Large field of view fringe projection visual three-dimensional measurement system and method

Abstract

The invention relates to optical three-dimensional detection technology. It aims to design a large-field distributed fringe projection visual three-dimensional measurement system, expand the measurement range of the fringe projection visual measurement system, realize multi-camera synchronous distributed acquisition, and quickly obtain global three-dimensional point cloud data of objects. . The present invention provides a large field of view fringe projection visual three-dimensional measurement system and method, including a laser projector, 4 complementary metal oxide semiconductor CMOS industrial cameras, a computer and a target. The laser projector is a high-brightness short-focus laser projector with a brightness greater than 3000lm , the projection ratio is less than 1, and the projected stripes can cover the effective measurement range of all cameras; the 4 cameras are evenly distributed in the entire field of view, and the camera acquisition range covers the entire field of view; the projected stripes of the projector and the images collected by the cameras are synchronized by software timing control. The invention is mainly applied to three-dimensional detection occasions.

Description

technical field [0001] The invention relates to optical three-dimensional detection technology, and proposes a novel large-field distributed fringe projection three-dimensional measurement system, a novel planar target, and a novel system global calibration method using the target. Background technique [0002] The fringe projection three-dimensional measurement technology uses a projector to project fringes to the measured object. The surface shape of the object will modulate the fringes, and then the modulated fringes will be collected by the camera, and finally the three-dimensional point cloud of the object surface will be back-calculated. The measurement accuracy of the fringe projection measurement system mainly depends on its calibration accuracy. At present, the mainstream system calibration methods include photogrammetry based on matrix transformation and polynomial fitting method. The application of photogrammetry completes the system calibration by establishing th...

AI Technical Summary

Problems solved by technology

Solve the problems that the traditional fringe projection system cannot obtain global 3D point cloud data, the calibration in large field of view measurement is complicated, the efficiency is low, the accuracy is poor, and it does not meet the needs of fast and precise visual 3D measurement of large field of view, etc.

Images