Vehicular Traffic Congestion Classification by Visual Features and Deep Learning Approaches: A Comparison

Abstract

Automatic traffic flow classification is useful to reveal road congestions and accidents. Nowadays, roads and highways are equipped with a huge amount of surveillance cameras, which can be used for real-time vehicle identification, and thus providing traffic flow estimation. This research provides a comparative analysis of state-of-the-art object detectors, visual features, and classification models useful to implement traffic state estimations. More specifically, three different object detectors are compared to identify vehicles. Four machine learning techniques are successively employed to explore five visual features for classification aims. These classic machine learning approaches are compared with the deep learning techniques. This research demonstrates that, when methods and resources are properly implemented and tested, results are very encouraging for both methods, but the deep learning method is the most accurately performing one reaching an accuracy of 99.9% for binary traffic state classification and 98.6% for multiclass classification.

Deep Learning-Based Congestion Detection at Urban Intersections

“…Most of the related researches on traffic state discrimination [ 20 , 21 ] were based on the single parameter evaluation, and the detection results are not accurate. With the development of artificial neural networks [ 22 ], the deep learning model is introduced into congestion detection to detect vehicle targets. However, it needs to build a deeper network to improve the detection accuracy, and the computational complexity is greatly increased, which is infeasible for real-time target detection.…”

Reliable and Rapid Traffic Congestion Detection Approach Based on Deep Residual Learning and Motion Trajectories

“…In the literature, several traffic congestion detection methods have used SVMs and achieved good results (e.g., [4], [35]). SVMs can be used for both linearly separable and nonlinearly separable data based on a kernel trick [36].…”

Advanced Computational Intelligence for Object Detection, Feature Extraction and Recognition in Smart Sensor Environments

“…The proposed complex neural architecture was based on a time convolutional layer model which helped to compare the extracted ship features. In [ 3 ], the authors discuss a model of vehicular traffic congestion with various approaches. As a result of this, a study presented a set of comparative results for different deep learning models.…”