Three-dimensional surface type measurement method for single-frame color fringe projection based on deep learning

Abstract

The invention discloses a three-dimensional surface type measurement method for single-frame color fringe projection based on deep learning, and the method comprises the steps: inputting gray-scale fringe images in red, green and blue channels which comprise three channels and are color fringe images based on a model CNN of a convolutional neural network; projecting three 12-step phase shift stripes with different frequencies by adopting a projector, and generating training data required by the CNN by utilizing a phase shift (PS) method and a projection minimum distance method (PDM) to train the CNN; when the method is in use, inputting three channel gray scale fringe images of a color fringe image into the CNN to obtain a molecular item, a denominator item and a low-precision absolute phase containing fringe level information; substituting the molecular item and the denominator item into an arc tangent function, and performing calculating to obtain high-precision absolute phase information in combination with the low-precision absolute phase. According to the invention, more accurate phase information and more reliable phase unwrapping can be provided without any complex pre- / post-processing.

Description

technical field [0001] The invention belongs to the technical field of optical measurement, and specifically relates to a three-dimensional surface shape measurement method based on deep learning-based single-frame color fringe projection. Background technique [0002] Fringe projection profilometry (FPP) has become one of the most widely used three-dimensional (3D) measurement techniques due to its simple hardware facilities, flexible implementation and high measurement accuracy. In recent years, with the increasing requirements for 3D information acquisition in high-speed scenes in applications such as online quality inspection and rapid reverse engineering, FPP-based high-speed 3D shape measurement technology has become crucial (Robust dynamic 3-d measurements with motion- compensated phase-shifting profilometry, author S Feng et al). [0003] In order to achieve 3D imaging in high-speed scenes, it is necessary to improve measurement efficiency and reduce the number of f...

AI Technical Summary

Problems solved by technology

The former is still unable to stably unfold the phase due to the difficulty in identifying the edges of the Gray code pattern (Projected fringe profilometry using the area-encoded algorithm for spatially isolated and dynamic objects, by W H Su)

The latter can restore the absolute phase by the 3-stripe number selection method (Optical imaging of physical objects, author D Towers, etc.), but due to the use of the Fourier transform (FT) method (a single-frame imaging method, but this method is not in the phase diagram poor quality of discontinuities or isolated regions) resulting in poor phase accuracy

In addition, color-coded projection methods have some inherent drawbacks, such as color difference between channels and color crosstalk, which can affect the quality of phase calculations

Although researchers have proposed some preprocessing methods to compensate for this defect, the impact of these defects on the measurement can only be reduced to a certain extent.

[0004] From the above analysis, it can be seen that although the color-coded projection technology has great potential to achieve single-frame 3D measurement, the only three color channels are not enough to encode fringe images that satisfy both high-quality phase information acquisition and stable phase unwrapping. In addition, The inherent chromatic aberration and color crosstalk problems of this technology are also difficult to solve through traditional methods

Images