A Gray Relational Analysis-Based Motion Detection Algorithm for Real-World Surveillance Sensor Deployment

• Published in: IEEE sensors journal Vol. 19; no. 3; pp. 1019 - 1027
• Main Authors: Huang, Shih-Chia, Liu, Huibin, Chen, Bo-Hao, Fang, Zhijun, Tan, Tan-Hsu, Kuo, Sy-Yen
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Feature extraction; Image motion analysis; Image segmentation; jitter cameras; Motion detection; Motion detectors; Motion perception; Moving object recognition; Object detection; State of the art; Streaming media; surveillance sensor; Transportation systems; Vibration; Video data; Video surveillance
• ISSN: 1530-437X; 1558-1748
• DOI: 10.1109/JSEN.2018.2879187

The automated detection of moving objects is an essential task for any intelligent transportation system. To achieve reliable and accurate motion detection in video streams acquired from either jitter or static cameras in real-world scenarios, a novel motion detection approach based on gray relational analysis is proposed in this paper, which integrates a multi-sample background generation module and a moving object detection module. As our experimental results demonstrate, the proposed approach attains superior motion detection performance compared to other state-of-the-art techniques based on qualitative and quantitative evaluations. In addition, the processing speed of the proposed approach makes it suitable for real-time applications.