerj-cs.3593\/ref-19","first-page":"480","article-title":"CAE-YOLOv8: occlusion object detection based on improved YOLOv8","author":"Liu","year":"2024a"},{"issue":"7","key":"10.7717\/peerj-cs.3593\/ref-20","doi-asserted-by":"publisher","first-page":"1105","DOI":"10.1049\/iet-ipr.2018.5198","article-title":"Graph clustering and variational image segmentation for automated firearm detection in X-ray images","volume":"13","author":"Mallia-Parfitt","year":"2019","journal-title":"IET Image Processing"},{"key":"10.7717\/peerj-cs.3593\/ref-21","first-page":"57","article-title":"A logarithmic X-ray imaging model for baggage inspection: simulation and object detection","author":"Mery","year":"2017"},{"issue":"1","key":"10.7717\/peerj-cs.3593\/ref-22","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1784\/insi.2012.55.1.16","article-title":"Detection of regular objects in baggage using multiple X-ray views","volume":"55","author":"Mery","year":"2013","journal-title":"Insight-Non-Destructive Testing and Condition Monitoring"},{"key":"10.7717\/peerj-cs.3593\/ref-23","first-page":"709","article-title":"Object recognition in baggage inspection using adaptive sparse representations of X-ray images","author":"Mery","year":"2016"},{"key":"10.7717\/peerj-cs.3593\/ref-24","article-title":"Threat image projection\u2014an overview","author":"Mitckes","year":"2003"},{"key":"10.7717\/peerj-cs.3593\/ref-25","article-title":"sixray_yolov8: an annotated X-ray security imagery dataset in YOLOv8 format","author":"MPU Project","year":"2024"},{"issue":"29","key":"10.7717\/peerj-cs.3593\/ref-26","doi-asserted-by":"publisher","first-page":"46","DOI":"10.5120\/ijca2018916708","article-title":"A review: object detection using deep learning","volume":"180","author":"Naik","year":"2018","journal-title":"International Journal of Computer Applications"},{"key":"10.7717\/peerj-cs.3593\/ref-27","first-page":"377","article-title":"Segmentation of handguns in dual energy X-ray imagery of passenger carry-on baggage","volume":"1","author":"Paranjape","year":"1998"},{"issue":"9","key":"10.7717\/peerj-cs.3593\/ref-28","doi-asserted-by":"publisher","first-page":"112681","DOI":"10.1016\/j.radphyschem.2025.112681","article-title":"Real-time CNN-based object detection of prohibited items for X-ray security screening","volume":"232","author":"Park","year":"2025","journal-title":"Radiation Physics and Chemistry"},{"key":"10.7717\/peerj-cs.3593\/ref-29","first-page":"457","article-title":"Assessment of object detection using deep convolutional neural networks","author":"Pathak","year":"2018"},{"issue":"7\u20138","key":"10.7717\/peerj-cs.3593\/ref-30","doi-asserted-by":"publisher","first-page":"108283","DOI":"10.1016\/j.compeleceng.2022.108283","article-title":"Lightray: lightweight network for prohibited items detection in X-ray images during security inspection","volume":"103","author":"Ren","year":"2022","journal-title":"Computers and Electrical Engineering"},{"key":"10.7717\/peerj-cs.3593\/ref-31","first-page":"1","article-title":"Threat image projection (tip) into X-ray images of cargo containers for training humans and machines","author":"Rogers","year":"2016"},{"key":"10.7717\/peerj-cs.3593\/ref-32","first-page":"2573","article-title":"Visual cortex inspired features for object detection in X-ray images","author":"Schmidt-Hackenberg","year":"2012"},{"key":"10.7717\/peerj-cs.3593\/ref-33","first-page":"720","article-title":"Model-based probabilistic relaxation segmentation applied to threat detection in airport X-ray imagery","volume":"2","author":"Sluser","year":"1999"},{"key":"10.7717\/peerj-cs.3593\/ref-34","first-page":"1","article-title":"Going deeper with convolutions","author":"Szegedy","year":"2015"},{"key":"10.7717\/peerj-cs.3593\/ref-35","first-page":"2820","article-title":"Mnasnet: platform-aware neural architecture search for mobile","author":"Tan","year":"2019"},{"key":"10.7717\/peerj-cs.3593\/ref-36","first-page":"10923","article-title":"Towards real-world X-ray security inspection: a high-quality benchmark and lateral inhibition module for prohibited items