Regional average value kernel density estimation-based moving target detecting method in dynamic scene

Abstract

The invention discloses a regional average value kernel density estimation-based moving target detecting method in a dynamic scene. The method comprises the following steps of: firstly, initializing a background model; secondly, building a time and space background model for describing the dynamic complex scene by using a training sample in a background modelling process and considering the time sequence characteristics of pixel points in a video frame and the space characteristics in the adjacent regions of the pixel points; thirdly, continuously updating the background model by using the new video frame sample in a moving target detecting process; fourthly, adapting to the instantaneous background change by the regional kernel density estimating method and adapting to the continuous background change by using single Gauss background model, wherein the combination of the two models can fast and accurately adapt to the continuous change of the background and increases the executing efficiency of the method at the same time; and finally performing a foreground detecting method by providing an adjacent region information amount-based method so as to further remove noise points and inanition of a moving target in the background region in the detecting process and more completely extract the moving object in the foreground. The method can be widely applied to alarming the suspicious moving target in an intelligent monitoring system in an outdoor scene or a prohibited military zone and has wide market prospect and application value.

Description

technical field The invention relates to a moving target detection algorithm in an intelligent video monitoring system, in particular to a moving target detection algorithm based on area mean kernel density estimation in a dynamic scene. Background technique Moving target detection is the basis of target recognition, tracking and later understanding of object behavior, and is a key issue in intelligent monitoring systems. In video surveillance systems, fixed cameras are used to monitor the target area, so most detection algorithms assume a static background and only consider background changes such as illumination and shadows. But in real complex scenes, the assumption of static scenes does not exist, for example: fountains, swaying leaves, water waves, etc. all lead to dynamic changes in the background. In addition, the camera may shake slightly due to various reasons. Therefore, moving object detection algorithms in dynamic scenes have become a research hotspot in the fi...

AI Technical Summary

Problems solved by technology

To sum up, the existing detection algorithms contain the following problems: the pixel-based background model update method only considers the temporal characteristics of the background, and in the case of dynamic scene background changes, if a certain function or multiple functions are used The mixed method cannot accurately describe the time-domain characteristics of pixels. If the kernel density estimation method is used, although it does not need to estimate and determine the function, it directly uses the sample to describe the background, but its execution efficiency is low during the update process, which cannot meet the real-time requirements. ; The method based on the region or the overall video frame only considers the spatial correlation of the background, and its execution efficiency is high, but the detection accuracy is lower than the pixel-based detection method

Images