{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T05:42:00Z","timestamp":1761198120091,"version":"3.41.2"},"reference-count":53,"publisher":"Emerald","issue":"2","license":[{"start":{"date-parts":[[2022,9,26]],"date-time":"2022-09-26T00:00:00Z","timestamp":1664150400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.emerald.com\/insight\/site-policies"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IJICC"],"published-print":{"date-parts":[[2023,5,15]]},"abstract":"<jats:sec><jats:title content-type=\"abstract-subheading\">Purpose<\/jats:title><jats:p>The purpose of the study is to address the problems of low accuracy and missed detection of occluded pedestrians and small target pedestrians when using the YOLOX general object detection algorithm for pedestrian detection. This study proposes a multi-level fine-grained YOLOX pedestrian detection algorithm.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Design\/methodology\/approach<\/jats:title><jats:p>First, to address the problem of the original YOLOX algorithm in obtaining a single perceptual field for the feature map before feature fusion, this study improves the PAFPN structure by adding the ResCoT module to increase the diversity of the perceptual field of the feature map and divides the pedestrian multi-scale features into finer granularity. Second, for the CSPLayer of the PAFPN, a weight gain-based normalization-based attention module (NAM) is proposed to make the model pay more attention to the context information when extracting pedestrian features and highlight the salient features of pedestrians. Finally, the authors experimentally determined the optimal values for the confidence loss function.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Findings<\/jats:title><jats:p>The experimental results show that, compared with the original YOLOX algorithm, the AP of the improved algorithm increased by 2.90%, the Recall increased by 3.57%, and F1 increased by 2% on the pedestrian dataset.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Research limitations\/implications<\/jats:title><jats:p>The multi-level fine-grained YOLOX pedestrian detection algorithm can effectively improve the detection of occluded pedestrians and small target pedestrians.<\/jats:p><\/jats:sec><jats:sec><jats:title content-type=\"abstract-subheading\">Originality\/value<\/jats:title><jats:p>The authors introduce a multi-level fine-grained ResCoT module and a weight gain-based NAM attention module.<\/jats:p><\/jats:sec>","DOI":"10.1108\/ijicc-05-2022-0161","type":"journal-article","created":{"date-parts":[[2022,9,24]],"date-time":"2022-09-24T00:13:42Z","timestamp":1663978422000},"page":"295-313","source":"Crossref","is-referenced-by-count":6,"title":["Research on pedestrian detection based on multi-level fine-grained YOLOX algorithm"],"prefix":"10.1108","volume":"16","author":[{"given":"Hong","family":"Wang","sequence":"first","affiliation":[]},{"given":"Yong","family":"Xie","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5837-330X","authenticated-orcid":false,"given":"Shasha","family":"Tian","sequence":"additional","affiliation":[]},{"given":"Lu","family":"Zheng","sequence":"additional","affiliation":[]},{"given":"Xiaojie","family":"Dong","sequence":"additional","affiliation":[]},{"given":"Yu","family":"Zhu","sequence":"additional","affiliation":[]}],"member":"140","published-online":{"date-parts":[[2022,9,26]]},"reference":[{"journal-title":"arXiv preprint arXiv:2004.10934","article-title":"Yolov4: optimal speed and accuracy of object detection","year":"2020","key":"key2023071410503910300_ref001"},{"journal-title":"arXiv preprint arXiv:2004.03580","article-title":"Feature pyramid grids","year":"2020","key":"key2023071410503910300_ref002"},{"first-page":"12214","article-title":"Detection in crowded scenes: one proposal, multiple predictions","year":"2020","key":"key2023071410503910300_ref003"},{"first-page":"399","article-title":"Corner proposal network for anchor-free, two-stage object detection","year":"2020","key":"key2023071410503910300_ref004"},{"journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence","article-title":"Res2net: a new multi-scale backbone architecture","year":"2019","key":"key2023071410503910300_ref005"},{"journal-title":"arXiv preprint arXiv:2107.08430","article-title":"Yolox: exceeding yolo series in 2021","year":"2021","key":"key2023071410503910300_ref006"},{"first-page":"1160","article-title":"Effective fusion factor in FPN for tiny object detection","year":"2021","key":"key2023071410503910300_ref007"},{"first-page":"12595","article-title":"Augfpn: improving multi-scale feature