Method for detecting heat source center in three-dimensional space

Abstract

The invention discloses a method for detecting a heat source center in the three-dimensional space, which belongs to the field of computer vision, and solves the problem of detecting the coordinates of the heat source center in the three-dimensional space. The method comprises the steps of firstly, calibrating an ideal space center as a space coordinate origin; rotating two angles at will; shooting an infrared picture at each of the two positions; then extracting the R channel values of the two infrared pictures to obtain a new picture; cutting the new picture according to the small checks ofa certain size, calculating an R channel average value of each small check, sorting the small checks according to the R values after calculation, calculating the size of a frame of a target heat source according to a heat source proportion condition, at the same time, obtaining the coordinates of the heat source on a two-dimensional image. After the coordinate of the heat source center on a two-dimensional plane is obtained, the depth information can be calculated according to the corresponding geometrical relationship, and finally the coordinate position of the heat source in the three-dimensional space is obtained.

Description

technical field [0001] The invention relates to the field of computer vision, and more specifically, to a method for detecting the center of a heat source in a three-dimensional space. Background technique [0002] In three-dimensional space, in order to accurately obtain the spatial coordinates of the center of the heat source, it is necessary to perform object detection in three-dimensional space. In the past thirty years, image recognition is mainly aimed at the recognition of digitized grayscale images. In recent years, it has become one of the new development directions of image recognition to recognize three-dimensional objects through distance images (or three-dimensional images, depth images). Distance data can be obtained from active or passive sensors, given in digital form, and the quality of distance data is gradually improving. The depth image data contains the depth information of the point, that is, the distance from the focal plane of the sensor to the targ...

AI Technical Summary

Problems solved by technology

Generally, laser cameras are used to obtain the depth data of three-dimensional objects, which can obtain accurate surface descriptions, but the equipment is relatively expensive, and the method based on depth data has a large amount of calculation.

Images