Vehicle detection even in poor visibility conditions using infrared thermal images and its application to road traffic flow monitoring

Abstract

We propose an algorithm for detecting vehicle positions and their movements by using thermal images obtained through an infrared thermography camera. The proposed algorithm specifies the area of moving vehicles based on the variations of pixel values, i.e. the standard deviations of pixel values along the time direction of spatio-temporal images. It also specifies vehicle positions by applying the pattern recognition algorithm which uses Haar-like features per frame of the images. Moreover, to increase the accuracy of vehicle detection, correction procedures for misrecognition of vehicles are employed. The results of our experiments show that the information about both vehicle positions and their movements can be obtained by combining those two kinds of detection, and the vehicle detection accuracy is 96.3 %. As an application of the algorithm, we also propose a method for estimating traffic flow conditions based on the results obtained by the algorithm. By use of the method for estimating traffic flow conditions, automatic traffic flow monitoring can be achieved. In addition, there is a possibility that traffic accidents, vehicle troubles, and illegal parking can be detected with the proposed method. By using the traffic information obtained from the proposed method, we also expect to realize an optimized traffic signal control around the clock even in changeable weather. © 2013 Springer Science+Business Media.