Improved adaptive Gaussian mixture foreground detection method

Abstract

The invention provides an improved adaptive Gaussian mixture foreground detection method. The method comprises: firstly, performing learning by utilizing a Gaussian mixture model to form an initialized Gaussian mixture background model; secondly, for a new input video sequence, performing sampling at an interval of N frames, obtaining an image frame by utilizing weighted time-domain mean filtering, and performing background model updating by taking the image frame as an input of Gaussian mixture modeling; automatically determining whether background mutation exists in a current frame by Poisson distribution, if the background mutation does not exist, keeping normal sampling interval and learning rate, otherwise, reducing an interval frame number and increasing the learning rate, updating the background model, and extracting a current background frame; and finally, performing difference by utilizing the current frame and the current background frame, obtaining an adaptive threshold with a maximum entropy method, performing weighted mean on the obtained threshold, and performing foreground detection. According to the method, motion interferences of tree leaf shake, water ripples and the like in a video scene are effectively overcome, the calculation amount of frames is reduced through periodic sampling, and the timeliness is improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, and more specifically, to an improved adaptive mixed Gaussian foreground detection method. Background technique [0002] With the strengthening of public safety construction and the improvement of people's safety awareness, intelligent video surveillance has begun to receive people's attention and favor. This puts forward higher requirements for security systems and video surveillance systems. [0003] The intelligent video surveillance system detects, tracks, and recognizes targets in dynamic scenes by automatically analyzing the video recorded by the camera, and analyzes and judges the behavior of the target on this basis. It can not only complete daily monitoring but also respond in time when abnormal situations occur, and solve the problems of traditional monitoring such as heavy workload, low efficiency, and slow response speed. [0004] Moving target detection is a key step in an ...

AI Technical Summary

Problems solved by technology

This method is to simulate the complex background of reality by establishing multiple Gaussian models for each pixel in the image, which can effectively eliminate the influence of water ripples, camera shake, etc. Slow object detection buggy and more

In addition, foreign scholars proposed to adaptively determine the number of Gaussians required for modeling each pixel. Compared with the traditional mixed Gaussian background modeling, this method has improved operating speed, but it cannot eliminate water ripples very well. influences

Images