Binocular vision verification method, control device, image acquisition equipment and storage medium

Abstract

The embodiment of the invention discloses a binocular vision verification method, a control device, image acquisition equipment and a storage medium, and belongs to the technical field of equipment verification. The binocular vision verification method comprises the steps that a verification binocular image obtained when target binocular vision equipment collects a target object is acquired, the target object comprises a plurality of entity verification circles with the same circle center, and each verification circle comprises a plurality of color features; obtaining a circle boundary two-dimensional coordinate of a maximum check circle in the target object based on one eye image of the check binocular image; converting the circular boundary two-dimensional coordinates into circular boundary three-dimensional coordinates; performing circle center extraction on the check binocular image to obtain a circle center three-dimensional coordinate; and based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the circle center, performing precision verification on the target binocular vision equipment to obtain a verification result. The plurality of color features provide enough gradient features, and the imaging precision of binocular vision is considered, so that the reliability of the verification result of the binocular vision equipment is improved.

Description

Technical Field

[0001] This application relates to the technical field of equipment calibration, and specifically relates to a binocular vision calibration method, a control device, an image acquisition device, and a storage medium. Background Art

[0002] Binocular stereo vision is a form of machine vision. Based on the parallax principle and using imaging devices to obtain two images of the object to be measured from different positions, the three-dimensional geometric information of the object is obtained by calculating the position deviation between corresponding points in the images. Before using binocular vision devices such as binocular cameras, it is usually necessary to calibrate the binocular vision device so that the left-eye image and the right-eye image are on the same plane, eliminate lens distortion, and avoid image distortion.

[0003] Usually, the accuracy calibration of binocular vision devices is based on predetermined corner points and pre-designed image features. However, the clearly designed image features cannot robustly express the imaging ability of binocular vision devices, resulting in poor reliability of the calibration results of binocular vision devices. Summary of the Invention

[0004] The purpose of the embodiments of this application is to provide a binocular vision calibration method, a control device, an image acquisition device, and a storage medium to solve the problem of poor reliability of the calibration results of binocular vision devices in the prior art.

[0005] To achieve the above purpose, the first aspect of this application provides a binocular vision calibration method, which includes:

[0006] Obtain calibration binocular images obtained when a target binocular vision device captures a target object, where the target object includes multiple solid calibration circles with the same center, and each calibration circle includes multiple color features;

[0007] Based on one of the images of the calibration binocular images, obtain the two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object;

[0008] Convert the two-dimensional coordinates of the circle boundary into three-dimensional coordinates of the circle boundary;

[0009] Extract the center of the circle from the calibration binocular images to obtain the three-dimensional coordinates of the center of the circle;

[0010] Based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center of the circle, perform accuracy calibration on the target binocular vision device to obtain a calibration result.

[0011] In the embodiments of this application, the target object is a solid calibration plate, and the solid calibration plate is obtained by three-dimensional printing.

[0012] In an embodiment of the present application, based on verifying one of the binocular images, obtaining the two-dimensional coordinates of the circular boundary of the largest verification circle in the target object includes:

[0013] Performing contour extraction on one of the binocular images to be verified to obtain a plurality of verification circle contour images;

[0014] Performing Hough transform on the plurality of verification circle contour images to obtain a plurality of verification circle roundnesses;

[0015] Filtering the plurality of verification circle roundnesses based on a preset roundness threshold to obtain the filtered plurality of verification circle roundnesses;

[0016] Based on the filtered plurality of verification circle roundnesses, determining the largest verification circle contour in the target object to obtain the two-dimensional coordinates of the circular boundary of the largest verification circle.

[0017] In an embodiment of the present application, extracting the center coordinates of a circle from the binocular image to be verified to obtain the three-dimensional coordinates of the center of the circle includes:

[0018] Determining the smallest verification circle contour in the binocular image to be verified, and obtaining all the left-eye center pixels and right-eye center pixels corresponding to the smallest verification circle contour;

[0019] Calculating the average value of the left-eye center pixels to obtain the left-eye center coordinates, and calculating the average value of the right-eye center pixels to obtain the right-eye center coordinates;

[0020] According to the left-eye center coordinates and the right-eye center coordinates, obtaining the three-dimensional coordinates of the center of the circle.

