Traffic anomaly detection based on image descriptor in videos

Abstract

The huge and ever growing volume of traffic video poses a compelling demand for efficient automatic detection of traffic anomaly. In this paper, a new traffic anomaly detection algorithm is introduced. It firstly divides a traffic video into several video cubes in temporal domain, and each video cube is divided into video blocks in spatial domain. Each image block of a video block is described using the local invariant features and the visual codebook approach. Based on the descriptor of the image block, we count the category number of the block (CNB) of a video block. Then, a Gaussian distribution model for estimating the probability of normal traffic with respect to the CNB is learned. The learned Gaussian distribution model is then used to detect the traffic anomaly from the test traffic video. Eventually, the results of all video blocks are fused to achieve the final decision. Experimental results show that the proposed algorithm performs better than two existing algorithms on both the intersection traffic videos and main road traffic videos.

Dynamics of information diffusion and its applications on complex networks

“…The key problem is how to mine the propagation probability of the links, and even sometimes the network structure is incomplete which need to infer the coupled of network structure and propagation probability [182]. The second type is feature based method, which needs to extract a list of features that might affect the information diffusion, including the content, users, network structure and temporal features [183,184], and then apply different machine learning methods to make the prediction of the information popularity in the final state [92,179]. Although the feature based method could generate good performance of the prediction, the challenge is that the result is sensitive to the quality of the extracted features.…”

Detecting abnormal events in traffic video surveillance using superorientation optical flow feature

“…Vehicle with varying speed produces a different set of points [5]. Traffic scene contains all kinds of objects such as vehicles, humans, animals etc., which is very difficult to track based on object classification [19]. Labelling of normal and abnormal data in a traffic scene is very difficult [7].…”

Detection of Anomalous Traffic Patterns and Insight Analysis from Bus Trajectory Data

“…Nguyen et al [15] developed a real-time system using social media (Twitter) data for traffic incident detection. Li et al [11] introduced a traffic anomaly detection algorithm based on the massive traffic video. Riveiro et al [17] constructed a visual analytics framework that employs large amounts of multidimensional and heterogeneous road traffic data for traffic anomaly detection.…”