Article Preview
Top1. Introduction
The field of traffic monitoring has generated great excitement in recent years within the intelligent transport systems community due to the increase of hardware development, the low-cost sensor technologies and the improvement in the development and optimization of data processing algorithms. Specifically, the video detection and monitoring solutions for traffic applications can help to improve the performance in traffic management (Luque-Baena et al., 2015; Kamijo et al., 2000; Cheng et al., 2011; Zhou et al, 2007). Thus, for example, if a high frequency of heavy vehicles is detected in one of the analyzed road sections, it is possible to redirect the traffic in a previous point with the aim of avoiding traffic congestion.
Some recent works have used deep learning for diverse traffic environment purposes. In (Wshah et al., 2016), three popular convolutional neural networks (CNN) are adapted to the detection of passengers in high occupancy vehicles and high occupancy tolling lanes. On the other hand, (Amato et al., 2017) presents a new system based on a small deep CNN, whose aim is to detect the occupancy level in a parking lot. Another example can be found in (Lv et al., 2015), where a deep-learning-based method is proposed in order to predict the traffic flow, or a traffic light recognition algorithm for varying illumination conditions (John et al., 2014).
Automatic video surveillance systems can be divided into several phases ((Buch et al., 2011), (Baumann et al., 2008)). A first step involves the detection of moving objects within the scene; a second stage performs monitoring tasks to associate the same vehicle detected in all frames of the sequence in which it appears; and finally, a feature detection phase to extract relevant knowledge of the movement of these objects, their behavior and appearance.
Each stage builds on the previous one, which implies that it is needed to have implemented the steps of detecting and tracking objects if it is intended to conduct an analysis of the detected vehicle. In this paper, some methods previously developed by our research team for object detection and tracking are combined with other techniques to yield a vehicle classification system. Specifically, a self-organizing neural network is applied to cluster the pixels in background and foreground layers in order to detect which pixels are in motion inside the scene (LopezRubio et al., 2011). Subsequently, a Kalman model for multiple objects is applied to determine the trajectory of each vehicle which appears in the scene (Rad et al., 2005).
Therefore, the aim of this work is to classify the detected vehicles in four categories: car, motorcycle, truck and van. A feature extraction process is required in order to obtain robust and discriminant characteristics which can differentiate correctly among the groups of vehicles. This analysis would help to manage and distribute the traffic more efficiently in the analyzed area. Other works in the literature have the same aim, although some of them apply different classification or clustering methods (Crouzil et al., 2016; Huang et al., 2016) or start from other methodologies associated to video surveillance systems (Liang et al., 2015). In this case, the Growing Neural Gas model (GNG) is considered, since it has been used successfully in different classification problems, from novelty detection (Fink et al., 2015) to text classification (Wang et al., 2007), or even issues related to medicine or biology such as osteoporosis detection ((Podolak et al., 2013), (de Oliveira Martins et al., 2009), (Ölmez et al., 2003)).