A system and method for measuring human body parameters based on segmented three-dimensional reconstruction

Abstract

The invention relates to a system and method for measuring human body parameters based on segmented three-dimensional reconstruction. The system includes a horizontal motorized turntable, a depth sensor and a laptop computer. The operation steps of the method are: (1) collecting depth images, (2) segmentally reconstructing a three-dimensional model, (3) visualizing the model, (4) eliminating creases, and (5) measuring and calculating data. The invention converts the measurement of the volume of the human body and the circumference of different parts into voxels for calculation and processing by constructing a three-dimensional model of the human body. The method is easy to operate and the calculation results are accurate, which can better solve the problems of complicated and large measurement deviations in the current manual measurement of human body parameters, save the time of the measurement personnel, and avoid repeated measurements due to inaccurate results . The invention has an intuitive and beautiful interface, which provides great convenience for relevant detection departments.

Description

technical field [0001] The present invention relates to a method for segmental three-dimensional reconstruction and measurement of an object represented by the human body, and in particular to a system and method for measuring human body parameters based on segmental reconstruction. The human body is segmented and scanned by Kinect and completed in several threads. For image registration and fusion, it is recommended to divide into three segments, and then fit multi-segment data to eliminate boundary creases, and establish a 3D model that retains details and has no holes. It also expands the function of calculating the volume of the model by using voxels, and the function of measuring height, arm length, girth and other data combined with human bones. A system consisting of a depth sensor and a laptop computer interface was built. Background technique [0002] In computer vision and computer graphics, 3D reconstruction technology is the process of obtaining the shape and ap...

AI Technical Summary

Problems solved by technology

However, traditional reconstruction equipment based on the KinectFusion algorithm is mainly aimed at the 3D reconstruction of static scenes. In theory, it can only allow objects to rotate smoothly and at a uniform speed in the scene without losing tracking, which meets the reconstruction requirements.

However, in specific application scenarios, such as reconstructing the human body, it is difficult for the person to not lose tracking during the rotation in front of the camera, and the quality of the obtained object model is relatively poor.

In addition, the estimation of the camera pose in the algorithm is more dependent on the geometric features of the objects in the scene, and the difficulty of camera tracking will also increase.

Images