{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,24]],"date-time":"2026-10-24T14:10:52Z","timestamp":1792851052943,"version":"4.5.0"},"reference-count":45,"publisher":"PeerJ","license":[{"start":{"date-parts":[[2025,4,30]],"date-time":"2025-04-30T00:00:00Z","timestamp":1745971200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"XiJing University Research Fund","award":["XJ240205"],"award-info":[{"award-number":["XJ240205"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>In the intelligent transformation of education, accurate recognition of students\u2019 classroom behavior has become one of the key technologies for enhancing the quality of instruction and the efficacy of learning. However, in the recognition of target behavior in real classroom scenarios, due to the use of wide-angle or panoramic images for image acquisition, students in the back row are far away from monitoring devices, and their subtle body movements such as the small opening and closing of the mouth (to determine whether they are speaking), fine finger operations (to distinguish between reading books or operating mobile phones) are difficult to recognize. Moreover, there are occlusions and scale differences in the front and back rankings, which can easily cause confusion and interference with target features in the detection process, greatly limiting the accurate recognition ability of existing visual algorithms for classroom behavior. This article proposes a deformable multiscale adaptive classroom behavior recognition network. To improve the network\u2019s capacity to model minute behavioral phenomena, the backbone section introduces a deformable self-attention dattention module, dynamically modifying the receptive field\u2019s geometry to enhance the model\u2019s concentration on the region of interest. To improve the network\u2019s capacity for feature extraction and integration of behavior occlusion and classroom behavior at different scales, a proposal has been put forward the Multiscale Attention Feature Pyramid Structure (MSAFPS), to achieve multi-level feature aggregation after multiscale feature fusion, reducing the impact of mutual occlusion and scale differences in classroom behavior between front and back rows. In the detect section, we adopt the Wise Intersection Over Union (Wise-IoU) loss as our loss criterion, augmenting the evaluation framework with richer contextual cues to broaden its scope and elevate the network\u2019s detection prowess. Extensive experimentation reveals that our proposed method outperforms rival algorithms on two widely adopted benchmark datasets: SCB-Dataset3-S (the Student Classroom Behavior Dataset\u2013<jats:uri xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/Whiffe\/SCB-dataset\">https:\/\/github.com\/Whiffe\/SCB-dataset<\/jats:uri>) and we created object detection dataset DataMountainSCB (<jats:uri xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/Chunyu-Dong\/DataFountainSCB1\">https:\/\/github.com\/Chunyu-Dong\/DataFountainSCB1<\/jats:uri>) containing six types of behaviors.<\/jats:p>","DOI":"10.7717\/peerj-cs.2876","type":"journal-article","created":{"date-parts":[[2025,4,30]],"date-time":"2025-04-30T08:57:14Z","timestamp":1746003434000},"page":"e2876","source":"Crossref","is-referenced-by-count":7,"title":["DMSA-Net: a deformable multiscale adaptive classroom behavior recognition network"],"prefix":"10.7717","volume":"11","author":[{"given":"Chunyu","family":"Dong","sequence":"first","affiliation":[{"name":"School of Computing, Xijing University, Xi\u2019an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jing","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computing, Xijing University, Xi\u2019an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shenglong","family":"Xie","sequence":"additional","affiliation":[{"name":"School of Computer Science Technology, Xidian University, Xi\u2019an, Shaanxi, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"4443","published-online":{"date-parts":[[2025,4,30]]},"reference":[{"key":"10.7717\/peerj-cs.2876\/ref-1","doi-asserted-by":"publisher","first-page":"951","DOI":"10.1016\/j.procs.2021.08.098","article-title":"Student behavior recognition in classroom using deep transfer learning with VGG-16","volume":"192","author":"Abdallah","year":"2021","journal-title":"Procedia Computer Science"},{"issue":"7","key":"10.7717\/peerj-cs.2876\/ref-2","doi-asserted-by":"publisher","first-page":"3818","DOI":"10.1016\/j.procs.2022.09.443","article-title":"Discovering the effects of learning analytics dashboard on students\u2019 behavioral patterns using differential sequence mining","volume":"207","author":"Ak\u00e7ap\u0131nar","year":"2022","journal-title":"Procedia Computer Science"},{"issue":"2","key":"10.7717\/peerj-cs.2876\/ref-3","doi-asserted-by":"publisher","first-page":"282","DOI":"10.18608\/jla.2016.32.14","article-title":"Using multimodal learning analytics to model student behavior: a systematic analysis of epistemological framing","volume":"3","author":"Andrade","year":"2016","journal-title":"Journal of Learning Analytics"},{"key":"10.7717\/peerj-cs.2876\/ref-4","article-title":"Yolov4: optimal speed and accuracy of object