Machine vision calibration board with sine characteristic lines
By designing a machine vision calibration board with sinusoidal feature lines, combining straight and curved features, the problems of limited accuracy and insufficient feature point verification of existing calibration boards are solved, achieving high-precision calibration and distortion correction, and is suitable for a variety of industrial vision systems.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-14
- Publication Date
- 2026-03-13
AI Technical Summary
Existing machine vision calibration boards have limited calibration accuracy due to their single feature type, and cannot simultaneously meet the calibration requirements of straight and curved features. The calibration effect is not ideal under complex distortion conditions, and there is a lack of effective feature point verification mechanism.
Design a machine vision calibration board with sinusoidal feature lines. Combining a checkerboard pattern on a rectangular substrate with a through sinusoidal feature line, the system identifies checkerboard corner points and sinusoidal feature points, jointly calculates camera parameters, and uses concentric circle positioning marks and intersection circle recognition marks for positioning and feature verification.
It improves calibration accuracy and enhances the reliability of calibration quality. It is highly adaptable and suitable for various industrial vision systems, including two-dimensional and three-dimensional applications.
Smart Images
Figure CN223993084U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of machine vision technology, and in particular to a machine vision calibration board with sinusoidal feature lines for camera calibration, which is especially suitable for industrial vision inspection systems that require high-precision calibration. Background Technology
[0002] Traditional machine vision calibration boards are mainly divided into two categories: checkerboard calibration boards and circular calibration boards. Checkerboard calibration boards are calibrated by identifying corner points. The advantage is that the algorithm is mature, but the disadvantage is that it lacks curve features and is not sensitive to tangential distortion. Circular calibration boards are calibrated by identifying the center of the circle. The advantage is that it has rotation invariance, but the disadvantage is that it has fewer feature points.
[0003] Therefore, existing machine vision calibration boards have the following shortcomings:
[0004] 1. The single feature type limits the calibration accuracy;
[0005] 2. It cannot simultaneously meet the calibration requirements for both straight and curved features;
[0006] 3. Calibration results are not ideal under complex distortion conditions;
[0007] 4. Lack of an effective feature point verification mechanism. Utility Model Content
[0008] To solve the above-mentioned technical problems, this utility model designs a machine vision calibration plate with sinusoidal feature lines.
[0009] The present invention adopts the following technical solution:
[0010] A machine vision calibration board with a sinusoidal feature line includes a rectangular substrate with a uniformly distributed checkerboard pattern on its surface. A sinusoidal feature line running through the calibration board is provided in the central area of the rectangular substrate. Concentric circle positioning marks are provided at the four corners of the rectangular substrate. Circular identification marks are provided at the intersections of the sinusoidal feature line and the checkerboard pattern.
[0011] Preferably, the amplitude of the sine curve feature line is the width of 2-3 checkerboard patterns, and the wavelength is the width of 8-10 checkerboard patterns.
[0012] Preferably, the diameter of the circular identification mark is 1 / 3 to 1 / 2 of the width of the checkerboard pattern.
[0013] Preferably, the checkerboard pattern uses alternating black and white squares of the same size.
[0014] The calibration method for this machine vision calibration board with sinusoidal feature lines is as follows:
[0015] (1) Acquire images of the calibration board in different poses;
[0016] (2) Identify the corner points of the chessboard and the feature points of the sine curve;
[0017] (3) Calculate camera parameters using both straight line and curve features;
[0018] (4) Perform distortion correction and accuracy verification.
[0019] The principle of this invention is as follows: Positioning marks: Concentric circle positioning marks are set at the four corners of the calibration plate for rapid positioning and direction identification. Feature verification points: Small dots are set at the intersections of the sine curve and the checkerboard lines to verify the accuracy of feature point identification.
[0020] The beneficial effects of this utility model are: (1) Improved calibration accuracy: Combining straight line and curve features, it can more comprehensively correct various distortions; (2) Enhanced reliability: The calibration quality is ensured through feature verification points; (3) Convenient to use: The positioning marks facilitate quick alignment and reduce operation time; (4) Strong adaptability: It is suitable for various industrial vision systems, including two-dimensional and three-dimensional applications. Attached Figure Description
[0021] Figure 1 This is a schematic diagram of the structure of this utility model;
[0022] In the diagram: 1. Rectangular substrate, 2. Checkerboard pattern, 3. Sine curve feature line, 4. Concentric circle positioning mark, 5. Circular identification mark. Detailed Implementation
[0023] The technical solution of this utility model will be further described in detail below through specific embodiments and with reference to the accompanying drawings:
[0024] Example: Figure 1 As shown, a machine vision calibration board with a sinusoidal feature line includes a rectangular substrate 1, a uniformly distributed checkerboard pattern 2 on the surface of the rectangular substrate, a sinusoidal curve feature line 3 running through the calibration board in the central area of the rectangular substrate, concentric circle positioning marks 4 at the four corners of the rectangular substrate, and circular identification marks 5 at the intersection of the sinusoidal curve feature line and the checkerboard pattern.
[0025] The amplitude of the sine curve characteristic line is 2.5 times the width of the checkerboard pattern, and the wavelength is 9 times the width of the checkerboard pattern. The diameter of the circular identification mark is 1 / 3 to 1 / 2 of the width of the checkerboard pattern. The checkerboard pattern uses black and white alternating squares of the same size.
[0026] When using this machine vision calibration board with sinusoidal feature lines, concentric circle positioning marks are set at the four corners of the calibration board for rapid positioning and orientation recognition. Small dots are set at the intersections of the sine curve and the checkerboard lines to verify the accuracy of feature point recognition.
[0027] The embodiments described above are merely preferred solutions of this utility model and are not intended to limit this utility model in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.
Claims
1. A machine vision calibration board with sinusoidal characteristic lines, comprising a rectangular substrate, the surface of which is provided with a uniformly distributed checkerboard pattern, characterized in that, The rectangular substrate center region is provided with a sinusoidal characteristic line penetrating the calibration plate, four corners of the rectangular substrate are provided with concentric circle positioning marks, and the sinusoidal characteristic line is provided with a circular identification mark at an intersection point of the sinusoidal characteristic line and the chessboard pattern.
2. The machine vision calibration panel with sinusoidal feature lines of claim 1, wherein, The amplitude of the sinusoidal characteristic line is 2-3 widths of the chessboard pattern, and the wavelength is 8-10 widths of the chessboard pattern.
3. The machine vision calibration board with sinusoidal feature lines of claim 1, wherein, The diameter of the circular identification mark is 1 / 3-1 / 2 of the width of the chessboard pattern.
4. The machine vision calibration panel with sinusoidal feature lines of claim 1, wherein, The chessboard pattern adopts black and white square grids of the same size.