Deep anomaly detection through visual attention in surveillance videos
Abstract This paper describes a method for learning anomaly behavior in the video by finding an attention region from spatiotemporal information, in contrast to the full-frame learning. In our proposed method, a robust background subtraction (BG) for extracting motion, indicating the location of att...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2020-10-01
|
| Series: | Journal of Big Data |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s40537-020-00365-y |
| _version_ | 1828495130435256320 |
|---|---|
| author | Nasaruddin Nasaruddin Kahlil Muchtar Afdhal Afdhal Alvin Prayuda Juniarta Dwiyantoro |
| author_facet | Nasaruddin Nasaruddin Kahlil Muchtar Afdhal Afdhal Alvin Prayuda Juniarta Dwiyantoro |
| author_sort | Nasaruddin Nasaruddin |
| collection | DOAJ |
| description | Abstract This paper describes a method for learning anomaly behavior in the video by finding an attention region from spatiotemporal information, in contrast to the full-frame learning. In our proposed method, a robust background subtraction (BG) for extracting motion, indicating the location of attention regions is employed. The resulting regions are finally fed into a three-dimensional Convolutional Neural Network (3D CNN). Specifically, by taking advantage of C3D (Convolution 3-dimensional), to completely exploit spatiotemporal relation, a deep convolution network is developed to distinguish normal and anomalous events. Our system is trained and tested against a large-scale UCF-Crime anomaly dataset for validating its effectiveness. This dataset contains 1900 long and untrimmed real-world surveillance videos and splits into 950 anomaly events and 950 normal events, respectively. In total, there are approximately ~ 13 million frames are learned during the training and testing phase. As shown in the experiments section, in terms of accuracy, the proposed visual attention model can obtain 99.25 accuracies. From the industrial application point of view, the extraction of this attention region can assist the security officer on focusing on the corresponding anomaly region, instead of a wider, full-framed inspection. |
| first_indexed | 2024-12-11T12:07:44Z |
| format | Article |
| id | doaj.art-3a7831f4d3204bd98a28df4a4e02aa20 |
| institution | Directory Open Access Journal |
| issn | 2196-1115 |
| language | English |
| last_indexed | 2024-12-11T12:07:44Z |
| publishDate | 2020-10-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Journal of Big Data |
| spelling | doaj.art-3a7831f4d3204bd98a28df4a4e02aa202022-12-22T01:07:53ZengSpringerOpenJournal of Big Data2196-11152020-10-017111710.1186/s40537-020-00365-yDeep anomaly detection through visual attention in surveillance videosNasaruddin Nasaruddin0Kahlil Muchtar1Afdhal Afdhal2Alvin Prayuda Juniarta Dwiyantoro3Department of Electrical and Computer Engineering, Syiah Kuala UniversityDepartment of Electrical and Computer Engineering, Syiah Kuala UniversityDepartment of Electrical and Computer Engineering, Syiah Kuala UniversityNodefluxAbstract This paper describes a method for learning anomaly behavior in the video by finding an attention region from spatiotemporal information, in contrast to the full-frame learning. In our proposed method, a robust background subtraction (BG) for extracting motion, indicating the location of attention regions is employed. The resulting regions are finally fed into a three-dimensional Convolutional Neural Network (3D CNN). Specifically, by taking advantage of C3D (Convolution 3-dimensional), to completely exploit spatiotemporal relation, a deep convolution network is developed to distinguish normal and anomalous events. Our system is trained and tested against a large-scale UCF-Crime anomaly dataset for validating its effectiveness. This dataset contains 1900 long and untrimmed real-world surveillance videos and splits into 950 anomaly events and 950 normal events, respectively. In total, there are approximately ~ 13 million frames are learned during the training and testing phase. As shown in the experiments section, in terms of accuracy, the proposed visual attention model can obtain 99.25 accuracies. From the industrial application point of view, the extraction of this attention region can assist the security officer on focusing on the corresponding anomaly region, instead of a wider, full-framed inspection.http://link.springer.com/article/10.1186/s40537-020-00365-yVisual attention approachConvolutional neural network (CNN)Integrated surveillance systemAnomaly classification |
| spellingShingle | Nasaruddin Nasaruddin Kahlil Muchtar Afdhal Afdhal Alvin Prayuda Juniarta Dwiyantoro Deep anomaly detection through visual attention in surveillance videos Journal of Big Data Visual attention approach Convolutional neural network (CNN) Integrated surveillance system Anomaly classification |
| title | Deep anomaly detection through visual attention in surveillance videos |
| title_full | Deep anomaly detection through visual attention in surveillance videos |
| title_fullStr | Deep anomaly detection through visual attention in surveillance videos |
| title_full_unstemmed | Deep anomaly detection through visual attention in surveillance videos |
| title_short | Deep anomaly detection through visual attention in surveillance videos |
| title_sort | deep anomaly detection through visual attention in surveillance videos |
| topic | Visual attention approach Convolutional neural network (CNN) Integrated surveillance system Anomaly classification |
| url | http://link.springer.com/article/10.1186/s40537-020-00365-y |
| work_keys_str_mv | AT nasaruddinnasaruddin deepanomalydetectionthroughvisualattentioninsurveillancevideos AT kahlilmuchtar deepanomalydetectionthroughvisualattentioninsurveillancevideos AT afdhalafdhal deepanomalydetectionthroughvisualattentioninsurveillancevideos AT alvinprayudajuniartadwiyantoro deepanomalydetectionthroughvisualattentioninsurveillancevideos |