detection","author":"Tao","year":"2021"},{"key":"10.7717\/peerj-cs.3593\/ref-37","article-title":"Yolov8","author":"Ultralytics","year":"2023"},{"key":"10.7717\/peerj-cs.3593\/ref-38","first-page":"1206","article-title":"Structural X-ray image segmentation for threat detection by attribute relational graph matching","volume":"2","author":"Wang","year":"2005"},{"issue":"5","key":"10.7717\/peerj-cs.3593\/ref-39","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1016\/j.jvcir.2019.01.017","article-title":"Research on automatic target detection and recognition based on deep learning","volume":"60","author":"Wang","year":"2019","journal-title":"Journal of Visual Communication and Image Representation"},{"issue":"16","key":"10.7717\/peerj-cs.3593\/ref-40","doi-asserted-by":"publisher","first-page":"3238","DOI":"10.3390\/electronics13163238","article-title":"Improved YOLOv8 for dangerous goods detection in X-ray security images","volume":"13","author":"Wang","year":"2024","journal-title":"Electronics"},{"key":"10.7717\/peerj-cs.3593\/ref-41","doi-asserted-by":"publisher","first-page":"45415","DOI":"10.1109\/access.2023.3273736","article-title":"Object detection and X-ray security imaging: a survey","volume":"11","author":"Wu","year":"2023","journal-title":"IEEE Access"},{"issue":"7","key":"10.7717\/peerj-cs.3593\/ref-42","doi-asserted-by":"publisher","first-page":"2420","DOI":"10.1007\/s11263-024-01989-w","article-title":"Oriented R-CNN and beyond","volume":"132","author":"Xie","year":"2024","journal-title":"International Journal of Computer Vision"},{"issue":"4","key":"10.7717\/peerj-cs.3593\/ref-43","doi-asserted-by":"publisher","first-page":"53","DOI":"10.53106\/199115992023083404005","article-title":"X-ray image prohibited item detection algorithm based on improved PP-YOLO","volume":"34","author":"Zhang","year":"2023","journal-title":"Journal of Computer Education"},{"issue":"1","key":"10.7717\/peerj-cs.3593\/ref-44","doi-asserted-by":"publisher","first-page":"107421","DOI":"10.1016\/j.jfranklin.2024.107421","article-title":"Lightweight prohibited items detection model in X-ray images based on improved YOLOv7-tiny","volume":"362","author":"Zhang","year":"2025","journal-title":"Journal of the Franklin Institute"},{"key":"10.7717\/peerj-cs.3593\/ref-45","first-page":"266","article-title":"Joint shape and texture based X-ray cargo image classification","author":"Zhang","year":"2014"},{"issue":"1","key":"10.7717\/peerj-cs.3593\/ref-46","doi-asserted-by":"publisher","first-page":"13171","DOI":"10.1038\/s41598-025-96035-1","article-title":"A lightweight x-ray-YOLO-Mamba model for prohibited item detection in X-ray images using selective state space models","volume":"15","author":"Zhao","year":"2025","journal-title":"Scientific Reports"},{"key":"10.7717\/peerj-cs.3593\/ref-47","doi-asserted-by":"publisher","first-page":"998","DOI":"10.1109\/tifs.2022.3154287","article-title":"Detecting overlapped objects in X-ray security imagery by a label-aware mechanism","volume":"17","author":"Zhao","year":"2022","journal-title":"IEEE Transactions on Information Forensics and Security"},{"key":"10.7717\/peerj-cs.3593\/ref-48","first-page":"546","article-title":"X-ray security inspection image detection algorithm based on improved YOLOv4","author":"Zhou","year":"2021"},{"key":"10.7717\/peerj-cs.3593\/ref-49","first-page":"27","article-title":"Amod-net: attention-based multi-scale object detection network for X-ray baggage security inspection","author":"Zhu","year":"2021"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-3593.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3593.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3593.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-3593.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,17]],"date-time":"2026-02-17T08:16:53Z","timestamp":1771316213000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-3593"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,17]]},"references-count":49,"alternative-id":["10.7717\/peerj-cs.3593"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.3593","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,17]]},"article-number":"e3593"}}