[0021] In an embodiment of the present application, converting the two-dimensional coordinates of the circular boundary into the three-dimensional coordinates of the circular boundary includes:

[0022] Converting the binocular image to be verified into a target image, where the target image is one of a depth image, a disparity image, and a three-dimensional point cloud image;

[0023] Mapping the two-dimensional coordinates of the circular boundary according to the target image to obtain the three-dimensional coordinates of the circular boundary.

[0024] In an embodiment of the present application, the verification result includes at least one of a reconstruction integrity verification result, a dimensional accuracy verification result, and a planar distribution accuracy verification result. Based on the three-dimensional coordinates of the circular boundary and the three-dimensional coordinates of the center of the circle, performing accuracy verification on the target binocular vision device to obtain the verification result, including:

[0025] Determining the number of first boundary points and the number of second boundary points, where the first boundary point is the point corresponding to the two-dimensional coordinates of the circular boundary, and the second boundary point is the point corresponding to the three-dimensional coordinates of the circular boundary;

[0026] According to the number of first boundary points and the number of second boundary points, obtaining the reconstruction integrity verification result;

[0027] Based on the three-dimensional coordinates of the circular boundary, the three-dimensional coordinates of the center of the circle, and the standard radius of the maximum verification circle, calculate the cumulative probability distribution of the dimension, and calculate the cumulative probability distribution of the error value to obtain the dimension accuracy verification result, where the dimension is the distance between the second boundary point and the center of the circle, and the error value is the difference between the dimension and the standard radius of the maximum verification circle;

[0028] Construct a fitting plane according to the second boundary point;

[0029] Based on the three-dimensional coordinates of the circular boundary, calculate the cumulative probability distribution of the perpendicular distance from each second boundary point to the fitting plane to obtain the plane distribution accuracy verification result.

[0030] In the embodiments of the present application, based on verifying one of the binocular images, obtaining the two-dimensional coordinates of the circular boundary of the maximum verification circle in the target object includes:

[0031] Perform binarization processing on one of the binocular images to be verified to obtain a target binary image;

[0032] Based on the target binary image, obtain the two-dimensional coordinates of the circular boundary of the maximum verification circle in the target object.

[0033] The second aspect of the present application provides a control device, including:

[0034] A memory configured to store instructions;

[0035] A processor configured to call instructions from the memory and be able to implement the above binocular vision verification method when executing the instructions.

[0036] The third aspect of the present application provides an image acquisition device, including:

[0037] According to the above control device;

[0038] A binocular vision device configured to acquire binocular vision images.

[0039] The fourth aspect of the present application provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to cause the machine to execute the above binocular vision verification method.

[0040] The present application provides a binocular vision calibration method, including: obtaining calibration binocular images obtained when a target binocular vision device captures a target object, where the target object includes a plurality of physical calibration circles with the same center, and each calibration circle includes a plurality of color features; obtaining two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object based on one of the binocular images of the calibration binocular images; converting the two-dimensional coordinates of the circle boundary into three-dimensional coordinates of the circle boundary; extracting the three-dimensional coordinates of the center of the circle from the calibration binocular images; and performing accuracy calibration on the target binocular vision device based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center of the circle to obtain a calibration result. When performing accuracy calibration based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center of the circle, the calibration points with continuous and uniform distribution in the image take into account the imaging accuracy in all directions and positions of the binocular vision, thereby improving the reliability of the calibration result of the binocular vision device.

