Fast Foot Measurement System Based on 3D Measurement Network

Abstract

A multi-node 3D foot measurement network system is proposed based on digital speckle correlation method. For each node, two cameras build a binocular system, the corresponding points between the left and right cameras are established through the spatial correlation method. In order to improve the search speed of corresponding points, the epipolar and parallax constraints are adopted. The sub-pixel corresponding point location is achieved by Newton-Raphson iterative optimization method to improve the reconstruction accuracy, and the single-node depth data is finally reconstructed based on the stereo vision principle. Multi-view depth data is matched by means of the measurement network calibration technology. The whole foot data acquisition takes less than one second, which has a great commercial application prospect.