{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T15:54:04Z","timestamp":1774540444042,"version":"3.50.1"},"reference-count":31,"publisher":"Emerald","issue":"4","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,11,11]]},"abstract":"<jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>The abnormal behaviors of staff at petroleum stations pose significant safety hazards. Addressing the challenges of high parameter counts, lengthy training periods and low recognition rates in existing 3D ResNet behavior recognition models, this paper proposes GTB-ResNet, a network designed to detect abnormal behaviors in petroleum station staff.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Design\/methodology\/approach<\/jats:title>\n                    <jats:p>Firstly, to mitigate the issues of excessive parameters and computational complexity in 3D ResNet, a lightweight residual convolution module called the Ghost residual module (GhostNet) is introduced in the feature extraction network. Ghost convolution replaces standard convolution, reducing model parameters while preserving multi-scale feature extraction capabilities. Secondly, to enhance the model's focus on salient features amidst wide surveillance ranges and small target objects, the triplet attention mechanism module is integrated to facilitate spatial and channel information interaction. Lastly, to address the challenge of short time-series features leading to misjudgments in similar actions, a bidirectional gated recurrent network is added to the feature extraction backbone network. This ensures the extraction of key long time-series features, thereby improving feature extraction accuracy.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Findings<\/jats:title>\n                    <jats:p>The experimental setup encompasses four behavior types: illegal phone answering, smoking, falling (abnormal) and touching the face (normal), comprising a total of 892 videos. Experimental results showcase GTB-ResNet achieving a recognition accuracy of 96.7% with a model parameter count of 4.46\u00a0M and a computational complexity of 3.898\u00a0G. This represents a 4.4% improvement over 3D ResNet, with reductions of 90.4% in parameters and 61.5% in computational complexity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Originality\/value<\/jats:title>\n                    <jats:p>Specifically designed for edge devices in oil stations, the 3D ResNet network is tailored for real-time action prediction. To address the challenges posed by the large number of parameters in 3D ResNet networks and the difficulties in deployment on edge devices, a lightweight residual module based on ghost convolution is developed. Additionally, to tackle the issue of low detection accuracy of behaviors amidst the noisy environment of petroleum stations, a triple attention mechanism is introduced during feature extraction to enhance focus on salient features. Moreover, to overcome the potential for misjudgments arising from the similarity of actions, a Bi-GRU model is introduced to enhance the extraction of key long-term features.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1108\/ijicc-07-2024-0317","type":"journal-article","created":{"date-parts":[[2024,10,29]],"date-time":"2024-10-29T16:26:49Z","timestamp":1730219209000},"page":"869-889","source":"Crossref","is-referenced-by-count":2,"title":["A\u00a0method for recognizing abnormal\u00a0behaviors of personnel at\u00a0petroleum stations based on\u00a0GTB-ResNet"],"prefix":"10.1108","volume":"17","author":[{"given":"Huiling","family":"Yu","sequence":"first","affiliation":[{"name":"Changzhou University , Changzhou,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-9720-0039","authenticated-orcid":false,"given":"Sijia","family":"Dai","sequence":"additional","affiliation":[{"name":"School of Computer Science and Artificial Intelligence , , Changzhou,","place":["China"]},{"name":"Changzhou University , , Changzhou,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shen","family":"Shi","sequence":"additional","affiliation":[{"name":"Changzhou University , Changzhou,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-2146-6044","authenticated-orcid":false,"given":"Yizhuo","family":"Zhang","sequence":"additional","affiliation":[{"name":"Changzhou University , Changzhou,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"140","published-online":{"date-parts":[[2024,10,29]]},"reference":[{"key":"2025102222552059200_ref001","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.120519","article-title":"Consistency-and dependence-guided knowledge distillation for object detection in remote sensing images","volume":"229","author":"Chen","year":"2023","journal-title":"Expert Systems with Applications"},{"key":"2025102222552059200_ref002","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/tifs.2024.3424303","article-title":"FIBNet: privacy-Enhancing approach for face biometrics based on the information bottleneck principle","volume":"19","author":"Chen","year":"2024","journal-title":"IEEE Transactions on Information Forensics and Security"},{"issue":"5","key":"2025102222552059200_ref003","doi-asserted-by":"publisher","DOI":"10.3390\/s21051656","article-title":"AR3D: attention residual 3D network for human action recognition","volume":"21","author":"Dong","year":"2021","journal-title":"Sensors"},{"key":"2025102222552059200_ref004","first-page":"203","article-title":"X3d: expanding architectures for efficient video recognition","author":"Feichtenhofer","year":"2020"},{"key":"2025102222552059200_ref032","first-page":"1580","author":"Han","year":"2020","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"2025102222552059200_ref005","first-page":"6546","article-title":"Can spatiotemporal 3D CNNs retrace the history of 2D CNNs and ImageNet?","author":"Hara","year":"2018"},{"key":"2025102222552059200_ref006","doi-asserted-by":"publisher","first-page":"1093","DOI":"10.1016\/j.ins.2022.06.091","article-title":"Deconv-transformer (DecT): a histopathological image classification model for breast cancer based on color deconvolution and transformer architecture","volume":"608","author":"He","year":"2022","journal-title":"Information Sciences"},{"issue":"1","key":"2025102222552059200_ref007","doi-asserted-by":"publisher","first-page":"221","DOI":"10.1109\/tpami.2012.59","article-title":"3D convolutional neural