[0041] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification, and are used to explain the embodiments of the present application together with the following specific implementation, but do not constitute a limitation to the embodiments of the present application. In the drawings:

[0043] Figure 1 Schematically shows a flowchart of a binocular vision calibration method according to an embodiment of the present application;

[0044] Figure 2 Schematically shows an example diagram of a target object according to an embodiment of the present application;

[0045] Figure 3 Schematically shows an example diagram of a calibration binocular image according to an embodiment of the present application;

[0046] Figure 4 Schematically shows an example diagram of a target image according to an embodiment of the present application;

[0047] Figure 5 Schematically shows an example diagram of the cumulative probability distribution of a size according to an embodiment of the present application;

[0048] Figure 6 Schematically shows an example diagram of the cumulative probability distribution of an error value according to an embodiment of the present application;

[0049] Figure 7 Schematically shows an example diagram of the cumulative probability distribution of the perpendicular distance between a second boundary point and a fitting plane according to an embodiment of the present application;

[0050] Figure 8Schematic diagram showing an example of a machine-readable storage medium according to an embodiment of the present application. Detailed implementation manners

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It should be understood that the specific implementation manners described herein are only for explaining and illustrating the embodiments of the present application, and are not used to limit the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts belong to the scope of protection of the present application.

[0052] It should be noted that the acquisition, transmission, storage, use, processing, etc. of data in the technical solutions of the present application all comply with the relevant regulations of national laws and regulations. In the embodiments of the present application, some industry-existing solutions such as certain software, components, models, etc. may be mentioned. They should be regarded as exemplary, and their purpose is only to illustrate the feasibility in the implementation of the technical solutions of the present application, but it does not mean that the applicant has already or necessarily used this solution.

[0053] It should be noted that if there are directional indications involved in the embodiments of the present application, the directional indications are only used to explain the relative positional relationship and movement conditions between components in a specific posture. If this specific posture changes, the directional indications will also change accordingly.

[0054] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present application.

[0055] Figure 1 Schematic diagram showing the flowchart of a binocular vision verification method according to an embodiment of the present application. As Figure 1 shown, the embodiments of the present application provide a binocular vision verification method, and the binocular vision verification method includes:

[0056] S110, obtaining a verification binocular image obtained when a target binocular vision device collects a target object, where the target object includes a plurality of entity verification circles with the same center, and each verification circle includes a plurality of color features.

[0057] Please refer to Figure 2 , Figure 2 which schematically shows an example diagram of a target object according to an embodiment of the present application.

[0058] In this embodiment, a target object is pre-designed. The target object includes multiple calibration circles, and the multiple calibration circles correspond to the same center of the circle. The target binocular vision device is a binocular vision device that needs to be calibrated for accuracy. The type of the target binocular vision device is set according to actual needs and can be a binocular vision camera, etc., which is not limited here.

[0059] Obtain the calibrated binocular images obtained when the target binocular vision device captures the target object. Since each calibration circle includes multiple color features, that is, the target object is composed of circles with ribbon features, the multiple color features provide sufficient gradient features for the target object to be matched. Furthermore, the calibration circles can be extracted by experimental methods for accuracy calibration. Both the color features and the number of calibration circles are set according to actual needs and are not limited here.

[0060] In the embodiment of the present application, the target object is a physical calibration board, and the physical calibration board is obtained by three-dimensional printing.

[0061] Printing the target object into a physical structure to obtain a physical calibration board can convert the planar continuous image features into spatially connected planar features, and better provide a uniform corner distribution in all directions for the target binocular vision device. In this embodiment, the target object is three-dimensionally printed by a 3D (Dimension) printing device to obtain a physical calibration board. After printing the three-dimensional structure of the target object, the image of the physical calibration board is captured by the target binocular vision device to obtain the calibrated binocular images.

[0062] S120, based on one of the binocular images of the calibrated binocular images, obtain the two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object.

[0063] The obtained calibrated binocular images include a left-eye image and a right-eye image. One of the binocular images can be the left-eye image or the right-eye image, which is not limited here. Based on one of the binocular images of the calibrated binocular images, obtain the two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object. The largest calibration circle is the circumscribed calibration circle of the target object. Based on one of the left-eye image and the right-eye image, determine the largest calibration circle in the target object, and then obtain the two-dimensional coordinates of the circle boundary of the largest calibration circle.

[0064] In the embodiment of the present application, obtaining the two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object based on one of the binocular images of the calibrated binocular images includes:

[0065] Perform contour extraction on one of the binocular images of the calibrated binocular images to obtain multiple calibration circle contour images;

[0066] Perform the Hough transform on multiple calibration circle contour images to obtain multiple calibration circle roundness values;

[0067] Filter the multiple calibration circle roundness values based on a preset roundness threshold to obtain the filtered multiple calibration circle roundness values;

[0068] Based on the filtered multiple calibration circle roundness values, determine the largest calibration circle contour in the target object to obtain the two-dimensional coordinates of the circular boundary of the largest calibration circle.

