Research And Application Of Moving Target Detection Method Based On Background Modeling

With the rapid development of computer technology,as a key technology of machine vision,moving target detection has become one of the hot topics in this field.Moving target detection technology has broad application prospects in the fields of intelligent transportation,security monitoring,and military equipment.Such as vehicle detection of intelligent transportation systems,behavior monitoring of security systems,and precise guidance of military equipment.In this paper,the moving target detection method based on background modeling is deeply studied,and two improved background modeling target detection algorithms are proposed.The main research contents are as follows:(1)The basic theory and algorithm principle of five classical background modeling algorithms are studied.The detection results of five algorithms in static scene,dynamic scene and moving target shadow scene are respectively given by experiments,and the detection results are quantitatively analyzed and qualitatively evaluated.(2)An improved Vibe algorithm is proposed for the shortcomings of Vibe algorithm in dealing with moving target "ghost",moving target shadow and dynamic background.In the background modeling stage,a foreground counter is set for each pixel to eliminate the target "ghost" caused by the changes of moving state of the moving target;In the foreground detection stage,a shadow removal algorithm based on the HSV color space combined with a gradient operator is used to remove the moving target shadow;In the post-processing stage of detection results,an image morphology open operation filter is used to enhance the robustness of the algorithm to dynamic background.The experimental results show that the improved Vibe algorithm is significantly improved in terms of detection accuracy compared with the original algorithm.(3)The method of camera motion compensation is introduced,for the problem of camera motion and algorithm time complexity,an improved Gaussian mixture background modeling algorithm based on motion compensation is proposed.In the background modeling stage,a motion compensation method based on KLT is used to obtain a motion compensation background model;In the foreground detection stage,the current pixel is first compared with its own background model,and then compared with the background model of its 8neighborhood pixels,the algorithm is speeded up without reducing the detection accuracy in this way;In the update stage of the background model,in order to adapt to the background changes caused by camera motion,an age value variable is set for each pixel.The experimental results show that the improved algorithm has a significant improvement in detection accuracy and running time compared to Gaussian mixture background modeling algorithm.