learning for object detection","year":"2020","key":"key2023071410503910300_ref008"},{"first-page":"11328","article-title":"Generalizable pedestrian detection: the elephant in the room","year":"2021","key":"key2023071410503910300_ref009"},{"first-page":"770","article-title":"Deep residual learning for image recognition","year":"2016","key":"key2023071410503910300_ref010"},{"first-page":"13713","article-title":"Coordinate attention for efficient mobile network design","year":"2021","key":"key2023071410503910300_ref011"},{"first-page":"7132","article-title":"Squeeze-and-excitation networks","year":"2018","key":"key2023071410503910300_ref012"},{"first-page":"10750","article-title":"Nms by representative region: towards crowded pedestrian detection by proposal pairing","year":"2020","key":"key2023071410503910300_ref013"},{"issue":"1","key":"key2023071410503910300_ref014","doi-asserted-by":"publisher","first-page":"55","DOI":"10.19713\/j.cnki.43-1423\/u.T20200236","article-title":"Pedestrian detection method for station based on improved YOLOv3","volume":"18","year":"2021","journal-title":"Journal of Railway Science and Engineering"},{"issue":"8","key":"key2023071410503910300_ref015","doi-asserted-by":"publisher","DOI":"10.11959\/j.issn.1000\u22120801.2021198","article-title":"An improved YOLOv4 algorithm for pedestrian detection in complex visual scenes","year":"2021","journal-title":"Telecommunications Science"},{"issue":"8","key":"key2023071410503910300_ref016","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3778\/j.issn.1002-8331.2001-0163","article-title":"Survey of vision based object detection methods","volume":"56","year":"2020","journal-title":"Computer Engineering and Applications"},{"key":"key2023071410503910300_ref017","first-page":"21002","article-title":"Generalized focal loss: learning qualified and distributed bounding boxes for dense object detection","volume":"33","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"first-page":"2378","article-title":"Pseudo-iou: improving label assignment in anchor-free object detection","year":"2021","key":"key2023071410503910300_ref018"},{"journal-title":"arXiv preprint arXiv:2107.12292","article-title":"Contextual transformer networks for visual recognition","year":"2021","key":"key2023071410503910300_ref019"},{"first-page":"2117","article-title":"Feature pyramid networks for object detection","year":"2017","key":"key2023071410503910300_ref020"},{"first-page":"2980","article-title":"Focal loss for dense object detection","year":"2017","key":"key2023071410503910300_ref021"},{"first-page":"21","article-title":"Ssd: single shot multibox detector","year":"2016","key":"key2023071410503910300_ref022"},{"first-page":"618","article-title":"Learning efficient single-stage pedestrian detectors by asymptotic localization fitting","year":"2018","key":"key2023071410503910300_ref023"},{"first-page":"8759","article-title":"Path aggregation network for instance segmentation","year":"2018","key":"key2023071410503910300_ref024"},{"first-page":"1058","article-title":"Analysis of anchor-based and anchor-free object detection methods based on deep learning","year":"2020","key":"key2023071410503910300_ref025"},{"journal-title":"arXiv preprint arXiv:2111.12419","article-title":"NAM: normalization-based attention module","year":"2021","key":"key2023071410503910300_ref026"},{"first-page":"14065","article-title":"Where, What, Whether: multi-modal learning meets pedestrian detection","year":"2020","key":"key2023071410503910300_ref027"},{"journal-title":"arXiv preprint arXiv:2103.10643","article-title":"CE-FPN: enhancing channel information for object detection","year":"2021","key":"key2023071410503910300_ref028"},{"journal-title":"arXiv preprint arXiv:1807.06514","article-title":"Bam: bottleneck attention module","year":"2018","key":"key2023071410503910300_ref029"},{"first-page":"7263","article-title":"YOLO9000: better, faster, stronger","year":"2017","key":"key2023071410503910300_ref030"},{"journal-title":"arXiv preprint arXiv:1804.02767","article-title":"Yolov3: an incremental improvement","year":"2018","key":"key2023071410503910300_ref031"},{"first-page":"779","article-title":"You only look once: unified, real-time object detection","year":"2016","key":"key2023071410503910300_ref032"},{"journal-title":"arXiv. preprint arXiv:2104.03106","article-title":"V2F-Net: explicit decomposition of occluded pedestrian detection","year":"2021","key":"key2023071410503910300_ref033"},{"issue":"3","key":"key2023071410503910300_ref034","doi-asserted-by":"publisher","DOI":"10.14177\/j.cnki.32-1397n.2021.45.03.001","article-title":"Improved pedestrian