Spatio-temporal anomaly detection for real-time video surveillance using SpatioGuard-YOLO framework.
Saved in:
| Title: | Spatio-temporal anomaly detection for real-time video surveillance using SpatioGuard-YOLO framework. |
|---|---|
| Authors: | Krishnan, Sreedevi R.1 (AUTHOR) sreedevi.cs@adishankara.ac.in, Amudha, P2 (AUTHOR) amudha_cse@avinuty.ac.in |
| Source: | Multimedia Tools & Applications. Jun2026, Vol. 85 Issue 6, p1-33. 33p. |
| Abstract: | Detecting anomalous events in real-time video surveillance is challenging due to cluttered backgrounds, overlapping objects, low-light conditions, and complex spatio-temporal variations. To address these issues, this paper proposes SpatioGuard-YOLO, an integrated framework that combines adaptive contrast enhancement, efficient object detection, and multi-scale spatial feature learning for robust video anomaly detection. A Dynamic Contrast Enhancer is introduced to improve visibility under adverse lighting conditions, while a novel Spatially Expanded Neural Network captures both global contextual information and fine-grained local details without compromising real-time performance. The proposed framework is evaluated on a real-time CCTV surveillance dataset designed for anomaly detection in pedestrian-restricted pathways, demonstrating strong generalization under diverse environmental conditions. By integrating these components with a YOLO-based detection pipeline, the proposed framework effectively reduces false detections in complex surveillance environments. Experimental results demonstrate that SpatioGuard-YOLO achieves high detection accuracy, exceeding 99% overall performance and outperforming existing state-of-the-art methods, confirming its effectiveness and reliability for real-world video surveillance applications. [ABSTRACT FROM AUTHOR] |
| Copyright of Multimedia Tools & Applications is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
Be the first to leave a comment!
