Study On Auto-focusing Method Of Traffic Camera Based On Image Processing

As one of the main parts of the intelligent transportation system,the automatic focusing method based on image processing has played an important role.It can provide the basis for the traffic monitoring command center to formulate reliable and effective scientific pre-control programs.This article studies the development history of autofocus technology at home and abroad,and analyzes that traditional focusing methods often have problems such as poor focusing accuracy and poor resolution.However,the auto-focusing method of traffic camera based on image processing directly analyzes the quality of the image according to the collected image,so as to determine the current imaging state,and then complete the autofocus operation.It not only achieves high focusing accuracy,but also far exceeds the former in terms of sensitivity and clarity.Therefore,in order to improve the intelligent level of traffic camera auto-focusing technology,this paper studies the auto-focusing method based on image processing.In order to reduce the influence of noise and exposure on the focusing of traffic cameras,this paper studies the solution of two-dimensional histogram equalization to deal with the interference of light intensity,so that the image with strong exposure is effectively preprocessed;when subjected to the noise size is affected,median filtering is used to effectively reduce the effect of noise on auto focus.When improving the construction of the focusing window,in order to reduce the background interference in the target area and reduce the operation time of focusing,this paper studies a focusing window method of the biological visual perception model,through the DOG filtering of the image information,DCT/PCT transformation,normalization of feature maps,image binarization,and other processing steps complete the improved construction process of the focusing window in different focusing states.Since the selection of the focus evaluation function directly affects the success or failure of the entire focus of the system,this paper studies an optimized SML focusing evaluation function and an adaptive threshold segmentation idea to effectively combine the strategy,in the original SML function image edge information Based on the above,consider adding two gradient information values in the diagonal direction,so as to improve the sharpness value and stability of the evaluation function.For the algorithm search strategy of focus function,this paper studies an optimizedclimbing search method(a combination of coarse and fine).This search algorithm can not only show high sensitivity and good anti-noise performance,but also the precise positioning of the focus position of the image is achieved,so the focusing effect is ideal.Finally,in order to realize the automatic focusing method based on image processing,the overall hardware structure and software flow of the focusing system are designed in detail,and a large number of theoretical experimental data verifications are performed using MATLAB simulation software.Finally,by comparing the image effects obtained before and after the simulation experiment,it is confirmed that the system is feasible and can meet the practical application requirements.