Machine vision calibration board with concentric circle array and sine characteristic line

Abstract

The utility model discloses a machine vision calibration board with concentric circle arrays and sine characteristic lines, which comprises a rectangular substrate, the surface of the rectangular substrate is provided with concentric circle pattern arrays arranged regularly, and at least one sine curve characteristic line is arranged above the concentric circle pattern arrays. The concentric circle pattern array is regularly arranged in the whole calibration plate area, stable circle center feature points are provided, the sine curve feature lines cover the concentric circle array, and a high-contrast curve contour can be presented in an image. And the concentric circle positioning marks are used for assisting board surface direction detection and alignment. According to the utility model, the characteristic identification rate is obviously improved; the concentric circle pattern has strong rotation invariance and is easy to identify; multi-feature fusion improves precision: curve and circular geometric features are combined to improve fitting capability; the graph layout is reasonable; robust calibration under a wider visual angle and complex distortion can be realized; the method is more suitable for modern industrial scenes, and supports AI image analysis and rapid deployment of edge vision equipment.

Description

Technical Field

[0001] This utility model relates to the field of computer vision and image processing technology, and in particular to a machine vision calibration board with a concentric circle array and sinusoidal feature lines. Background Technology

[0002] Visual measurement technology is well adapted to the new standards and requirements of modern industry for workpiece shape and size inspection. It is a non-contact shape inspection method that combines accuracy and efficiency.

[0003] Checkerboard calibration targets are widely used for camera calibration in vision measurement imaging systems. The process typically involves taking multiple photographs of the calibration target in different poses using the vision measurement imaging system. By extracting and comparing the pixel coordinates of the corner points on the calibration target with the known world coordinates of those corner points, the internal and external parameters of the camera model of the vision measurement imaging system are determined, thus completing the calibration of the system. Therefore, the accuracy of the corner point extraction from the checkerboard calibration target directly affects the final calibration accuracy of the camera.

[0004] However, traditional checkerboard calibration boards suffer from problems such as blurred edges and corner point recognition errors, and are prone to inaccurate feature extraction, especially in scenes with large lens distortion. Utility Model Content

[0005] To solve the above-mentioned technical problems, this utility model designs a machine vision calibration board with a concentric circle array and sinusoidal feature lines.

[0006] The present invention adopts the following technical solution:

[0007] A machine vision calibration board with a concentric circle array and a sinusoidal feature line includes a rectangular substrate. The surface of the rectangular substrate is provided with a regularly arranged array of concentric circle patterns, and at least one sinusoidal feature line is disposed above the array of concentric circle patterns. The concentric circle pattern array is regularly arranged throughout the calibration board area, providing stable center feature points, and the sinusoidal feature line covers the concentric circle array, which can present a high-contrast curve contour in the image.

[0008] Preferably, the rectangular substrate has concentric circle positioning marks at its four corners.

[0009] Preferably, the size of the concentric circle positioning mark is larger than the concentric circles of the concentric circle pattern array. This is used to assist in board surface orientation detection and alignment.

[0010] Preferably, the center point spacing of the concentric circle pattern array is 1 / 5 to 1 / 3 of the width of the rectangular substrate.

[0011] Preferably, the amplitude of the sine curve is the center-to-center distance of 1-2 concentric circle pattern arrays, and the wavelength is the center-to-center distance of 3-4 concentric circle pattern arrays.

[0012] The calibration method for this machine vision calibration board with concentric circle array and sinusoidal feature lines is as follows:

[0013] (1) Acquire calibration board images from multiple perspectives;

[0014] (2) Identify the center point of the concentric circles and the characteristic points of the sine curve;

[0015] (3) Estimate the camera's intrinsic and extrinsic parameters based on the geometric relationships of the point set;

[0016] (4) Perform distortion modeling and correction of the imaging system.

[0017] The beneficial effects of this utility model are: (1) Significantly improve the feature recognition rate: the concentric circle pattern has strong rotational invariance and is easy to recognize; (2) Multi-feature fusion enhances accuracy: the curve and circular geometric features are combined to improve the fitting ability; (3) Reasonable graphic layout: it helps to achieve robust calibration under wider viewing angle and complex distortion; (4) More suitable for modern industrial scenarios: it supports the rapid deployment of AI image analysis and edge vision devices. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of this utility model;

[0019] In the figure: 1. Rectangular substrate, 2. Concentric circle pattern array, 3. Sine curve feature line, 4. Concentric circle positioning mark. Detailed Implementation

[0020] The technical solution of this utility model will be further described in detail below through specific embodiments and with reference to the accompanying drawings:

[0021] Example: Figure 1 As shown, a machine vision calibration board with a concentric circle array and a sinusoidal feature line includes a rectangular substrate 1. The surface of the rectangular substrate is provided with a regularly arranged concentric circle pattern array 2, and a sinusoidal feature line 3 is provided above the concentric circle pattern array.

[0022] The rectangular substrate has concentric circle positioning marks 4 at its four corners. The size of the concentric circle positioning marks is larger than the concentric circles of the concentric circle pattern array. The center point spacing of the concentric circle pattern array is 1 / 5 to 1 / 3 of the width of the rectangular substrate. The amplitude of the sine curve is the center point spacing of 1-2 concentric circle pattern arrays, and the wavelength is the center point spacing of 3-4 concentric circle pattern arrays.

[0023] This invention features a concentric circle pattern array regularly arranged across the entire calibration plate area, providing stable center feature points. A sinusoidal curve feature line covers the concentric circle array, presenting a high-contrast curve outline in the image. Concentric circle positioning marks are used to assist in plate orientation detection and alignment.

[0024] Significantly improves feature recognition rate: concentric circle patterns have strong rotation invariance and are easy to recognize; multi-feature fusion enhances accuracy: curve and circular geometric features are combined to improve fitting ability; reasonable graphic layout: helps to achieve robust calibration under wider viewing angles and complex distortions; more suitable for modern industrial scenarios: supports rapid deployment of AI image analysis and edge vision devices.

[0025] 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 target with concentric circle arrays and sinusoidal features, comprising a rectangular substrate, characterized by, The rectangular substrate surface is provided with a regularly arranged concentric circle pattern array, and at least one sinusoidal characteristic line is arranged above the concentric circle pattern array.

2. The machine vision calibration target of claim 1 having concentric arrays and sinusoidal features, characterized by, The four corners of the rectangular substrate are provided with concentric circle positioning marks.

3. The machine vision calibration target of claim 2 having concentric arrays and sinusoidal features, characterized by, The size of the concentric circle positioning marks is greater than that of the concentric circles of the concentric circle pattern array.

4. The machine vision calibration panel with concentric arrays and sinusoidal features of claim 1, wherein, The center point spacing of the concentric circle pattern array is 1 / 5 to 1 / 3 of the width of the rectangular substrate.

5. The machine vision calibration panel with concentric arrays and sinusoidal features of claim 1, wherein, The wave amplitude of the sinusoidal curve is 1-2 center point spacings of the concentric circle pattern array, and the wavelength is 3-4 center point spacings of the concentric circle pattern array.

Images