Image Preprocessing In Intelligent Traffic And Research On Specific Illegal Vehicles

In recent years, with the progress of science and technology and the development of computer and communication technology, people have begun to focus on the traffic reform by video technology. This change calls for the intelligent transportation system, in which the preprocessing of images and the illegal vehicles detection are the two key technologies. Specifically, the effect of image preprocessing directly influences the follow-up series of tests; and detection of illegal vehicles, which directly concerns with the safety of traffic transportation, acts as the core function in intelligent transportation. This paper will deeply discuss the two aspects, and the main work follows below:For preprocessing of images, this paper proposes a new concept of the Mean Neighbor (MN) based on the analysis of the popular filtering algorithms, then designs and implements the corresponding filtering algorithm. Firstly, the new method chooses the window size of the filter adaptively, and then calculates the mean value after removing the noise. Secondly, it proposes three ways of calculating the MN and two ways of judging the noise, then combines them to design nine filtering algorithms. The experiments show that the7th filtering algorithm has got the best filtering performance after comparing the above nine filtering algorithms with the mean filter, the median filter and some other improved filtering algorithms. Also the7th filtering algorithm is more pervasive and can deal with different noise types.For detection of illegal vehicles, a improved detection algorithm for vehicle center was proposed by working over the traditional algorithm; and the new algorithm, which judge the coming car by the illegal condition of its former vehicle, greatly improve the rate of detection. In addition, the background subtraction technique was used to separate the background and moving objects; consequently filtrating of motor vehicles gets achieved by referring to the geometric characteristics of the moving object. The experiment show that comparing with the traditional detection methods the new algorithm can detect illegal vehicles better.