detection","author":"Bochkovskiy","year":"2020"},{"key":"10.7717\/peerj-cs.2876\/ref-5","first-page":"6154","article-title":"Cascade R-CNN: delving into high quality object detection","author":"Cai","year":"2018"},{"key":"10.7717\/peerj-cs.2876\/ref-6","first-page":"6299","article-title":"Quo vadis, action recognition? a new model and the kinetics dataset","author":"Carreira","year":"2017"},{"issue":"20","key":"10.7717\/peerj-cs.2876\/ref-7","doi-asserted-by":"publisher","first-page":"8385","DOI":"10.3390\/s23208385","article-title":"Student behavior detection in the classroom based on improved YOLOv8","volume":"23","author":"Chen","year":"2023","journal-title":"Sensors"},{"key":"10.7717\/peerj-cs.2876\/ref-8","article-title":"How to train an ultralytics YOLOv8 oriented bounding box (OBB) model","author":"Gallagher","year":"2024"},{"key":"10.7717\/peerj-cs.2876\/ref-9","first-page":"42","article-title":"Automatic academic emotion recognition and its evolution analysis in classroom","author":"Gao","year":"2023"},{"key":"10.7717\/peerj-cs.2876\/ref-10","first-page":"1440","article-title":"Fast R-CNN","author":"Girshick","year":"2015"},{"key":"10.7717\/peerj-cs.2876\/ref-11","first-page":"580","article-title":"Rich feature hierarchies for accurate object detection and semantic segmentation","author":"Girshick","year":"2014"},{"key":"10.7717\/peerj-cs.2876\/ref-12","first-page":"1","article-title":"Human activity recognition using combinatorial deep belief networks","author":"Gowda","year":"2017"},{"issue":"8","key":"10.7717\/peerj-cs.2876\/ref-13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3472290","article-title":"A survey on deep learning for human activity recognition","volume":"54","author":"Gu","year":"2021","journal-title":"ACM Computing Surveys (CSUR)"},{"key":"10.7717\/peerj-cs.2876\/ref-14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.00752","article-title":"Mamba: linear-time sequence modeling with selective state spaces","author":"Gu","year":"2023"},{"issue":"5","key":"10.7717\/peerj-cs.2876\/ref-15","doi-asserted-by":"publisher","first-page":"823","DOI":"10.18280\/ts.370515","article-title":"Detection of head raising rate of students in classroom based on head posture recognition","volume":"37","author":"Guo","year":"2020","journal-title":"Traitement du Signal"},{"issue":"16","key":"10.7717\/peerj-cs.2876\/ref-16","doi-asserted-by":"publisher","first-page":"5932","DOI":"10.3390\/s22165932","article-title":"Bimodal learning engagement recognition from videos in the classroom","volume":"22","author":"Hu","year":"2022","journal-title":"Sensors"},{"issue":"2","key":"10.7717\/peerj-cs.2876\/ref-17","doi-asserted-by":"publisher","first-page":"133","DOI":"10.6180\/jase.202104_24(2).0002","article-title":"A new feature fusion network for student behavior recognition in education","volume":"24","author":"Jisi","year":"2021","journal-title":"Journal of Applied Science and Engineering"},{"issue":"6","key":"10.7717\/peerj-cs.2876\/ref-18","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1145\/3065386","article-title":"ImageNet classification with deep convolutional neural networks","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Communications of the ACM"},{"key":"10.7717\/peerj-cs.2876\/ref-19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01146","article-title":"Generalized focal loss V2: learning reliable localization quality estimation for dense object detection","author":"Li","year":"2021","journal-title":"IEEE"},{"key":"10.7717\/peerj-cs.2876\/ref-20","first-page":"2980","article-title":"Focal loss for dense object detection","author":"Lin","year":"2017"},{"key":"10.7717\/peerj-cs.2876\/ref-21","first-page":"21","article-title":"Single shot multibox detector","author":"Liu","year":"2016"},{"key":"10.7717\/peerj-cs.2876\/ref-22","first-page":"664","article-title":"Student abnormal behavior recognition in classroom video based on deep learning","author":"Liu","year":"2021"},{"issue":"2","key":"10.7717\/peerj-cs.2876\/ref-23","doi-asserted-by":"publisher","first-page":"36391","DOI":"10.1109\/ACCESS.2024.3373536","article-title":"YOLOv8n_BT: research on classroom learning behavior recognition algorithm based on improved YOLOv8n","volume":"12","author":"Liu","year":"2024","journal-title":"IEEE Access"},{"issue":"12","key":"10.7717\/peerj-cs.2876\/ref-24","doi-asserted-by":"publisher","first-page":"2358","DOI":"10.3390\/electronics13122358","article-title":"SRFAD-Net: scale-robust feature aggregation and diffusion network for object detection in remote sensing images","volume":"13","author":"Liu","year":"2024","journal-title":"Electronics"},{"key":"10.7717\/peerj-cs.2876\/ref-25","first-page":"4929","article-title":"DeepCut: joint subset partition and labeling for multi person pose estimation","author":"Pishchulin","year":"2016"},{"key":"10.7717\/peerj-cs.2876\/ref-26","first-page":"779","article-title":"You only look once: unified, real-time object detection","author":"Redmon","year":"2016"},{"key":"10.7717\/peerj-cs.2876\/ref-27","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1804.02767","article-title":"Yolov3: an incremental