Residue detection method, device, equipment and computer storage medium

Abstract

The invention discloses a method, device, equipment and computer storage medium for detecting residual matter based on video monitoring. The method includes: constructing two Gaussian background models with different update periods of the background image; when the second update period is reached, saving the short background difference image corresponding to the short background image obtained and updating the short background image; when the first update period is reached, obtaining The long background difference image corresponding to the long background image is updated and the long background image is updated; multiple short background difference images obtained in the update cycle of the long background image are fused to obtain a fused image; the long background difference image and the fused image are intersected to obtain an intersection image ; The difference between the intersection image and the last saved retentate image is used as the retentate image; the connected domains in the retentate image are first marked and displayed on the read-in video frame. According to the embodiment of the present invention, slowly changing non-residue information in the retentate image can be filtered out to reduce false detection, and the accuracy of retentate detection is high.

Description

technical field [0001] The invention belongs to the technical field of video monitoring, and in particular relates to a method, device, equipment and computer storage medium for detecting residual objects based on video monitoring. Background technique [0002] Detained matter detection is mainly to detect the accumulation of stranded matter in the market safety channel, and when jamming and road occupation behaviors are detected, an alarm will be issued. In the field of retention detection, the techniques used mainly include the time difference method, optical flow method and background difference method. Among them, the background difference method is more sensitive to the influence of light and so on. [0003] In the existing background difference method, the main method is to extract the current frame and the background for difference, and then accumulate the difference results in order to find the lingering target. However, although this method can temporarily detect ...

AI Technical Summary

Problems solved by technology

However, although this method can temporarily detect stranded objects, it cannot filter out slow-moving non-residual objects and slowly changing light and shadows, resulting in false detections.

Images