[0069] Perform contour extraction on one of the calibration binocular images to obtain multiple calibration circle contour images. Since the graphic structures of the target objects are all circular, the contours in the image can be quickly extracted. During contour extraction, unknown contours in the image background are also extracted. Perform the Hough transform on the multiple calibration circle contour images to obtain multiple calibration circle roundness values to filter out abnormal image contours.

[0070] Filter the multiple calibration circle roundness values based on a preset roundness threshold to obtain the filtered multiple calibration circle roundness values. By presetting the roundness threshold in advance, unknown contours are filtered out and circular contours that meet the conditions are retained. Based on the filtered multiple calibration circle roundness values, determine the largest calibration circle contour in the target object. During the imaging process of the target binocular vision device, the image of the calibration circle occupies a relatively large proportion of the picture, so that the largest calibration circle contour is the contour of the circumscribed calibration circle, and then the two-dimensional coordinates of the circumscribed calibration circle can be obtained.

[0071] S130, convert the two-dimensional coordinates of the circular boundary into three-dimensional coordinates of the circular boundary.

[0072] During the calibration process of the target binocular vision device, three-dimensional coordinates are required. Perform image reconstruction on the calibration binocular images, and then convert the two-dimensional coordinates of the circular boundary into three-dimensional coordinates of the circular boundary.

[0073] In the embodiments of the present application, obtaining the two-dimensional coordinates of the circular boundary of the largest calibration circle in the target object based on one of the calibration binocular images includes:

[0074] Perform binarization processing on one of the calibration binocular images to obtain a target binary image;

[0075] Based on the target binary image, obtain the two-dimensional coordinates of the circular boundary of the largest calibration circle in the target object.

[0076] Obtaining the three-dimensional coordinates of the circular boundary of the largest calibration circle in the target object based on one of the calibration binocular images includes:

[0077] Perform binarization processing on one of the calibration binocular images to obtain a target binary image;

[0078] Based on the target binary image, obtain the two-dimensional coordinates of the circular boundary of the largest calibration circle in the target object.

[0079] For ease of understanding, in an embodiment of the present application, one of the binocular images is the left-eye image. Perform binarization processing on one of the binocular images for calibration, that is, perform threshold binarization on the left-eye image to obtain a target binary image with a black-and-white binary distribution:

[0080]

[0081] Wherein, f(x, y) is the target binary image with a black-and-white binary distribution, I(x, y) is the pixel value of the pixel point with coordinates (x, y), and n is the binarization threshold.

[0082] Based on the target binary image, extract the contour of the largest calibration circle to obtain the two-dimensional coordinates of the circular boundary of the largest calibration circle in the target object. Performing binarization processing on the image reduces the influence of light and environment on imaging, and thus can more accurately locate and extract features in the image.

[0083] S140. Perform center extraction on the binocular images for calibration to obtain the three-dimensional coordinates of the center.

[0084] Determine the center pixels in the binocular images for calibration, and then perform center extraction on the binocular images for calibration to obtain the three-dimensional coordinates of the center. The three-dimensional coordinates of the center can be used to calibrate the dimensional accuracy of the binocular vision system, and thus obtain the calibration result of the binocular vision system.

[0085] In an embodiment of the present application, performing center extraction on the binocular images for calibration to obtain the three-dimensional coordinates of the center includes:

[0086] Determine the smallest calibration circle contour in the binocular images for calibration, and obtain all the left-eye center pixels and right-eye center pixels corresponding to the smallest calibration circle contour;

[0087] Perform mean calculation on the left-eye center pixels to obtain the left-eye center coordinates, and perform mean calculation on the right-eye center pixels to obtain the right-eye center coordinates;

[0088] Obtain the three-dimensional coordinates of the center according to the left-eye center coordinates and the right-eye center coordinates.

