An unsupervised approach for traffic motion patterns extraction

Abstract

Automatic analysis, understanding typical activities, and identifying vehicle behaviour in crowded traffic scenes are fundamental and challenging tasks for traffic video surveillance. Some recent researches have been using machine learning approaches to extract meaningful patterns occurring in a traffic scene, for example, intersection. In this regard, we convert visual patterns and features to visual words using dense and sparse optical flow and learning traffic motion patterns with group sparse topical coding (GSTC) algorithm. In the first step of the proposed algorithm, the input traffic video is divided into non‐overlapping clips. After that, motion vectors are extracted using dual TV‐L1 as a dense optical flow and Lucas–Kanade as a sparse optical flow and converted to flow words. For learning traffic motion patterns, the GSTC algorithm, that is, a non‐probabilistic topic model (TM) has been applied. These patterns represent priors on observable motion, which can be utilised to describe a scene and answer behaviour questions such as what are the motion patterns in a traffic scene and what is going on. The experimental results which have been obtained using a real dataset, QUML, show that the combination of the GSTC + dual TV‐L1 extracts more traffic motion patterns in comparison with the GSTC + Lucas–Kanade and previous studies.

Computer vision-based framework for pedestrian movement direction recognition

“…An unsupervised mobile robot have been developed to learn human motion patterns in a real-world environment to predict future behaviors [43]. In [44] authors discussed group sparse topical coding (GSTC) technique to study motion patterns. In [45] used trajectory clustering based on low level trajectory patterns and high level discrete transitions for predicting anomalous pedestrian detection.…”

Traffic Behavior Recognition from Traffic Videos under Occlusion Condition: A Kalman Filter Approach

“…The proposed architecture achieves an average of 1.359 s in relation to time-to-accident measure, with an average precision of 17.36%. Moradi et al employed an unsupervised approach to extract traffic motion patterns from optical flow, which is not capable of recognizing movement patterns at object level ( 19 ). In Chen et al, trajectories are divided into different kinds of behaviors based on B-spline control points ( 29 ).…”

“…Besides these sensors, surveillance cameras provide researchers and city planners with real scenes of traffic conditions. With traffic cameras, it is possible to not only extract vehicle count, density, and speed, but also vehicle motion patterns, and behavior and interaction between vehicle and pedestrians (14)(15)(16)(17)(18)(19)(20)(21)(22). With comprehensive and rich content, surveillance videos can provide the task of traffic control with detailed and accurate traffic information.…”