CN-121999058-A - Structured light camera calibration method and device, electronic equipment, storage medium and product

CN121999058ACN 121999058 ACN121999058 ACN 121999058ACN-121999058-A

Abstract

The application discloses a method and a device for calibrating a structured light camera, electronic equipment, a storage medium and a product. Relates to the technical field of image processing. The method comprises the steps of moving a calibration plate to different positions, respectively obtaining calibration plate images of all the positions, enabling each identification point in the calibration plate to correspond to a circle with a first radius, a circle with a second radius or a concentric circle with a second radius, determining circle center coordinates of feature points in the calibration plate according to the calibration plate images of all the positions, carrying out stereoscopic vision calibration according to the circle center coordinates to obtain camera parameters, and completing the calibration of the structured light camera if errors between projection points obtained based on the camera parameters and the circle center coordinates meet requirements. According to the scheme, the checkerboard calibration plate is updated to be the composite calibration plate consisting of circles and concentric circles with different sizes, the characteristic points are utilized for three-dimensional vision calibration, and the calibration result is verified based on the projection points, so that the light spot interference can be reduced, the error is reduced, and the robustness and the calibration precision of the calibration process are improved.

Inventors

YANG RONGJIAN
ZHENG LINGXIANG
HOU CHUNYU

Assignees

广州艾目易科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260121

Claims (10)

1. A structured light camera calibration method, comprising: The method comprises the steps of moving a calibration plate to positions with different object distances, and respectively obtaining calibration plate images of the positions, wherein each identification point in the calibration plate corresponds to a circle, the type of the circle comprises a circle with a first radius, a circle with a second radius and a concentric circle, and the first radius is larger than the second radius; determining center coordinates of feature points in the calibration plate according to the calibration plate images of all the positions, wherein the feature points comprise part of identification points in the calibration plate; performing stereoscopic vision calibration according to the center coordinates of the feature points to obtain camera parameters; And if the error between the projection point obtained based on the camera parameters and the center coordinates meets the requirement, the calibration of the structured light camera is completed.
2. The method of claim 1, wherein determining center coordinates of feature points in the calibration plate from calibration plate images for each location comprises: If the characteristic points correspond to a group of concentric circles, fitting an elliptical contour of each circle in the group of concentric circles according to the calibration plate image of each position, calculating to obtain circle center coordinates of the corresponding circles, and carrying out weighted fusion on the corresponding circle center coordinates of each circle to obtain circle center coordinates of the group of concentric circles; if the characteristic points correspond to the circles with the first radius or the circles with the second radius, fitting the elliptical outlines of the circles with the first radius or the circles with the second radius according to the calibration plate images of all the positions, and calculating to obtain the circle center coordinates of the circles with the first radius or the circles with the second radius.
3. The method according to claim 2, wherein the step of performing stereoscopic calibration according to the center coordinates of the feature points to obtain camera parameters includes: In the first calibration, according to the center coordinates of the feature points, taking the minimized re-projection error as a target, carrying out first projection on the known 3D world points, and determining a distortion coefficient according to the result of the first projection; In the secondary calibration, performing de-distortion treatment on the calibration plate image at each position according to the distortion coefficient, and updating the center coordinates of the feature points according to the de-distorted image; and carrying out secondary projection on the known three-dimensional world point according to the updated circle center coordinates, and obtaining camera parameters according to the secondary projection result.
4. The method of claim 2, wherein fitting an elliptical profile to the circles corresponding to the feature points comprises: carrying out median filtering denoising on the circle corresponding to the characteristic point and the local neighborhood marked by the circle; extracting the outline of the circle corresponding to the characteristic point; sampling gray level sections along the radial direction or tangential direction of the contour in the local adjacent region, and obtaining a sub-pixel edge point set of a circle corresponding to the characteristic point through interpolation according to the gray level sections; and fitting the elliptical outline of the circle corresponding to the characteristic point by adopting a least square method according to the sub-pixel edge point set.
5. The method of claim 1, further comprising, prior to performing stereoscopic calibration based on the center coordinates of the feature points to obtain camera parameters: determining the included angle between the main axis direction and the x axis of the center points of all circles in any calibration plate image; Rotating the center points of all circles in opposite directions by the included angle so that the rows and columns of the calibration plates are approximately parallel to the image coordinate axes; Based on a clustering algorithm, clustering the circle center points of all circles according to an x-axis coordinate and a y-axis coordinate respectively to obtain a plurality of row groups and a plurality of column groups of the circle center points; Fitting the center points in each row grouping and each column grouping into straight lines; according to the straight line with the maximum number of the inner points, eliminating the abnormal points in the corresponding row or the corresponding column; And determining the characteristic point topology in the calibration plate according to the inner point quantity alternation constraint and the point spacing consistency of the row grouping and the column grouping.
6. The method of claim 1, further comprising, prior to performing stereoscopic calibration based on the center coordinates of the feature points to obtain camera parameters: based on a maximum inter-class variance method, binarizing the calibration plate image of each position to obtain a rough segmentation threshold; Performing morphological opening operation on the binarized calibration plate image by adopting square structural elements to obtain the calculated calibration plate image at each position; determining a subdivision threshold based on a maximum inter-class variance method according to the histogram; and re-binarizing the calibration plate image of each position according to the subdivision threshold value to obtain the preprocessed calibration plate image of each position.
7. A structured light camera calibration apparatus, comprising: The image acquisition module is used for moving the calibration plate to positions with different object distances and respectively acquiring the images of the calibration plate at each position, wherein each identification point in the calibration plate corresponds to a circle, the type of the circle comprises a circle with a first radius, a circle with a second radius and a concentric circle, and the first radius is larger than the second radius; the coordinate determining module is used for determining the center coordinates of the characteristic points in the calibration plate according to the calibration plate images of all the positions, wherein the characteristic points comprise part of identification points in the calibration plate; The calibration module is used for carrying out stereoscopic vision calibration according to the circle center coordinates of the characteristic points to obtain camera parameters; and the checking module is used for completing the calibration of the structured light camera if the error between the projection point obtained based on the camera parameters and the center coordinates meets the requirement.
8. An electronic device, comprising: at least one processor; a memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the structured light camera calibration method of any one of claims 1-6.
9. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the structured light camera calibration method according to any of claims 1-6.
10. A computer program product comprising a computer program and/or instructions which, when executed by a processor, implements the structured light camera calibration method as claimed in any one of claims 1 to 6.