[0089] Each calibration circle in the target object is circular in the plane. The calibration circle is an ellipse with a relatively large roundness in the visual imaging of the binocular vision device, and the imaging of the center will also be an ellipse. Perform contour extraction on the binocular images for calibration to determine the smallest calibration circle contour in the binocular images for calibration. Obtain all the center pixels within the smallest calibration circle contour through binarization. Since the binocular images for calibration include the left-eye image and the right-eye image, obtain all the left-eye center pixels and right-eye center pixels corresponding to the smallest calibration circle contour.

[0090] Obtain the average value to get the sub-pixel center coordinates of the circle, that is, calculate the average value of the left-eye center pixels in the left-eye image to obtain the left-eye center coordinates, and calculate the average value of the right-eye center pixels in the right-eye image to obtain the right-eye center coordinates. In this embodiment, according to the left-eye center coordinates, the right-eye center coordinates, the image center of the left-eye image, the focal length, and the baseline length, the three-dimensional coordinates of the center of the circle are obtained:

[0091]

[0092] Among them, (x, y, z) are the three-dimensional coordinates of the center of the circle, (x l , y l ) are the left-eye center coordinates, (x r , y r ) are the right-eye center coordinates, (c x , c y ) is the image center of the left-eye image, f is the focal length, d is the planar parallax, and b is the baseline length.

[0093] S150. Based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center of the circle, perform accuracy verification on the target binocular vision device to obtain a verification result.

[0094] The checkerboard calibration method means placing multiple checkerboard patterns with known sizes and shapes in front of the binocular vision device and taking multiple checkerboard patterns through the binocular vision device. The reverse process of the checkerboard calibration method can be used for the verification of the binocular vision device, and the accuracy of the binocular vision device is verified through the accurate positioning points of the checkerboard. However, due to the unclear gradient features of the checkerboard, the pure-color flat area lacking feature regions cannot be effectively reconstructed by the reconstruction algorithm, and the sparse point set is difficult to accurately verify the imaging accuracy of the target binocular vision device.

[0095] During the reconstruction process of the target object, verification points that are continuously and evenly distributed are provided. Based on the three-dimensional coordinates of the center of the circle and the three-dimensional coordinates of the circle boundary of the largest verification circle for accuracy verification, the complete imaging accuracy of the target binocular vision device in all directions and positions can be verified, and a verification result with higher reliability can be obtained. During the accuracy verification process, the target object can combine the image plane with space, and multiple color features provide sufficient gradient features. When performing accuracy verification based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center of the circle, the verification points with continuous and uniform distribution in the image take into account the imaging accuracy of the binocular vision in all directions and positions, thereby improving the reliability of the verification result of the binocular vision device.

[0096] In the embodiments of the present application, converting the two-dimensional coordinates of the circle boundary into three-dimensional coordinates of the circle boundary includes:

[0097] Convert the verification binocular image into a target image, where the target image is one of a depth image, a disparity image, and a three-dimensional point cloud image;

[0098] Map the two-dimensional coordinates of the circle boundary according to the target image to obtain the three-dimensional coordinates of the circle boundary.

[0099] Please refer to Figure 3 , Figure 3 which schematically shows an example diagram of verifying a binocular image according to an embodiment of the present application.

[0100] The target binocular vision device performs visual image acquisition on the physical calibration board to obtain the calibrated binocular image. The calibrated binocular image is a two-dimensional planar image, where x l is the coordinate of the left-eye image along the x-axis, x r is the coordinate of the right-eye image along the x-axis, P(x, y, z) is the three-dimensional coordinate of the physical calibration board, f is the focal length, b is the baseline length, and d = x l - x r is the planar parallax.

[0101] Reconstruct the calibrated binocular image to convert it into a target image. The target image is one of a depth image, a disparity image, and a three-dimensional point cloud image. Based on one of the images of the calibrated binocular image, contour extraction is performed, that is, the contour of the circumscribed calibration circle is extracted, that is, the two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object are obtained. Map the two-dimensional coordinates of the circle boundary according to the reconstructed target image to obtain the three-dimensional coordinates of the circle boundary.

