Multi-modal heat source identification and three-dimensional space positioning system and method and storage medium

Abstract

The invention discloses a multi-mode heat source identification and three-dimensional space positioning system and method and a storage medium. The method comprises the following steps: acquiring respective information of an input visible light image, a thermal image and a depth point cloud; and calibrating the visible light characteristic image, the thermal image and the depth information in theinput device, enabling color images with different depths of field to correspond to calibration of the thermal image, and eliminating information dislocation after double-light fusion, wherein the temperature data on the thermal image coordinates correspond to the depth information, correcting the detected temperature deviation, finally obtaining the calibration method and device of the internal and external parameters of the sensor, and generating a fused complete data set / relationship, thereby obtaining the corresponding multi-dimensional data in space; Performance is improved in the aspectsof moving fire source recognition and position positioning, moving non-contact human body temperature detection, accurate alignment of thermal image data and image data in a larger distance interval,object temperature measurement accuracy in dynamic distance change and the like.

Description

technical field [0001] The invention relates to the fields of heat source tracking, thermal imaging, visual perception, image recognition, and image positioning, and more specifically relates to a method for multimodal spectral information identification and three-dimensional space positioning. Background technique [0002] At present, thermal imaging is the main heat source image recognition, and there are many applications, such as fire detection, heat source detection, thermal infrared night vision, etc., but there are few technologies that can accurately integrate thermal imaging, visible light, and depth information. [0003] The existing pure thermal imaging image recognition technology has incomplete information and small business application scenarios. It only has temperature information and lacks visible light information. It is inconvenient to judge objects, such as human faces, license plates and other information. [0004] The existing visible light (including or...

AI Technical Summary

Problems solved by technology

[0002] At present, thermal imaging is the main heat source image recognition, and there are many applications, such as fire detection, heat source detection, thermal infrared night vision, etc., but there are few technologies that can accurately integrate thermal imaging, visible light, and depth information.

[0003] The existing pure thermal imaging image recognition technology has incomplete information and small business application scenarios. It only has temperature information and lacks visible light information. It is inconvenient to judge objects, such as human faces, license plates and other information.

[0004] The existing visible light (including ordinary infrared light) object recognition technology is difficult to accurately identify some specific objects, such as fire, but thermal images can quickly make judgments through temperature recognition

[0005] The existing dual-light fusion technology of visible light and thermal infrared lacks distance information, especially the depth information within the overlapping viewing angle range of thermal image and visible light camera, which leads to the matching of thermal image data and visible light pixels, and the temperature value of thermal image data varies with There is a large measurement error for distance changes

Images