3D human body reconstruction method based on infrared thermal imaging
A technology of infrared thermal imaging and human body, which is applied in 3D modeling, image enhancement, image analysis, etc., can solve the problems that do not meet the needs of real-time infrared image human body model reconstruction, and achieve the goal of increasing contrast and visual effect, and improving enhancement effect Effect
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Publication Date
- 2021-07-13
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] A 3D human body reconstruction method based on infrared thermal imaging belongs to the technical field of computer vision processing, and in particular relates to a 3D human body reconstruction method in image processing. Background technique
[0002] With the development of computer vision processing technology, 3D reconstruction technology is mainly used in augmented reality and virtual reality technology. The imaging principle of the infrared thermal imaging system is to convert the infrared specific band signal of the thermal radiation of the object into an image. The imaging of the infrared thermal imaging system is not affected by changes in illumination, smoke, fog, trees and other obstacles. It is suitable for target recognition and detection in dark and harsh weather environments. It has a strong role in search and rescue, military, security, driving assistance and other fields. Value. With the development of machine vision and artificial in...
Examples
Embodiment 1
[0054] Such as figure 1 and 5 As shown, the 3D human body reconstruction method based on infrared thermal imaging includes the following steps:
[0055] Step 1, making an infrared human body image;
[0056] Step 2. Enhance the infrared human body image, control the distribution of the lateral suppression coefficient by changing the bimodal Gaussian distribution function, and control the contrast by changing the gray scale of the image;
[0057] Step 3. Make a 3D human body model for infrared human body image registration, and perform SMPL-X parameterization on the 3D human body model to obtain human body shape, human body posture, and human face parameters. A single infrared human body image is used as input, and the output is human body shape parameters , the multidimensional vector of the combination of human body posture parameters and human face parameters, each model parameter is used as a real label, and paired to generate a data set;
[0058] Step 4. Construct the hu...
Embodiment 2
[0063] On the basis of Example 1, in step 1, an infrared thermal imager is used to collect an infrared human body image, and the infrared thermal imager is deployed in a dark environment without light at night and in a harsh environment with smog, and the collected infrared human body image Convert to a single-channel digital image format.
Embodiment 3
[0065] On the basis of embodiment 1, in step 2, the suppression coefficient refers to the distance between pixel units:
[0066]
[0067] In formula 1, x1, y1, x2, y2 are the vector coefficients of pixel units (x1, y1), (x2, y2);
[0068] bimodal Gaussian distribution function It is found on the basis of the Gaussian distribution function, and the specific form is:
[0069]
[0070] σ in formula 2 1 , σ 2 is the width of the bimodal Gaussian function, μ 1 , μ 2 is the location parameter of the bimodal Gaussian function, β, β 1 , β 2 , σ 1 , σ 2 , μ 1 , μ 2 are constants.
[0071] When β=1, β 1 = β 2 = 2, μ 1 =μ 2 =0, π=3.14, the form of the isotropic bimodal Gaussian distribution function is:
[0072]
[0073] Converting the isotropic bimodal Gaussian distribution function in formula 3 into the heterotropic bimodal Gaussian function, we get:
[0074]
[0075] Rotate the function in Equation 4 counterclockwise by angle α, define the coordinates bef...