<?xml version="1.1" encoding="utf-8"?>
<article xsi:noNamespaceSchemaLocation="http://jats.nlm.nih.gov/publishing/1.1/xsd/JATS-journalpublishing1-mathml3.xsd" dtd-version="1.1" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"><front><journal-meta><journal-id journal-id-type="publisher-id">ETI</journal-id><journal-title-group><journal-title>Education and Teaching Innovation</journal-title></journal-title-group><issn>2995-4894</issn><eissn>2995-4908</eissn><publisher><publisher-name>Art and Technology</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.61369/ETI.2026060024</article-id><article-categories><subj-group subj-group-type="heading"><subject>Article</subject></subj-group></article-categories><title>基于时序一致性约束的视频异常报警优化方法</title><url>https://artdesignp.com/journal/ETI/4/6/10.61369/ETI.2026060024</url><author>张师源</author><pub-date pub-type="publication-year"><year>2026</year></pub-date><volume>4</volume><issue>6</issue><history><date date-type="pub"><published-time>2026-06-20</published-time></date></history><abstract>针对弱监督视频异常检测中分数波动引起的报警频繁切换和短时误报，提出一种时序一致性报警后处理方法。该方法在16帧片段尺度上依次进行指数移动平均、双阈值判定、短间隔合并和最短持续时间约束，并利用校准子集确定工作点。基于DSANet 公开预训练模型，在UCF-Crime和XD-Violence上进行5次重复分层留出实验。F1工作点下，两个数据集的每百视频纯误报事件数分别由59.14降至24.91和10.81降至5.97，报警切换次数分别下降41.1%和19.9%，帧级F1基本保持。该方法无需重新训练，平均处理时间不超过1.94 ms/视频，但事件召回率略有下降，报警延迟有所增加。</abstract><keywords>视频异常检测,时序一致性,事件级误报</keywords></article-meta></front><body/><back><ref-list><ref id="B1" content-type="article"><label>1</label><element-citation publication-type="journal"><p>[1] 汪洋, 周脚根, 严俊, 等. 基于深度学习的监控视频异常检测方法综述[J].中国图象图形学报,2025,30(3):615-640.[2] SULTANI W, CHEN C, SHAH M. Real-world anomaly detection in surveillance videos[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018: 6479-6488.[3] WU P, LIU J, SHI Y, et al. Not only look, but also listen: learning multimodal violence detection under weak supervision[C]//European Conference on Computer Vision. Cham: Springer, 2020: 322-339.[4] TIAN Y, PANG G, CHEN Y, et al. Weakly-supervised video anomaly detection with robust temporal feature magnitude learning[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 4975-4986.[5] 朱新瑞, 钱小燕, 施俞洲, 等. 长短期时间序列关联的视频异常事件检测[J].中国图象图形学报,2024,29(7):1998-2010.[6] RADFORD A, KIM J W, HALLACY C, et al. Learning transferable visual models from natural language supervision[C]//International Conference on Machine Learning. PMLR, 2021: 8748-8763.[7] WU P, ZHOU X, PANG G, et al. VadCLIP: adapting vision-language models for weakly supervised video anomaly detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2024, 38(6): 6074-6082.[8] YIN W, ZHANG H, WANG X, et al. Learning to tell apart: weakly supervised video anomaly detection via disentangled semantic alignment[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2026, 40(14): 12027-12035.[9] KARIM H, DOSHI K, YILMAZ Y. Real-time weakly supervised video anomaly detection[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2024: 6848-6856.[10] CANNY J. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6): 679-698.</p><pub-id pub-id-type="doi"/></element-citation></ref></ref-list></back></article>
