Phase unwrapping method based on least square method

Abstract

The invention discloses a phase unwrapping method based on a least square method. Firstly, a two-dimensional wrapped image which contains three-dimensional information of a to-be-detected object and has M*N data points is obtained through shooting with a camera device with a corresponding image processing method; then, four directions including horizontal x, vertical y and p and q forming phi / 4 with the x and y directions clockwise are set according to the two-dimensional wrapped image; a two-dimensional light recovering field is set according to a two-dimensional wrapped phase value phi, and primary shearing is performed in each direction; then, a new two-dimensional light recovering field is set in the direction perpendicular to the direction after primary shearing, and secondary shearing is performed; finally, first difference in four directions of the two-dimensional wrapped image after secondary shearing is introduced into a Poisson equation, the Poisson equation is solved through discrete cosine transform, and the real phase phi of the wrapped image is obtained. The phase unwrapping method has the characteristics of being wide in application range, high in unwrapping speed, high in anti-interference capacity and the like, and the unwrapping phase operation speed is greatly increased.

Description

technical field [0001] The invention relates to a method for unwrapping a phase wrapping phase in structured light three-dimensional measurement, and belongs to the technical field of structured light three-dimensional measurement. Background technique [0002] With the rapid development of science and technology, in the process of industrial production, the detection of defects, dimensions, and free-form surfaces on the surface of objects is becoming more and more frequent, and the requirements for detection accuracy are also continuously improving. Classical contact detection methods are no longer available. No matter how applicable, in order to solve this problem, structured light three-dimensional measurement technology came into being. Structured light three-dimensional measurement technology is a non-contact, high-precision projected structured light or natural light surface topography measurement technology, which has been widely used in modern industries such as elec...

AI Technical Summary

Problems solved by technology

However, in practical applications, due to too many directions, the phase transformation in the local area is too fast, and the problem of undersampling will occur, and then the phase error will be introduced when calculating the phase difference, resulting in the failure of unwrapping.

[0004] Transverse shear least squares phase unwrapping algorithm (Qian Xiaofan, Li Bin, Li Xinghua, et al. Improvement of transverse shear least squares phase unwrapping algorithm[J]. China Laser, 2012,39(11):193-197 .) Solve the undersampling problem caused by the rapid change of the phase in space, but because the phase estimation is based on the two-way (horizontal and vertical) least squares method, it will pass through the phase inconsistency area during the unwrapping process, resulting in The phase error propagates in space, and there is also the defect of smoothing the real phase, which makes the wrapped phase tend to have a sharper peak and a slower steep slope

Images