[0102] In an embodiment of the present application, the verification result includes at least one of a reconstruction integrity verification result, a dimensional accuracy verification result, and a planar distribution accuracy verification result. Based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center of the circle, perform accuracy verification on the target binocular vision device to obtain the verification result, including:

[0103] Determine the number of first boundary points and the number of second boundary points, where the first boundary point is the point corresponding to the two-dimensional coordinates of the circle boundary, and the second boundary point is the point corresponding to the three-dimensional coordinates of the circle boundary;

[0104] Obtain the reconstruction integrity verification result according to the number of first boundary points and the number of second boundary points;

[0105] Based on the three-dimensional coordinates of the circle boundary, the three-dimensional coordinates of the center of the circle, and the standard radius of the largest calibration circle, calculate the cumulative probability distribution of the size and calculate the cumulative probability distribution of the error value to obtain the dimensional accuracy verification result, where the size is the distance between the second boundary point and the center of the circle, and the error value is the difference between the size and the standard radius of the largest calibration circle;

[0106] Construct a fitting plane according to the second boundary points;

[0107] Based on the three-dimensional coordinates of the circle boundary, calculate the cumulative probability distribution of the perpendicular distance from each second boundary point to the fitting plane to obtain the planar distribution accuracy verification result.

[0108] Please refer to Figure 4 , Figure 4 which schematically shows an example diagram of a target image according to an embodiment of the present application.

[0109] After obtaining the three-dimensional coordinates of the circular boundary and the three-dimensional coordinates of the center of the circle, at least one of the reconstruction integrity verification result, the dimensional accuracy verification result, and the planar distribution accuracy verification result is obtained. For ease of understanding, in the embodiments of the present application, the reconstruction integrity verification result, the dimensional accuracy verification result, and the planar distribution accuracy verification result are obtained together. The first boundary point is the point corresponding to the two-dimensional coordinates of the circular boundary, and the second boundary point is the point corresponding to the three-dimensional coordinates of the circular boundary. Determine the number of the first boundary points and the number of the second boundary points in the imaging. According to the number of the first boundary points and the number of the second boundary points, determine whether the corresponding spatial points are reconstructed in the reconstructed target image, and obtain the reconstruction integrity verification result. For ease of understanding, in the embodiments of the present application, the number of the first boundary points is M, the number of the second boundary points is m, and the reconstruction integrity verification result is m / M.

[0110] Please refer to Figure 5 , Figure 5 which schematically shows an example diagram of the cumulative probability distribution of a dimension according to an embodiment of the present application.

[0111] Based on the three-dimensional coordinates of the circular boundary, the three-dimensional coordinates of the center of the circle, and the standard radius of the maximum verification circle, calculate the cumulative probability distribution of the dimension, and calculate the cumulative probability distribution of the error value to obtain the dimensional accuracy verification result.

[0112] Specifically, calculate the distance between the second boundary point and the center of the circle, that is, obtain the size of the maximum verification circle radius in the imaging, and calculate the cumulative probability distribution of the distance from the second boundary point corresponding to the three-dimensional coordinates of the circular boundary to the center of the circle, that is, the cumulative probability distribution of the dimension. As shown in Figure 5, the abscissa of the image is the distance between the second boundary point and the center of the circle, that is, the abscissa is the dimension, and the ordinate is the cumulative probability distribution of the dimension.

[0113] Please refer to Figure 6 , Figure 6 which schematically shows an example diagram of the cumulative probability distribution of an error value according to an embodiment of the present application.

[0114] The standard radius is the actual radius value of the maximum verification circle, and the error value is the difference between the dimension and the standard radius of the maximum verification circle. Subtract the standard radius from the dimension, and then calculate the cumulative probability distribution of the error value to obtain the dimensional accuracy verification result. As shown in Figure 6, the abscissa of the image is the error value between the dimension and the standard radius, and the ordinate is the cumulative probability distribution of the dimension minus the standard radius.

[0115] Please refer to Figure 7 , Figure 7An exemplary diagram of the cumulative probability distribution of the perpendicular distance between a second boundary point and a fitted plane according to an embodiment of the present application is schematically shown.

