Garbage throwing behavior detection method based on urban management monitoring video

Abstract

The invention provides a garbage throwing behavior detection method based on an urban management monitoring video. The method comprises the steps of firstly carrying out the pavement garbage detectionof the urban management monitoring video at intervals, rechecking the monitoring video in the former time if the pavement garbage is found, and analyzing the monitoring video to find a pedestrian whothrows garbage; and if no pedestrian throws the garbage, searching the pedestrian throwing the garbage in the video in the front of the surrounding camera, so as to detect the behavior of throwing the garbage in the monitoring video.

Description

technical field [0001] The invention belongs to the field of video processing, and in particular relates to a littering behavior detection method based on urban management monitoring video. Background technique [0002] With the rapid development of science and technology, the acceleration of urbanization and the rapid improvement of people's living standards, the garbage and waste generated in daily life and production have also increased rapidly. The tension of community residents is not strong enough, and littering often occurs. situation, and a large amount of garbage discharge has indeed caused serious harm to community residents. [0003] In the prior art, the littering behavior is mainly detected by manual on-site discovery and snapshot, but this behavior is time-consuming and labor-intensive, and the detection efficiency is not high. Technically, the research on video target detection is not hot enough compared to the target detection in the image field, and because...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the littering behavior is mainly detected by manual on-site discovery and capture, but this behavior is time-consuming and labor-intensive, and the detection efficiency is not high

Technically, the research on video target detection is not hot enough compared to the target detection in the image field, and due to the randomness and transient nature of littering behavior, at the technical level, it is not only necessary to identify the garbage itself but also to find out who is littering. This brings a lot of inconvenience to video object detection for littering behavior detection; in addition, in some cases, the garbage will float to other areas due to wind or other reasons, which also leads to the detection of littering objects. There is trouble, so there is an urgent need for a method and system for automatically identifying littering behavior in surveillance video

Images