networks for human action recognition","volume":"35","author":"Ji","year":"2012","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"2025102222552059200_ref008","first-page":"3259","article-title":"Predicting human behavior using 3D loop ResNet","author":"Kakamu","year":"2022"},{"key":"2025102222552059200_ref009","article-title":"Would mega-scale datasets further enhance spatiotemporal 3D CNNs?","author":"Kataoka","year":"2020"},{"key":"2025102222552059200_ref010","first-page":"1910","article-title":"Resource efficient 3d convolutional neural networks","author":"Kopuklu","year":"2019"},{"key":"2025102222552059200_ref011","first-page":"4903","article-title":"Lp-3dcnn: unveiling local phase in 3d convolutional neural networks","author":"Kumawat","year":"2019"},{"key":"2025102222552059200_ref012","doi-asserted-by":"publisher","first-page":"646","DOI":"10.1109\/icdmw.2019.00098","article-title":"Dynamic spatio-temporal feature learning via graph convolution in 3D convolutional networks","author":"Li","year":"2019"},{"issue":"19","key":"2025102222552059200_ref013","doi-asserted-by":"publisher","DOI":"10.3390\/app131910700","article-title":"Faster R-CNN-LSTM construction site unsafe behavior recognition model","volume":"13","author":"Li","year":"2023","journal-title":"Applied Sciences"},{"key":"2025102222552059200_ref014","first-page":"7083","article-title":"TSM: temporal shift module for efficient video understanding","author":"Lin","year":"2019"},{"issue":"3","key":"2025102222552059200_ref015","doi-asserted-by":"publisher","first-page":"318","DOI":"10.1108\/ijicc-12-2018-0167","article-title":"Self-adaptive scale pedestrian detection algorithm based on deep residual network","volume":"12","author":"Liu","year":"2019","journal-title":"International Journal of Intelligent Computing and Cybernetics"},{"key":"2025102222552059200_ref016","doi-asserted-by":"publisher","first-page":"66797","DOI":"10.1109\/access.2022.3185112","article-title":"A multichannel CNN-GRU model for human activity recognition","volume":"10","author":"Lu","year":"2022","journal-title":"IEEE Access"},{"issue":"6","key":"2025102222552059200_ref017","doi-asserted-by":"publisher","first-page":"995","DOI":"10.1016\/j.eng.2019.08.013","article-title":"Opportunities and challenges of artificial intelligence for green manufacturing in the process industry","volume":"5","author":"Mao","year":"2019","journal-title":"Engineering"},{"key":"2025102222552059200_ref018","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1016\/j.ijinfomgt.2019.05.020","article-title":"Machine learning based digital twin framework for production optimization in petrochemical industry","volume":"49","author":"Min","year":"2019","journal-title":"International Journal of Information Management"},{"key":"2025102222552059200_ref019","first-page":"3139","article-title":"Rotate to attend: convolutional triplet attention module","author":"Misra","year":"2021"},{"key":"2025102222552059200_ref020","first-page":"5","article-title":"Study on 3D action recognition based on deep neural network","author":"Park","year":"2019"},{"key":"2025102222552059200_ref021","doi-asserted-by":"publisher","DOI":"10.1016\/j.apergo.2022.103697","article-title":"Validating an abnormal situation prediction model for smart manufacturing in the oil refining industry","volume":"101","author":"Shi","year":"2022","journal-title":"Applied Ergonomics"},{"key":"2025102222552059200_ref022","first-page":"568","article-title":"Two-stream convolutional networks for action recognition in videos","volume":"27","author":"Simonyan","year":"2014","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2025102222552059200_ref023","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/icecie47765.2019.8974824","article-title":"Abnormal behavior recognition using CNN-LSTM with attention mechanism","author":"Tay","year":"2019"},{"key":"2025102222552059200_ref024","first-page":"4489","article-title":"Learning spatiotemporal features with 3D convolutional networks","author":"Tran","year":"2015"},{"key":"2025102222552059200_ref025","first-page":"6450","article-title":"A closer look at spatiotemporal convolutions for action recognition","author":"Tran","year":"2018"},{"issue":"6","key":"2025102222552059200_ref026","doi-asserted-by":"publisher","first-page":"1510","DOI":"10.1109\/tpami.2017.2712608","article-title":"Long-term temporal convolutions for action recognition","volume":"40","author":"Varol","year":"2017","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"2025102222552059200_ref027","first-page":"7794","article-title":"Non-local neural networks","author":"Wang","year":"2018"},{"issue":"4","key":"2025102222552059200_ref028","doi-asserted-by":"publisher","first-page":"655","DOI":"10.1108\/aa-06-2019-0118","article-title":"3D long-term recurrent convolutional networks for human sub-assembly recognition in human-robot collaboration","volume":"40","author":"Wen","year":"2020","journal-title":"Assembly Automation"},{"key":"2025102222552059200_ref029","first-page":"284","article-title":"Long-term feature banks for detailed video understanding","author":"Wu","year":"2019"},{"key":"2025102222552059200_ref030","doi-asserted-by":"publisher","DOI":"10.1016\/j.imavis.2021.104108","article-title":"Motion saliency based multi-stream multiplier ResNets for action recognition","volume":"107","author":"Zong","year":"2021","journal-title":"Image and Vision Computing"}],"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-07-2024-0317\/full\/xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.emerald.com\/ijicc\/article-pdf\/17\/4\/869\/9479812\/ijicc-07-2024-0317.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/www.emerald.com\/ijicc\/article-pdf\/17\/4\/869\/9479812\/ijicc-07-2024-0317.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T02:55:27Z","timestamp":1761188127000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.emerald.com\/ijicc\/article\/17\/4\/869\/1216378\/A-method-for-recognizing-abnormal-behaviors-of"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,29]]},"references-count":31,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2024,11,11]]}},"URL":"https:\/\/doi.org\/10.1108\/ijicc-07-2024-0317","relation":{},"ISSN":["1756-378X","1756-3798"],"issn-type":[{"value":"1756-378X","type":"print"},{"value":"1756-3798","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,10,29]]}}}