[0116] Based on the second boundary points, a fitted plane is constructed. Based on the three-dimensional coordinates of the circular boundary, the cumulative probability distribution of the perpendicular distance between each second boundary point and the fitted plane is calculated to verify the plane reconstruction accuracy, and the plane distribution accuracy verification result is obtained. As shown in the figure, the abscissa of the image is the perpendicular distance between the second boundary point and the fitted plane, and the cumulative probability distribution of the perpendicular distance between the second boundary point and the fitted plane.

[0117] The present application provides a binocular vision verification method, including: obtaining a verification binocular image obtained when a target binocular vision device captures a target object, where the target object includes a plurality of physical verification circles with the same center, and each verification circle includes a plurality of color features; based on one of the binocular images of the verification binocular image, obtaining the two-dimensional coordinates of the circular boundary of the largest verification circle in the target object; converting the two-dimensional coordinates of the circular boundary into three-dimensional coordinates of the circular boundary; extracting the center coordinates of the circle from the verification binocular image to obtain the three-dimensional center coordinates; based on the three-dimensional coordinates of the circular boundary and the three-dimensional center coordinates, performing accuracy verification on the target binocular vision device to obtain a verification result. When performing accuracy verification based on the three-dimensional coordinates of the circular boundary and the three-dimensional center coordinates, the verification points with continuous and uniform distribution in the image take into account the imaging accuracy in all directions and positions of binocular vision, thereby improving the reliability of the verification result of the binocular vision device.

[0118] An embodiment of the present application further provides a control device, including:

[0119] A memory configured to store instructions;

[0120] A processor configured to call instructions from the memory and capable of implementing the above-mentioned binocular vision verification method when executing the instructions.

[0121] The processor contains a kernel, and the kernel retrieves the corresponding program unit from the memory. One or more kernels can be set, and the above-mentioned binocular vision verification method is implemented by adjusting the kernel parameters.

[0122] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of, for example, read-only memory (ROM) or flash memory (flash RAM), and the memory includes at least one storage chip.

[0123] An embodiment of the present application further provides an image acquisition device, including:

[0124] The above-mentioned control device;

[0125] A binocular vision device configured to obtain a binocular vision image.

[0126] When precision detection of an image acquisition device is required, a binocular vision device is used to obtain calibration binocular images corresponding to a target object. The control device obtains three-dimensional coordinates of a circle boundary and three-dimensional coordinates of a circle center based on the calibration binocular images, and performs precision calibration on the target binocular vision device based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the circle center to obtain a calibration result.

[0127] An embodiment of the present application further provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to cause a machine to execute the above binocular vision calibration method.

[0128] Please refer to Figure 8 , Figure 8 which schematically shows an example diagram of a machine-readable storage medium according to an embodiment of the present application.

[0129] The image acquisition device includes a control device 200 and a binocular vision device 300. The image acquisition device may further include other devices. The other devices are set according to actual needs and may be a human-computer interaction device or the like, which is not limited herein. The memory 210 and the processor 220 are connected through a system bus. The processor 220 is used to provide computing and control capabilities. The memory 210 includes a machine-readable storage medium 211, and the memory 210 provides an environment for the operation of the machine-readable storage medium 211. Instructions are stored on the machine-readable storage medium 211, and the instructions are used to cause a machine to execute the above binocular vision calibration method.

[0130] Those skilled in the art should understand that the embodiments of the present application may be provided as a method, a system, or a computer program product. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0131] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one or more flows or multiple flows and / or blocks Figure 1 one or more blocks or multiple blocks.

[0132] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the processes Figure 1 or processes and / or boxes Figure 1 or boxes specified.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the processes Figure 1 or processes and / or boxes Figure 1 or boxes specified.

[0134] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0135] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

[0136] Computer-readable media includes both permanent and non-permanent, removable and non-removable media implemented by any method or technology for storage of information such as computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), digital versatile discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0137] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0138] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A binocular vision verification method, characterized in that, The binocular vision calibration method includes: Obtaining calibration binocular images obtained by a target binocular vision device when collecting a target object, where the target object includes multiple solid calibration circles with the same center, and each calibration circle includes multiple color features; Based on one of the calibration binocular images, obtaining the two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object; Converting the two-dimensional coordinates of the circle boundary into three-dimensional coordinates of the circle boundary; Performing center extraction on the calibration binocular images to obtain three-dimensional coordinates of the center; Based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center, performing accuracy calibration on the target binocular vision device to obtain a calibration result.

