A system and method for detecting video moving objects crossing the line

Abstract

The invention relates to a system and method for detecting a video moving target crossing a line. The method comprises the following steps: inputting a monitoring video, drawing an alarm line segment in the monitoring video; Draw the target frame at the line segment; automatically select the detection area according to the size of the target frame and the alarm line segment, perform background modeling on the detection area, identify the target according to the background of the detection area, and track the target in the detection area; when the tracked target crosses the alarm line segment , an alarm is issued. Compared with the prior art, the present invention can reduce false detection and missed detection, and improve algorithm efficiency.

Description

technical field [0001] The invention relates to the field of video technology, in particular to a system and method for detecting a video moving object crossing a line. Background technique [0002] The basic process of the current security video algorithm for crossing the moving target is as follows: 1. The user selects the position of the line segment in the scene; 2. Performs background modeling on the scene and extracts the moving foreground; 3. Tracks the moving target; 4. Judgments Whether the target crosses the line. [0003] Existing patents and literatures are aimed at improving the algorithm used in the background modeling process and the algorithm used to extract the moving foreground, but the research and development costs and time costs are relatively high. Due to the limitations of the current background modeling algorithm, complex scenes It is difficult to balance the false detection and missed detection problems, resulting in a greatly reduced system practic...

AI Technical Summary

Problems solved by technology

[0003] Existing patents and literatures are aimed at improving the algorithm used in the background modeling process and the algorithm used to extract the moving foreground, but the research and development costs and time costs are relatively high. Due to the limitations of the current background modeling algorithm, complex scenes It is difficult to balance the problems of false detection and missed detection, which greatly reduces the practicability of the system

Most of the false detections are caused by the instability of the modeling algorithm, and the lighting has a greater impact; most of the missed detections are due to excessive shielding of false detections, resulting in some targets being shielded

In addition, the complex algorithm also brings time consumption. The main time consumption of the line-crossing algorithm comes from the size of the detection area. If the user is asked to select the detection area, the user does not understand the algorithm, and he will choose a large one. Caused great waste, resulting in too long algorithm processing time, affecting user experience

Images