Fast phase unwrapping method, device, equipment and medium based on edge detection

Abstract

The present invention provides a fast phase unwrapping method, device, equipment and medium based on edge detection. The method includes: acquiring a deformed fringe image; wherein, the deformed fringe image is generated by an ideal grating fringe modulated by an object under test; using edge detection The operator identifies and locates the boundary of the wrapped phase map, and obtains the edge detection result of the wrapped phase map; corrects the abnormal edge points in the edge detection result, and maps the corrected edge points into the initial fringe series; optimizes the initial fringe series , and then use the optimized fringe series to calculate the absolute phase corresponding to the grating fringe image. The invention realizes the fast recovery of the absolute phase.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a fast phase unwrapping method, device, equipment and medium based on edge detection. Background technique [0002] Fringe projection profilometry is a full-field, non-contact, high-efficiency and high-precision optical three-dimensional measurement technology, which has been widely used in additive manufacturing, virtual reality, biomedical and industrial inspection and other fields. Fringe projection profilometry projects one or more fringe patterns with certain coding rules to the surface of the measured object through the projection device, uses the camera to collect the deformed fringes in real time and transmits the collected fringe images to the computer for processing, and finally according to The geometric relationship of the system recovers the 3D point cloud of the object. When the coded stripes touch the surface of an object, the stripes deform. The deform...

AI Technical Summary

Problems solved by technology

However, due to the need to project an additional fringe pattern, the time-phase unwrapping method is less efficient and not suitable for dynamic measurements

[0012] The phase unwrapping algorithm based on deep learning can improve the efficiency of phase unwrapping to a certain extent, but due to the need for a large number of training samples and complex training process in the early stage, the versatility and real-time performance are not high

[0013] In summary, the existing methods are difficult to balance the relationship between phase recovery efficiency and accuracy

Images