EADN: An Efficient Deep Learning Model for Anomaly Detection in Videos
Surveillance systems regularly create massive video data in the modern technological era, making their analysis challenging for security specialists. Finding anomalous activities manually in these enormous video recordings is a tedious task, as they infrequently occur in the real world. We proposed...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/9/1555 |
| _version_ | 1797503802337656832 |
|---|---|
| author | Sareer Ul Amin Mohib Ullah Muhammad Sajjad Faouzi Alaya Cheikh Mohammad Hijji Abdulrahman Hijji Khan Muhammad |
| author_facet | Sareer Ul Amin Mohib Ullah Muhammad Sajjad Faouzi Alaya Cheikh Mohammad Hijji Abdulrahman Hijji Khan Muhammad |
| author_sort | Sareer Ul Amin |
| collection | DOAJ |
| description | Surveillance systems regularly create massive video data in the modern technological era, making their analysis challenging for security specialists. Finding anomalous activities manually in these enormous video recordings is a tedious task, as they infrequently occur in the real world. We proposed a minimal complex deep learning-based model named EADN for anomaly detection that can operate in a surveillance system. At the model’s input, the video is segmented into salient shots using a shot boundary detection algorithm. Next, the selected sequence of frames is given to a Convolutional Neural Network (CNN) that consists of time-distributed 2D layers for extracting salient spatiotemporal features. The extracted features are enriched with valuable information that is very helpful in capturing abnormal events. Lastly, Long Short-Term Memory (LSTM) cells are employed to learn spatiotemporal features from a sequence of frames per sample of each abnormal event for anomaly detection. Comprehensive experiments are performed on benchmark datasets. Additionally, the quantitative results are compared with state-of-the-art methods, and a substantial improvement is achieved, showing our model’s effectiveness. |
| first_indexed | 2024-03-10T03:55:32Z |
| format | Article |
| id | doaj.art-5a7b84dbb5784b55a4b2ce00e71111f5 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-10T03:55:32Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-5a7b84dbb5784b55a4b2ce00e71111f52023-11-23T08:46:05ZengMDPI AGMathematics2227-73902022-05-01109155510.3390/math10091555EADN: An Efficient Deep Learning Model for Anomaly Detection in VideosSareer Ul Amin0Mohib Ullah1Muhammad Sajjad2Faouzi Alaya Cheikh3Mohammad Hijji4Abdulrahman Hijji5Khan Muhammad6Digital Image Processing Laboratory, Department of Computer Science, Islamia College Peshawar, Peshawar 25000, PakistanSoftware, Data and Digital Ecosystems, Department of Computer Science, Norwegian University for Science and Technology (NTNU), 2815 Gjøvik, NorwayDigital Image Processing Laboratory, Department of Computer Science, Islamia College Peshawar, Peshawar 25000, PakistanSoftware, Data and Digital Ecosystems, Department of Computer Science, Norwegian University for Science and Technology (NTNU), 2815 Gjøvik, NorwayIndustrial Innovation and Robotic Center (IIRC), University of Tabuk, Tabuk 47711, Saudi ArabiaDepartment of Civil and Environmental Engineering, King Fahd University of Petroleum and Minerals (KFUPM), Dharan 31261, Saudi ArabiaVisual Analytics for Knowledge Laboratory (VIS2KNOW Lab), Department of Applied Artificial Intelligence, School of Convergence, College of Computing and Informatics, Sungkyunkwan University, Seoul 03063, KoreaSurveillance systems regularly create massive video data in the modern technological era, making their analysis challenging for security specialists. Finding anomalous activities manually in these enormous video recordings is a tedious task, as they infrequently occur in the real world. We proposed a minimal complex deep learning-based model named EADN for anomaly detection that can operate in a surveillance system. At the model’s input, the video is segmented into salient shots using a shot boundary detection algorithm. Next, the selected sequence of frames is given to a Convolutional Neural Network (CNN) that consists of time-distributed 2D layers for extracting salient spatiotemporal features. The extracted features are enriched with valuable information that is very helpful in capturing abnormal events. Lastly, Long Short-Term Memory (LSTM) cells are employed to learn spatiotemporal features from a sequence of frames per sample of each abnormal event for anomaly detection. Comprehensive experiments are performed on benchmark datasets. Additionally, the quantitative results are compared with state-of-the-art methods, and a substantial improvement is achieved, showing our model’s effectiveness.https://www.mdpi.com/2227-7390/10/9/1555anomaly detectionshots segmentationcomputer visiondeep learninghistogram differencekeyframe extraction |
| spellingShingle | Sareer Ul Amin Mohib Ullah Muhammad Sajjad Faouzi Alaya Cheikh Mohammad Hijji Abdulrahman Hijji Khan Muhammad EADN: An Efficient Deep Learning Model for Anomaly Detection in Videos Mathematics anomaly detection shots segmentation computer vision deep learning histogram difference keyframe extraction |
| title | EADN: An Efficient Deep Learning Model for Anomaly Detection in Videos |
| title_full | EADN: An Efficient Deep Learning Model for Anomaly Detection in Videos |
| title_fullStr | EADN: An Efficient Deep Learning Model for Anomaly Detection in Videos |
| title_full_unstemmed | EADN: An Efficient Deep Learning Model for Anomaly Detection in Videos |
| title_short | EADN: An Efficient Deep Learning Model for Anomaly Detection in Videos |
| title_sort | eadn an efficient deep learning model for anomaly detection in videos |
| topic | anomaly detection shots segmentation computer vision deep learning histogram difference keyframe extraction |
| url | https://www.mdpi.com/2227-7390/10/9/1555 |
| work_keys_str_mv | AT sareerulamin eadnanefficientdeeplearningmodelforanomalydetectioninvideos AT mohibullah eadnanefficientdeeplearningmodelforanomalydetectioninvideos AT muhammadsajjad eadnanefficientdeeplearningmodelforanomalydetectioninvideos AT faouzialayacheikh eadnanefficientdeeplearningmodelforanomalydetectioninvideos AT mohammadhijji eadnanefficientdeeplearningmodelforanomalydetectioninvideos AT abdulrahmanhijji eadnanefficientdeeplearningmodelforanomalydetectioninvideos AT khanmuhammad eadnanefficientdeeplearningmodelforanomalydetectioninvideos |