Video trajectory analysis using unsupervised clustering and multi-criteria ranking
- PDF / 3,739,691 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 64 Downloads / 163 Views
METHODOLOGIES AND APPLICATION
Video trajectory analysis using unsupervised clustering and multi-criteria ranking Arif Ahmed Sekh1 · Debi Prosad Dogra3 · Samarjit Kar2 · Partha Pratim Roy4
© The Author(s) 2020
Abstract Surveillance camera usage has increased significantly for visual surveillance. Manual analysis of large video data recorded by cameras may not be feasible on a larger scale. In various applications, deep learning-guided supervised systems are used to track and identify unusual patterns. However, such systems depend on learning which may not be possible. Unsupervised methods relay on suitable features and demand cluster analysis by experts. In this paper, we propose an unsupervised trajectory clustering method referred to as t-Cluster. Our proposed method prepares indexes of object trajectories by fusing high-level interpretable features such as origin, destination, path, and deviation. Next, the clusters are fused using multi-criteria decision making and trajectories are ranked accordingly. The method is able to place abnormal patterns on the top of the list. We have evaluated our algorithm and compared it against competent baseline trajectory clustering methods applied to videos taken from publicly available benchmark datasets. We have obtained higher clustering accuracies on public datasets with significantly lesser computation overhead. Keywords Unsupervised clustering · Object trajectory · Motion analysis
1 Introduction and related works Object motion pattern identification and trajectory analysis are two important steps in various computer vision applications (Ahmed et al. 2018b). Trajectory analysis is used in many video analysis tasks such as video summarization (Dogra et al. 2016; Ajmal et al. 2017), event detection (Reddy and Veena 2018), and visual surveillance (Vishwakarma and Communicated by V. Loia.
B
Arif Ahmed Sekh [email protected] Debi Prosad Dogra [email protected] Samarjit Kar [email protected] Partha Pratim Roy [email protected]
1
UiT The Arctic University of Norway, TromsØ, Norway
2
National Institute of Technology Durgapur, Durgapur, India
3
Indian Institute of Technology Bhubaneswar, Bhubaneswar, India
4
Indian Institute of Technology Roorkee, Roorkee, India
Agrawal 2013; Huang et al. 2018). Analysis of large volume trajectory can be effective in traffic analysis (Santhosh et al. 2018) and crowd monitoring (Bera et al. 2016). The primary application of such analysis is abnormality detection (Roshtkhari and Levine 2013; Mabrouk and Zagrouba 2018). However, unsupervised clustering of trajectories is a difficult task. Clustering using simple features extracted from object trajectories, e.g. object location (xi , yi , ti ), produces poor results (Xu et al. 2015). They cannot be used for complex and long-term analysis. High-level features like source, destination, path, and activity can be used to represent moving objects. These high-level features can help to find patterns and group them together. The objective can be to classify the trajectorie
Data Loading...
