Geographical Detection of Traffic Congestion Using Machine Learning Algorithms
It is extremely tedious and time-consuming to keep watching all the day and identify congestion from the current surveillance system using in traffic monitoring hall. Furthermore, it is impossible to watch all the cameras relies on human eyes considering numerous cameras covering a large-scale region using in the freeway. However, prompt detection of the traffic congestion in large-scale region is important. Prompt detection can prevent extended congestion with devastating evolution from the initial controllable traffic congestion, which is one of the important applications in intelligent transport system (ITS). Deep learning algorithms have the potential implementation meanings to be intensely used in many fields of the transportation system, from traffic flow prediction to traffic congestion recognition. Classification of traffic condition is one of the most important parts of an ITS, which can be widely utilized in traffic control strategies, traffic flow analysis and so on. Thus it is necessary to propose an intelligent transport system. So, as we use the convolutional neural network (CNN) to filter the traffic videos and images to form the better results to update in TrafficNet. Traffic feature is successfully extracted and interpreted for classification. In order to further improve the detection accuracy of deep learning approaches, residual learning is proposed and has successfully applied to various aspects. The residual network-based approaches make the deep network to get better performance. With the exploration of the state-of-art deep residual network, improvement technologies are investigated including typical milestone nets, data augmentation methods, transfer technology and supervised learning for higher detection accuracy. Consider the feature difference between traffic image and that in vision benchmarks, a systematic study should be conducted to push the congestion recognition accuracy to a new level for daily surveillance use. For this practical application, the generalized ability of the network for new input images and videos is significant to explore.