Projector calibration method based on red and blue checkerboard calibration plate

Abstract

The invention relates to a projector calibration method based on a red and blue checkerboard calibration plate. The method comprises the steps that the calibration plate is selected; a measurement system is set up; a corner point coordinate is extracted; deformation grating fringe images are collected; a filtering function is constructed for compensation, wherein Fourier transforming is carried out on the acquired deformation grating fringe images in horizontal and vertical directions to acquire the amplitude spectrum image of deformation grating stripes; a frequency domain filtering function is constructed for filtering to extract fundamental frequency component and direct current component in the amplitude spectrum image; inverse Fourier transforming is carried out on the extracted fundamental frequency component and direct current component to acquire compensated deformation grating fringe images in horizontal and vertical directions; the wrapped phase images of the compensated deformation grating fringe images in horizontal and vertical directions are calculated; a target image is acquired after dephasing; and a projector is calibrated. According to the invention, the constructed frequency domain filtering function can filter out components except the fundamental frequency component and the direct current component; and the influence of gamma nonlinearity and environmental noise can be obviously reduced.

Description

technical field [0001] The invention relates to a projector calibration method, in particular to a projector calibration method with a plane calibration plate with uniform reflectivity, and in particular to a projector calibration method based on a red and blue checkerboard calibration plate. Background technique [0002] In phase measurement profilometry, the calibration of the hardware system is a necessary step to complete the transformation from the pixel coordinates of the image plane and the projection plane to the three-dimensional coordinates of the space, and the calibration accuracy determines the measurement accuracy of the system. Hardware system calibration includes internal and external parameter matrix calibration of cameras and projectors. At present, camera calibration methods are becoming more and more mature. A typical example is the Matlab camera calibration toolbox established by Bouguet et al. Since the second-order radial and tangential distortion of th...

AI Technical Summary

Problems solved by technology

[0003] In actual measurement, the deformed grating fringe image collected by the camera will be affected by factors such as projector-camera gamma nonlinearity and environmental noise, so the phase error will inevitably be introduced

In order to reduce the impact of environmental noise and gamma nonlinear factors on the measurement accuracy of the system, Zhang et al. assumed that the main source of the phase error is the gamma nonlinearity of the projector, and used the look-up table method to increase the phase accuracy by 5.6 times. Good suppression of errors caused by gamma nonlinearity of the projector, but no significant suppression of phase errors caused by ambient noise

Zhou et al. established a model of the relationship between ambient light and phase error, so that the phase error was significantly reduced, but this method needs to project all-white and all-black patterns onto a uniform flat whiteboard to determine the average grayscale and Modulation, this method is not practical in industrial production sites with complex noise sources, such as forging and welding industrial production sites

Lei et al. proposed a multi-frequency inverse phase error method to compensate for the nonlinear error of the phase. The projector projects two sets of phase-shifted grating fringe images with an initial phase difference of π / 4 onto the surface of the object, and the wrapping phase of the two sets of grating fringe images Taking the average value can achieve the purpose of suppressing the γ non-linear error, but this method needs to project twice the number of grating fringe images, so the measurement efficiency is low

Images