improvement","author":"Redmon","year":"2018"},{"issue":"6","key":"10.7717\/peerj-cs.2876\/ref-28","doi-asserted-by":"publisher","first-page":"1137","DOI":"10.1109\/TPAMI.2016.2577031","article-title":"Faster R-CNN: towards real-time object detection with region proposal networks","volume":"39","author":"Ren","year":"2017","journal-title":"IEEE Transactions on Pattern Analysis & Machine Intelligence"},{"key":"10.7717\/peerj-cs.2876\/ref-29","first-page":"288","article-title":"Student behavior detection based on YOLOv4-Bi","author":"Ren","year":"2021"},{"key":"10.7717\/peerj-cs.2876\/ref-30","doi-asserted-by":"publisher","first-page":"116951","DOI":"10.1016\/j.measurement.2025.116951","article-title":"Using mask R-CNN and distance transform to detect elephants on railway tracks for collision prevention","volume":"248","author":"Sudhanan","year":"2025","journal-title":"Measurement"},{"issue":"13","key":"10.7717\/peerj-cs.2876\/ref-31","doi-asserted-by":"publisher","first-page":"6790","DOI":"10.3390\/app12136790","article-title":"Classroom behavior detection based on improved YOLOv5 algorithm combining multi-scale feature fusion and attention mechanism","volume":"12","author":"Tang","year":"2022","journal-title":"Applied Sciences"},{"key":"10.7717\/peerj-cs.2876\/ref-32","first-page":"10418","article-title":"Vits for sits: vision transformers for satellite image time series","author":"Tarasiou","year":"2023"},{"key":"10.7717\/peerj-cs.2876\/ref-33","article-title":"Evolution of Yolo algorithm and Yolov5: the state-of-the-art object detention algorithm. Bachelor\u2019s thesis","author":"Thuan","year":"2021"},{"key":"10.7717\/peerj-cs.2876\/ref-34","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1805.09512","article-title":"You only look twice: rapid multi-scale object detection in satellite imagery","author":"Van Etten","year":"2018"},{"key":"10.7717\/peerj-cs.2876\/ref-35","first-page":"7464","article-title":"YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors","author":"Wang","year":"2023"},{"key":"10.7717\/peerj-cs.2876\/ref-36","article-title":"YOLOV10: real-time end-to-end object detection","volume":"37","author":"Wang","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.7717\/peerj-cs.2876\/ref-37","first-page":"1","article-title":"Yolov9: learning what you want to learn using programmable gradient information","author":"Wang","year":"2024"},{"key":"10.7717\/peerj-cs.2876\/ref-38","first-page":"160","article-title":"Research on intelligent recognition algorithm of college students\u2019 classroom behavior based on improved SSD","author":"Wenchao","year":"2022"},{"key":"10.7717\/peerj-cs.2876\/ref-39","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2310.02522","article-title":"SCB-Dataset3: a benchmark for detecting student classroom behavior","author":"Yang","year":"2023"},{"key":"10.7717\/peerj-cs.2876\/ref-40","first-page":"41","article-title":"Student classroom behavior detection based on YOLOv7+ BRA and multi-model fusion","author":"Yang","year":"2023"},{"issue":"6","key":"10.7717\/peerj-cs.2876\/ref-41","doi-asserted-by":"publisher","first-page":"6947","DOI":"10.1007\/s10639-023-12079-8","article-title":"Using multimodal learning analytics to model students\u2019 learning behavior in animated programming classroom","volume":"29","author":"Yusuf","year":"2024","journal-title":"Education and Information Technologies"},{"key":"10.7717\/peerj-cs.2876\/ref-42","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2023.3258666","article-title":"SuperYOLO: super resolution assisted object detection in multimodal remote sensing imagery","volume":"61","author":"Zhang","year":"2023","journal-title":"IEEE Transactions on Geoscience and Remote Sensing"},{"key":"10.7717\/peerj-cs.2876\/ref-43","first-page":"93","article-title":"Classroom behavior recognition based on improved yolov3","author":"Zhang","year":"2020"},{"key":"10.7717\/peerj-cs.2876\/ref-44","first-page":"16965","article-title":"DETRs beat YOLOs on real-time object detection","author":"Zhao","year":"2024"},{"issue":"4","key":"10.7717\/peerj-cs.2876\/ref-45","doi-asserted-by":"publisher","first-page":"140","DOI":"10.1007\/s11554-024-01515-8","article-title":"Csb-yolo: a rapid and efficient real-time algorithm for classroom student behavior detection","volume":"21","author":"Zhu","year":"2024","journal-title":"Journal of Real-Time Image Processing"}],"container-title":["PeerJ Computer Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/peerj.com\/articles\/cs-2876.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-2876.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-2876.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/peerj.com\/articles\/cs-2876.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,30]],"date-time":"2025-04-30T08:57:28Z","timestamp":1746003448000},"score":1,"resource":{"primary":{"URL":"https:\/\/peerj.com\/articles\/cs-2876"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,30]]},"references-count":45,"alternative-id":["10.7717\/peerj-cs.2876"],"URL":"https:\/\/doi.org\/10.7717\/peerj-cs.2876","archive":["CLOCKSS","LOCKSS","Portico"],"relation":{},"ISSN":["2376-5992"],"issn-type":[{"value":"2376-5992","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4,30]]},"article-number":"e2876"}}