detection algorithm based on YOLOv3","year":"2021","journal-title":"Journal of Nanjing University of Science and Technology"},{"first-page":"536","article-title":"Small-scale pedestrian detection based on topological line localization and temporal feature aggregation","year":"2018","key":"key2023071410503910300_ref035"},{"first-page":"10781","article-title":"Efficientdet: scalable and efficient object detection","year":"2020","key":"key2023071410503910300_ref036"},{"issue":"12","key":"key2023071410503910300_ref037","doi-asserted-by":"crossref","first-page":"9066","DOI":"10.1007\/s10489-021-02373-8","article-title":"Coordinate-based anchor-free module for object detection","volume":"51","year":"2021","journal-title":"Applied Intelligence"},{"first-page":"9627","article-title":"Fcos: fully convolutional one-stage object detection","year":"2019","key":"key2023071410503910300_ref038"},{"key":"key2023071410503910300_ref039","article-title":"Attention is all you need","volume":"30","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"first-page":"7774","article-title":"Repulsion loss: detecting pedestrians in a crowd","year":"2018","key":"key2023071410503910300_ref040"},{"first-page":"15849","article-title":"End-to-end object detection with fully convolutional network","year":"2021","key":"key2023071410503910300_ref041"},{"first-page":"13029","article-title":"Scaled-yolov4: scaling cross stage partial network","year":"2021","key":"key2023071410503910300_ref042"},{"first-page":"3","article-title":"Cbam: convolutional block attention module","year":"2018","key":"key2023071410503910300_ref043"},{"first-page":"13430","article-title":"Temporal-context enhanced detection of heavily occluded pedestrians","year":"2020","key":"key2023071410503910300_ref044"},{"issue":"2","key":"key2023071410503910300_ref045","doi-asserted-by":"publisher","first-page":"418","DOI":"10.13873\/J.1000-9787(2021)02-0004-04","article-title":"Review of target Detection algorithms based on deep learning","volume":"40","year":"2021","journal-title":"Transducer and Microsystem Technologies"},{"first-page":"637","article-title":"Occlusion-aware R-CNN: detecting pedestrians in a crowd","year":"2018","key":"key2023071410503910300_ref046"},{"issue":"8","key":"key2023071410503910300_ref047","doi-asserted-by":"crossref","first-page":"3703","DOI":"10.1109\/TIP.2018.2818018","article-title":"Too far to see? Not really!\u2014pedestrian detection with scale-aware. localization policy","volume":"27","year":"2018","journal-title":"IEEE Transactions on Image Processing"},{"first-page":"9759","article-title":"Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection","year":"2020","key":"key2023071410503910300_ref048"},{"first-page":"11622","article-title":"Variational pedestrian detection","year":"2021","key":"key2023071410503910300_ref049"},{"first-page":"8514","article-title":"Varifocalnet: an iou-aware dense object detector","year":"2021","key":"key2023071410503910300_ref050"},{"first-page":"135","article-title":"Bi-box regression for pedestrian detection and occlusion estimation","year":"2018","key":"key2023071410503910300_ref051"},{"key":"key2023071410503910300_ref052","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1016\/j.patrec.2019.12.010","article-title":"Attention guided neural network models for occluded pedestrian detection","volume":"131","year":"2020","journal-title":"Pattern Recognition Letters"},{"issue":"2","key":"key2023071410503910300_ref053","doi-asserted-by":"publisher","first-page":"295","DOI":"10.13195\/j.kzyjc.2020.0124","article-title":"Anchor-free scale adaptive pedestrian detection algorithm","volume":"36","year":"2021","journal-title":"Control and Decision"}],"container-title":["International Journal of Intelligent Computing and Cybernetics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJICC-05-2022-0161\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/insight\/content\/doi\/10.1108\/IJICC-05-2022-0161\/full\/html","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,24]],"date-time":"2025-07-24T22:54:20Z","timestamp":1753397660000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.emerald.com\/ijicc\/article\/16\/2\/295-313\/122635"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,9,26]]},"references-count":53,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2022,9,26]]},"published-print":{"date-parts":[[2023,5,15]]}},"alternative-id":["10.1108\/IJICC-05-2022-0161"],"URL":"https:\/\/doi.org\/10.1108\/ijicc-05-2022-0161","relation":{},"ISSN":["1756-378X"],"issn-type":[{"type":"print","value":"1756-378X"}],"subject":[],"published":{"date-parts":[[2022,9,26]]}}}