A Complex Traffic Scene Image Enhancement And Its Application In The License Plate Recognition

With exclusive image processing technology has developed rapidly day month new science and technology, and gradually get more and more attention. At the same time, to promote social and economic prosperity of transport colorful scene showing a trend, people travel to provide a convenient, but also led to the complexity of the vast majority of our local transport scene. Faced with complex traffic scenarios, combined with more traffic and shuttling back and forth of vehicles and pedestrians are intertwined, very confusing, in large part led to difficulties in license plate recognition, but also to the work of people travel and transport sector to bring a lot of obstacles. In recent years, the rapid development of network technology presents the trend, license plate recognition technology is particularly important to highlight its position, its importance is self-evident. License plate recognition technology is not only related to the commonly used computer technology and image processing techniques described earlier, but also covers the emerging pattern recognition and artificial intelligence and neural networks and a series of advanced technologies. Currently, the license plate recognition technology in the field of intelligent transportation national advocacy plays an important role, not only that, but also has very important significance in today’s ETC and security in the field of intelligent vehicles. Generally speaking, the license plate recognition technology, including for the image preprocessing and positioning as well as the license plate number of the license plate character segmentation and, for some, the movement of the vehicle license plate recognition efficiency and the condition of the natural environment is closely related to the license plate location.Therefore, the study of complex traffic environment under image enhancement technology background and further to apply it to a very important significance license plate recognition. This paper describes mainly through image enhancement associated theoretical knowledge as a basis for image enhancement algorithms in depth, combined with practical, detailed analysis for image enhancement in complex traffic scenarios related application license plate recognition technology, through specific experiments show that solve complex traffic scene image enhancement technology and related algorithms, in order to enhance the recognition of license plates in complex traffic scenarios. In this paper, the process of the specific study, but also focus on the current license plate recognition in some of the core technology and algorithms, through analysis, integration finalized effective license plate recognition algorithm. Currently onthe market for the license plate recognition inefficiency, low accuracy and so on, this article will both be improved by appropriate algorithms. For the image pre-processing problems, such as: the presence of the image causes the image enhancement techniques to remove the noise, and there is an image blur caused by the movement of the noise produced by the vehicle a detailed exposition. Furthermore the introduction of a common image enhancement algorithms, to lay the foundation for the later image to identify and enhanced with increased effectiveness. Several image enhancement algorithm based on the above, finally proposed an improved image blur parameter estimation algorithm, which is an algorithm based on radon transform. With respect to the results of the first parameter estimation was carried out after the Wiener filter, followed by image restoration.In the arrangement of the whole structure, the first complex traffic scenarios for image enhancement technology and some of the key technologies will be used in the vehicle license plate recognition elaborate. And introduces the key technologies in image enhancement technology and license plate recognition technology will be used for matching image segmentation algorithm and text, etc. were studied and experimental simulation, and finally, the complex traffic scene image enhancement technology in specific applications license plate recognition process, specifically including image enhancement algorithm in license plate recognition, fuzzy scale estimates based image deblurring fuzzy parameters and simulation experiments demonstrate through practical experiments in this study can be a good solution to the current vehicle license plate recognition exist part of the problem, the program for practical application has a strong reference value.