Abnormal behavior detection using sparse representations through sequential generalization of k-means
The potential capability to automatically detect and classify human behavior as either normal or abnormal events is an important aspect in intelligent monitoring/surveillance systems. This study presents a new high-performance framework for detecting behavioral abnormalities in video streams by util...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Turkiye Klinikleri
2021
|
Subjects: | |
Online Access: | http://eprints.utm.my/94025/1/AhlamAlDhamari2021_AbnormalBehaviorDetectionUsingSparse.pdf |
_version_ | 1796865740079366144 |
---|---|
author | Al-Dhamari, Ahlam Sudirman, Rubita Mahmood, Nasrul Humaimi |
author_facet | Al-Dhamari, Ahlam Sudirman, Rubita Mahmood, Nasrul Humaimi |
author_sort | Al-Dhamari, Ahlam |
collection | ePrints |
description | The potential capability to automatically detect and classify human behavior as either normal or abnormal events is an important aspect in intelligent monitoring/surveillance systems. This study presents a new high-performance framework for detecting behavioral abnormalities in video streams by utilizing only the patterns for normal behaviors. In this paper, we used a hybrid descriptor, called a foreground optical flow energy (FGOFE), which makes use of two effective motion techniques in order to extract the most descriptive spatiotemporal features in video sequences. The FGOFE descriptor can effectively capture both weak and sudden incidents in a scene. The sequential generalization of k-means (SGK) algorithm was applied in this study to generate the dictionary set that can sparsely represent each signal; in addition, the orthogonal matching pursuit algorithm was utilized to recover high-dimensional sparse features when referring to a few numbers of noisy linear measurements. Using the SGK allows gaining a less complex and quicker implementation compared to other dictionary learning methods. We conducted comprehensive experiments to analyze and evaluate the ability of our framework in detecting abnormalities using several public benchmarks, which contain different abnormal samples and various contextual compositions. The experimental results show that the proposed framework achieved high detection accuracy (up to 95.33%) and low frame processing time (31 ms on average) compared to the relevant related work. |
first_indexed | 2024-03-05T21:01:40Z |
format | Article |
id | utm.eprints-94025 |
institution | Universiti Teknologi Malaysia - ePrints |
language | English |
last_indexed | 2024-03-05T21:01:40Z |
publishDate | 2021 |
publisher | Turkiye Klinikleri |
record_format | dspace |
spelling | utm.eprints-940252022-02-28T13:31:13Z http://eprints.utm.my/94025/ Abnormal behavior detection using sparse representations through sequential generalization of k-means Al-Dhamari, Ahlam Sudirman, Rubita Mahmood, Nasrul Humaimi TK Electrical engineering. Electronics Nuclear engineering The potential capability to automatically detect and classify human behavior as either normal or abnormal events is an important aspect in intelligent monitoring/surveillance systems. This study presents a new high-performance framework for detecting behavioral abnormalities in video streams by utilizing only the patterns for normal behaviors. In this paper, we used a hybrid descriptor, called a foreground optical flow energy (FGOFE), which makes use of two effective motion techniques in order to extract the most descriptive spatiotemporal features in video sequences. The FGOFE descriptor can effectively capture both weak and sudden incidents in a scene. The sequential generalization of k-means (SGK) algorithm was applied in this study to generate the dictionary set that can sparsely represent each signal; in addition, the orthogonal matching pursuit algorithm was utilized to recover high-dimensional sparse features when referring to a few numbers of noisy linear measurements. Using the SGK allows gaining a less complex and quicker implementation compared to other dictionary learning methods. We conducted comprehensive experiments to analyze and evaluate the ability of our framework in detecting abnormalities using several public benchmarks, which contain different abnormal samples and various contextual compositions. The experimental results show that the proposed framework achieved high detection accuracy (up to 95.33%) and low frame processing time (31 ms on average) compared to the relevant related work. Turkiye Klinikleri 2021-01 Article PeerReviewed application/pdf en http://eprints.utm.my/94025/1/AhlamAlDhamari2021_AbnormalBehaviorDetectionUsingSparse.pdf Al-Dhamari, Ahlam and Sudirman, Rubita and Mahmood, Nasrul Humaimi (2021) Abnormal behavior detection using sparse representations through sequential generalization of k-means. Turkish Journal of Electrical Engineering and Computer Sciences, 29 (1). pp. 152-168. ISSN 1300-0632 http://dx.doi.org/10.3906/ELK-1904-187 DOI:10.3906/ELK-1904-187 |
spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Al-Dhamari, Ahlam Sudirman, Rubita Mahmood, Nasrul Humaimi Abnormal behavior detection using sparse representations through sequential generalization of k-means |
title | Abnormal behavior detection using sparse representations through sequential generalization of k-means |
title_full | Abnormal behavior detection using sparse representations through sequential generalization of k-means |
title_fullStr | Abnormal behavior detection using sparse representations through sequential generalization of k-means |
title_full_unstemmed | Abnormal behavior detection using sparse representations through sequential generalization of k-means |
title_short | Abnormal behavior detection using sparse representations through sequential generalization of k-means |
title_sort | abnormal behavior detection using sparse representations through sequential generalization of k means |
topic | TK Electrical engineering. Electronics Nuclear engineering |
url | http://eprints.utm.my/94025/1/AhlamAlDhamari2021_AbnormalBehaviorDetectionUsingSparse.pdf |
work_keys_str_mv | AT aldhamariahlam abnormalbehaviordetectionusingsparserepresentationsthroughsequentialgeneralizationofkmeans AT sudirmanrubita abnormalbehaviordetectionusingsparserepresentationsthroughsequentialgeneralizationofkmeans AT mahmoodnasrulhumaimi abnormalbehaviordetectionusingsparserepresentationsthroughsequentialgeneralizationofkmeans |