| With the rapid development of the national economy and the gradual improvement of people’s living standards,computer technology and multimedia technology based on image information have developed rapidly.The field of traffic supervision is more and more focused on using video and image data to solve practical problems,such as: road traffic statistics,vehicle violation judgments,and so on.However,due to the adverse weather environment,low illumination or excessive exposure to the external environment,often the images taken are too dark and the details are seriously lost.Therefore,how to effectively solve this problem becomes the top priority in the field of traffic supervision.Because the traditional histogram equalization in the process of processing at the expense of part of the gray level in exchange for the expansion of contrast,it will lead to the loss of some image details and easy to form false contours.In order to deal with this shortcoming,this paper proposes based on A histogram enhancement algorithm at the edges.The gray level distribution of the image enhanced by this algorithm is more uniform,which can better expand the dynamic range of the image.The algorithm first converts the image to the HSV color space and extracts the color information of the V channel.Next,the edge image of the V channel image is completely histogram equalized to obtain the edge image mapping function.Finally,this function is applied to the global image.The second half of the article mainly analyzes the detection part of the vehicle.Through comparative analysis,this article uses a more efficient and flexible HOG feature and SVM classifier to detect the vehicle.The work mainly includes the pre-processing of samples,the Extraction,training of the SVM classifier,etc.finally combined with the image enhancement algorithm proposed in this paper to enhance the original low-illumination and low-resolution images,and classify and detect the enhanced images.The final experimental results show that the original image after image enhancement will have better expressiveness in subjective vision,and the image contrast can be better enhanced,especially in terms of the contrast and brightness of the edge contours of objects in the image At the same time,it can also enhance the details of the color image while ensuring the naturalness of the color of the image.In terms of vehicle detection,the vehicle detection recognition rate after image enhancement has been improved to a certain extent.This research result is of great significance for the realization of intelligent transportation. |
