Vehicle License Plate Location And Characters Segmentation

In recent years,with China's rapid economic development of cities in the number of cars increases,the traffic management faces increasing pressure,an urgent requirement to establish an intelligent,efficient traffic management system.License plate recognition system to effectively alleviate the pressure,its rapid development greatly facilitates the management and maintenance of road infrastructure on the vehicle,has important practical significance.A basic license plate recognition system typically includes four processing modules.Three processing modules that license plate image preprocessing,license plate localization and license plate character segment are studied deeply.First,this paper preprocesses on the license plate image,including the image gray level,denoising,histogram equalization,contrast enhancement,tilt correction and binarization.Among them,the tilt correction is mainly based on Radon transform to detect the tilt angle of the plate image plate image correction.Secondly,at the license plate location stage,mathematical morphology is used to roughly locate the license plate candidate region to reduce the computational complexity of the algorithm in order to save storage space in system,and improved the vertical projection of the candidate plate region is used to further process it,then precise positioning get the real goal of the license plate.Finally,at plate characters segmentation stage,combining with license plate features of the intrinsic ratio and vertical projection realizes license plate character segmentation processing.The experimental results show that the algorithm can more accurately locate the license plate,separator character,able to meet actual needs.