Perimeter alarm method based on two-dimensional laser radar and video analysis

Abstract

The invention provides a perimeter alarm method based on a two-dimensional laser radar and video analysis. The perimeter alarm method comprises the steps of acquiring a monitoring video image in a target region, performing corresponding point mapping with a three-dimensional model of the target region, and calculating a perspective projection matrix M; extracting a plurality of moving targets of the monitoring video image, obtaining two-dimensional coordinates, calculating corresponding three-dimensional coordinates to the moving targets, and determining whether the three-dimensional coordinate of each moving target satisfies a preset video alarm rule; using a two-dimensional laser radar for laser scanning of the target region, determining that a preset radar alarm rule is satisfied when the moving targets intersect with a laser scanning surface, and converting radar alarm information corresponding to the moving targets into a three-dimensional coordinate; and comparing the video alarm three-dimensional coordinate and the radar alarm three-dimensional coordinate, and sending out an alarm signal if the two are consistent. The invention can be used for intrusion alarm of abnormal moving targets in a perimeter defense region through bi-directional examination of video analysis results and radar alarm detection results.

Description

technical field [0001] The invention relates to the technical field of perimeter alarms, in particular to a perimeter alarm method based on two-dimensional laser radar and video analysis. Background technique [0002] The video-based target detection part is based on computer vision analysis technology to classify and identify target objects or use foreground and background separation for current extraction, and detect restricted areas or mixed lines for separated image targets to realize alarms. Due to problems such as the contrast between the target and the background, the texture characteristics of moving targets, the shaking of leaves, light, and shadows, there will be many false positives and false negatives in the current video analysis method, which is generally difficult to meet the needs of practical applications. For example, based on pressure sensors or vibration sensors, etc., due to the insensitivity of pressure detection, remote vibration and other problems, a ...

AI Technical Summary

Problems solved by technology

The current method of video analysis will have many false positives and false positives due to the contrast between the target and the background, the texture characteristics of the moving target, the shaking of the leaves, the light, the shadow, etc., and it is generally difficult to meet the actual application requirements.

For example, based on pressure sensors or vibration sensors, etc., due to the insensitivity of pressure detection, remote vibration and other issues, a certain degree of false positives and false negatives will be caused.

[0003] In addition, relying on infrared, microwave and other equipment on the market, there will also be many false positives and negative negatives due to light changes, temperature changes, leaf occlusion, beam reflection and other problems, and the usability is not high.

Images