Research And Implementation Of License Plate Positioning System In Foggy Environmen

In the modern intelligent transportation system,electronic license plate identification has become its core and is widely used in people’s life.The license plate identification system generally includes three main functions: license plate positioning,character division and character identification.As the first link of the intelligent transportation system,license plate positioning will directly determine the result of subsequent division and identification operations.However,in the real scene,affected by the haze weather,the license plate images collected by the capture device appear contrast and sharpness degradation,which greatly reduces the accuracy and robustness of the license plate location.In order to further improve the accuracy and robustness of license plate location under foggy conditions,this paper will study image dehazing technology and license plate location technology.The main research work of this paper is as follows:Image dehazing: Aiming at the problems that the processing results of the dehazing algorithms such as the dark primary color prior are distorted in strong light areas such as the sky and the body-in-white,and the color of the processed image is darker,the original image is first regenerated using local entropy.It is divided into two parts,the sky and the non-sky,and then their transmittances are calculated respectively,and the transmittance of the distortion domain is corrected.Finally,the haze-free image is restored using the dark primary color prior model;and the restored image is restored by the improved enhancement method.Brightness adjustment has been added.The actual measurement results show that the proposed method has better clarity and brightness in the recovered image,which provides a better original image source for later license plate location work,thus greatly improving the accuracy of license plate location.License plate localization: Aiming at the foggy environment,a license plate localization algorithm based on the combination of color features and single-player phase consistency edge detection is proposed.First,the restored license plate image is mapped to the HSV color space;then,according to the geometric priori characteristics of the license plate,each color component is thresholded and fused into an HSV binary image,and then positioned by morphological filtering and projection analysis.license plate.Experimental results show that the method can precisely locate the most common blue-bottomed license plates.When the color feature-based method fails to locate the license plate whose color is close to the color of the car body,the single-play phase-consistency edge detection method is used to locate the license plate.Firstly,the phase consistency model is improved,that is,the Riesz transform is used instead of the Hilbert transform to obtain a single phase consistency model.Then,the edge of the license plate image is extracted by this model,and the result is binarized.Finally,the license plate is realized by combining morphological filtering.position.The experimental results show that the algorithm can locate the license plate accurately and robustly without being affected by the contrast and brightness of the image when extracting the edge features of the image.Finally,for the successfully located license plate,if there is an inclined situation,a low-rank texture transformation invariance model is used to correct it.The experimental results show that the correction effect of the proposed method is very ideal,and it has strong anti-interference ability.In addition,a license plate positioning system in foggy environment is built based on MATLAB GUI,and the simulation test shows that the system can easily realize human-computer interaction and data visualization.