|
License plate recognition system is based on digital image processing and character recognition technology, an intelligent traffic management system , which is mainly used in automotive electronic billing areas, such as automated highway toll system, parking fees management system, greatly improved its efficiency . License plate recognition system can be divided into plate location, character segmentation and character recognition three parts, of which the license plate positioning plate recognition system is a key first step. License plate positioning system 's goal is to extract from the original plate image area of the vehicle license ( or coordinate ) , for the next character segmentation and recognition . However, due to weather, light , car models , license plate hanging position , and stains wear various objective reasons for plate positioning accuracy rate had a negative impact , resulting in false consciousness or even rejection plate. Therefore, how to accurately locate the license plate license plate recognition system has been the focus of the study , with high economic value and practical value. In this paper, a detailed study abroad representative license plate recognition system , based on the license plate location issues in-depth study and discussion . First, the license plate positioning system more commonly used image processing method for the study and simulation , respectively, image preprocessing, edge detection test several methods for programming , focusing on the color edge detection and positioning system according to the needs of the license plate carried out the necessary improvements. Secondly, the introduction of the basic operations of mathematical morphology and other related knowledge , and adopt a new horizontal and vertical structural elements , the use of the structural elements on the edge detection of binary image expansion, corrosion and a series of morphological operations , to obtain a number of plate candidate region ( coordinates). Which explores the structural element shape , size selection and the importance of structural elements used in this paper are representative . Finally, the plates own characteristics , such as the license plate area , aspect ratio , etc. of the plate candidate region comprehensive analysis and comparison and selection, accurately locate the position of the plate ( or coordinates ) , and returns to the original image is divided plate , complete positioning plate .
|