Description

Structured light camera calibration method and device, electronic equipment, storage medium and product Technical Field The embodiment of the application relates to the technical field of image processing, in particular to a method and a device for calibrating a structured light camera, electronic equipment, a storage medium and a product. Background The binocular vision technology synchronously acquires images from different visual angles by simulating the principle of binocular parallax of human beings by utilizing two cameras, so that three-dimensional geometric information of a scene is acquired, and the technology becomes a core foundation in the fields of robot navigation, automatic driving, high-precision three-dimensional reconstruction, augmented reality, virtual reality, industrial vision measurement and the like. The structured light binocular camera refers to a stereoscopic vision system consisting of a left visible light camera and a right visible light camera, and can be used for high-precision three-dimensional measurement or depth perception in a low-illumination or full-black environment. The main method of calibrating the existing structured light camera is realized based on a plane checkerboard (black-white checkerboard pattern), such as Zhang Zhengyou calibration method. The method has the defects that active facula interference exists, the contrast and texture are weak, the feature consistency is poor, errors are easy to transfer and amplify, the robustness is insufficient, and the calibration precision is required to be improved. Disclosure of Invention The application provides a method, a device, electronic equipment, a storage medium and a product for calibrating a structured light camera, so as to improve the robustness and the calibration precision of the calibration process. In a first aspect, an embodiment of the present application provides a method for calibrating a structured light camera, including: The method comprises the steps of moving a calibration plate to positions with different object distances, and respectively obtaining calibration plate images of the positions, wherein each identification point in the calibration plate corresponds to a circle, the type of the circle comprises a circle with a first radius, a circle with a second radius and a concentric circle, and the first radius is larger than the second radius; determining center coordinates of feature points in the calibration plate according to the calibration plate images of all the positions, wherein the feature points comprise part of identification points in the calibration plate; performing stereoscopic vision calibration according to the center coordinates of the feature points to obtain camera parameters; And if the error between the projection point obtained based on the camera parameters and the center coordinates meets the requirement, the calibration of the structured light camera is completed. In a second aspect, an embodiment of the present application further provides a structured light camera calibration device, including: The image acquisition module is used for moving the calibration plate to positions with different object distances and respectively acquiring the images of the calibration plate at each position, wherein each identification point in the calibration plate corresponds to a circle, the type of the circle comprises a circle with a first radius, a circle with a second radius and a concentric circle, and the first radius is larger than the second radius; the coordinate determining module is used for determining the center coordinates of the characteristic points in the calibration plate according to the calibration plate images of all the positions, wherein the characteristic points comprise part of identification points in the calibration plate; The calibration module is used for carrying out stereoscopic vision calibration according to the circle center coordinates of the characteristic points to obtain camera parameters; and the checking module is used for completing the calibration of the structured light camera if the error between the projection point obtained based on the camera parameters and the center coordinates meets the requirement. In a third aspect, an embodiment of the present application provides an electronic device, including: One or more processors; a storage means for storing one or more programs; the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the structured light camera calibration method as described in the first aspect. In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the structured light camera calibration method according to the first aspect. In a fifth aspect, embodiments of the present application further provide a computer program produc