Moving target detection method and device based on improved Gaussian mixture model

Abstract

The invention discloses a moving object detection method and device based on an improved Gaussian mixture model, and the method comprises the steps: obtaining a video sequence image, carrying out thepartitioning of each frame of image in the video sequence image, and obtaining a plurality of pixel blocks; constructing a Gaussian mixture model according to the pixel value of each pixel block in each frame of image; matching the pixel value of each pixel block in the current frame image with the Gaussian mixture model; updating the Gaussian mixture model according to a matching result; and matching the pixel value of each pixel block in the current frame image with the updated Gaussian mixture model to detect a moving object. According to the method, the video image is partitioned, and thepixel values of the pixel blocks are matched with the Gaussian mixture model, so that the calculation amount of model matching is reduced, and the moving object detection efficiency is improved; and the updating rate of the background model can be adaptively valued according to the number of frames through which the current video flows.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a moving target detection method and device based on an improved mixed Gaussian model. Background technique [0002] In recent years, with the vigorous development of computer and Internet technology, people's daily life is increasingly inseparable from various smart devices. For example, using computer vision instead of human eyes to identify targets in videos can not only greatly improve work efficiency, but also solve many problems that are difficult for human eyes to solve. Therefore, the detection algorithm of moving targets in videos is becoming more and more important. Now The moving target detection algorithms mainly include frame difference method, background difference method and background modeling method. Among them, the background modeling method is the most widely used method. It models the background of the image, compares the current image with the backgr...

AI Technical Summary

Problems solved by technology

If the value of the update rate is too small, the proportion of the Gaussian distribution that can match the pixel in the current video is too small, the Gaussian model under this parameter will not work well, and the background update speed will be reduced. , to extend the time for new background modeling; on the contrary, if the value of the update rate is too large, the update speed of the background model is too fast, and the pixels in the current video can always be successfully matched with the Gaussian model, so it will not be possible to continue Moving object detection, there will be a lot of noise

However, in the existing mixed Gaussian model, the value of the update rate is generally a fixed value, which cannot meet the needs of the actual situation.

Images