Target object identification method and device based on infrared thermal imaging system

Abstract

The invention discloses a target object identification method and device based on an infrared thermal imaging system. The method comprises steps that infrared image data shot by the infrared thermal imaging system is acquired, and the infrared image data comprises the pixel point position and a pixel point gray value; the acquired pixel point gray value in an alarm region is compared with a preset threshold of the alarm region, if the pixel point gray value is larger than the preset threshold, the next step is performed; the identity information of break-in targets of the infrared image data and the quantity can be acquired according to the pixel point position, and the corresponding alarm information is generated. The method is advantaged in that precise control on alarm output can be realized, target objects triggering alarm are locked through setting a vector scope corresponding to the alarm target objects when a user need aim at only one type or several close alarm target object types, other target objects do not trigger alarm, so the user is more convenient to use, an application scene of a system is made to be wider, and an error report rate is smaller.

Description

technical field [0001] The invention belongs to the technical field of computer vision and machine vision, and in particular relates to a method and device for object recognition based on an infrared thermal imager. Background technique [0002] Intelligent video monitoring system is an application direction that has developed rapidly in the field of computer vision and machine vision in recent years, and has been studied more. This system can use computer or machine vision technology to process, analyze and even understand the collected image signals, and further operate and process the video surveillance system on this basis, and make different responses according to different situations. Make the intelligent video surveillance system have better intelligence and wider application fields. In recent years, intelligent video monitoring systems mainly involve knowledge in machine vision, computer vision, image processing, artificial intelligence and pattern recognition. [...

AI Technical Summary

Problems solved by technology

[0004] At present, most of the intelligent video monitoring devices on the market are based on the application of visible light, and most of them use the color distribution or other characteristic information to process the image. Currently, the commonly used algorithms for moving target monitoring include background difference method, Inter-frame difference method and optical flow algorithm, etc., but these algorithms will have great difficulties in practical applications, because these algorithms are susceptible to the huge complexity of target motion in the actual environment and the complexity and particularity of video data. Among them, the size of the target object, the speed of movement, the trajectory of movement, the similarity between the color of the target and the background color, and the stability of the background are all important influencing factors.

[0005] On the other hand, various intelligent monitoring and alarm methods based on various infrared thermal imaging cameras in the market are different, but the current intelligent monitoring and alarm systems all have a large alarm range, and the alarm conditions are uncontrollable, which has caused High false alarm rate, and inability to accurately identify the status of targets entering the warning range

Images