Ultra-wideband radar human body posture reconstruction method based on point cloud
A technology of ultra-wideband radar and human body posture, which is applied in the direction of radio wave reflection/re-radiation, utilization of re-radiation, measurement devices, etc. It can solve the difficulties in displaying 3D imaging results of 3D ultra-wideband radar, multi-computing resources, large amount of data, etc. problems, to achieve the effect of enriching the detailed information of human targets and expanding the application scenarios
Inactive Publication Date: 2021-06-29
NAT UNIV OF DEFENSE TECH
View PDF4 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the 3D imaging results of human targets in 3D ultra-wideband radar systems are difficult to display, and are not conducive to everyone's understanding, and the readability is poor
Some scholars use deep learning tools to directly reconstruct the pose of the human target from the 3D imaging results of the human target, but this method requires a huge amount of data to be processed and consumes a lot of computing resources
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0020] The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments.
[0021] Refer to attached figure 1 , a kind of UWB radar human posture reconstruction method based on point cloud proposed by the present invention is specifically realized through the following steps:
[0022] Step 1, the 3D UWB radar transmits radar signals, receives echo signals, and performs rapid 2D imaging.
[0023] In this embodiment, the three-dimensional UWB radar transmission signal is denoted as S T (t), the received echo signal is denoted as S R (t), where t represents the time, using the back projection (BP) method to receive the echo signal S R (t) Perform range-azimuth rapid two-dimensional imaging of the entire detection scene to obtain a two-dimensional imaging result I(x, y), where (x, y) is any pixel in the range-azimuth imaging area.
[0024] In this embodiment, the operating frequency of the three-dimensional u...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides an ultra-wideband radar human body posture reconstruction method based on point cloud, and belongs to the field of radar signal processing. In allusion to the problem of poor readability of an ultra-wideband radar imaging result, the method comprises the steps of firstly, receiving a radar echo signal, and carrying out distance-azimuth two-dimensional imaging on a detection scene; preprocessing a two-dimensional imaging result to lock a target position; then performing three-dimensional imaging on the area where the target is located, constructing a scattering intensity contour surface, and performing feature point sampling to obtain point cloud data of the human body target; inputting the point cloud data into a trained neural network to reconstruct the posture of the human body target; and finally, fusing the detected target position with the posture reconstruction result to obtain the posture of the target at the real position. According to the method, a radar signal processing method and a deep learning method are combined, the posture of the human body target is quickly reconstructed from the radar point cloud data by using the lightweight neural network, and richer human body target detail information is provided for system users.
Description
technical field [0001] The invention relates to the field of radar signal processing, in particular to a point cloud-based ultra-wideband radar human posture reconstruction method. Background technique [0002] Human pose estimation plays a vital role in a wide range of applications in human-computer interaction, robotics, surveillance, and virtual reality, and has received increasing attention. At present, most of the research on human pose estimation algorithms is based on optical sensors. Scholars have designed various pose estimation methods according to different input data, such as single optical RGB images, continuous frame video images, and RGBD images. However, the use of optical sensors depends on a good working environment. They cannot work in dark and occluded environments, and their use in complex urban scenarios such as anti-terrorism, stability maintenance, and emergency rescue is limited. [0003] Low-frequency ultra-wideband radar is a target detection sens...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G01S13/42
CPCG01S13/42
Inventor 金添宋永坤戴永鹏宋勇平周小龙
Owner NAT UNIV OF DEFENSE TECH