Occlusion vehicle detection algorithm in crowded scene for Traffic Surveillance System

Traffic Surveillance System (TSS) plays an important role in extracting necessary information (count, type, speed, etc.). In the area of Traffic Surveillance System (TSS), vehicle detection has emerged as an influential field of study. So far there has been a considerable amount of research to accommodate this subject. However, these studies almost address problems in developed countries where the traffic infrastructure is constructed to appropriate automobiles. Detecting moving vehicles in urban areas is difficult because the inter-vehicle space is significantly reduced, increasing the occlusion between vehicles. This issue is more challenging in developing countries where the roads are crowded with 2-wheeled motorbikes in rush hours. This paper presents a method to improve the occlusion vehicle detection from static surveillance cameras. The proposed method is a vision-based approach in which undefined blobs of occluded vehicles are examined to extract the vehicles individually based on the geometric and the ellipticity characteristic of objects' shapes. Experiments have been carried out with the real-world data to evaluate the performance and the accuracy of our method. The assessment results are promising for a detection rate of 84.10% at daytime.

I. Introduction

The past decade has seen the explosion of intelligent and expert systems, especially in the area of transportation management. Traffic surveillance system (TSS) has gained popularity among researchers and authorities. A key aspect of TSS is to derive the traffic information (count, average speed, and the density of each vehicle type) for further analysis related to traffic management and planning. In this context, many studies have been conducted in developed countries where the transportation frameworks are constructed primarily for automobiles. These systems [1], [2] were developed with the advanced equipment and sensors to optimize the incoming signal including radar, infrared camera and so on. However, in developing countries, the application of these systems has trouble with high cost and incompatible infrastructures. On the contrary, the vision-based TSSs which are built from computer vision and image processing techniques [3]–[5] have shown more superior capability with lower cost and easier to maintain. Moreover, they are extremely versatile as algorithms are designed to cope with a broad range of operations such as detect, identify, count, track, and classify vehicles.