2. The binocular vision verification method according to claim 1, characterized in that, The target object is a solid calibration plate, and the solid calibration plate is obtained by three-dimensional printing.

3. The binocular vision verification method according to claim 1, wherein The obtaining the two-dimensional coordinates of the circle boundary of the largest calibration circle in the target object based on one of the calibration binocular images includes: Performing contour extraction on one of the calibration binocular images to obtain multiple calibration circle contour images; Performing Hough transformation on the multiple calibration circle contour images to obtain multiple calibration circle roundnesses; Filtering the multiple calibration circle roundnesses based on a preset roundness threshold to obtain the filtered multiple calibration circle roundnesses; Based on the filtered multiple calibration circle roundnesses, determining the contour of the largest calibration circle in the target object to obtain the two-dimensional coordinates of the circle boundary of the largest calibration circle.

4. The binocular vision verification method according to claim 1, wherein The performing center extraction on the calibration binocular images to obtain three-dimensional coordinates of the center includes: Determining the contour of the smallest calibration circle in the calibration binocular images, and obtaining all left-eye center pixels and right-eye center pixels corresponding to the contour of the smallest calibration circle; Performing mean calculation on the left-eye center pixels to obtain left-eye center coordinates, and performing mean calculation on the right-eye center pixels to obtain right-eye center coordinates; Based on the left-eye center coordinates and the right-eye center coordinates, obtaining three-dimensional coordinates of the center.

5. The binocular vision verification method according to claim 1, wherein The converting the two-dimensional coordinates of the circle boundary into three-dimensional coordinates of the circle boundary includes: Converting the calibration binocular images into a target image, where the target image is one of a depth image, a disparity image, and a three-dimensional point cloud image; Mapping the two-dimensional coordinates of the circle boundary according to the target image to obtain three-dimensional coordinates of the circle boundary.

6. The binocular vision verification method according to claim 1, wherein The calibration result includes at least one of a reconstruction integrity calibration result, a dimensional accuracy calibration result, and a planar distribution accuracy calibration result. The performing accuracy calibration on the target binocular vision device based on the three-dimensional coordinates of the circle boundary and the three-dimensional coordinates of the center to obtain a calibration result includes: Determining the number of first boundary points and the number of second boundary points, where the first boundary point is the point corresponding to the two-dimensional coordinates of the circle boundary, and the second boundary point is the point corresponding to the three-dimensional coordinates of the circle boundary; Based on the number of the first boundary points and the number of the second boundary points, obtaining a reconstruction integrity calibration result; Based on the three-dimensional coordinates of the circular boundary, the three-dimensional coordinates of the center of the circle, and the standard radius of the maximum verification circle, calculate the cumulative probability distribution of the dimension and calculate the cumulative probability distribution of the error value to obtain the dimension accuracy verification result, where the dimension is the distance between the second boundary point and the center of the circle, and the error value is the difference between the dimension and the standard radius of the maximum verification circle; Construct a fitting plane according to the second boundary point; Based on the three-dimensional coordinates of the circular boundary, calculate the cumulative probability distribution of the perpendicular distance from each second boundary point to the fitting plane to obtain the plane distribution accuracy verification result.

7. The binocular vision verification method according to claim 1, wherein The obtaining the two-dimensional coordinates of the circular boundary of the maximum verification circle in the target object based on one of the verification binocular images includes: Perform binarization processing on one of the verification binocular images to obtain a target binary image; Based on the target binary image, obtain the two-dimensional coordinates of the circular boundary of the maximum verification circle in the target object.

8. A control device, characterized in that, Comprising: A memory configured to store instructions; A processor configured to call the instructions from the memory and capable of implementing the binocular vision verification method according to any one of claims 1 to 7 when executing the instructions.

9. An image acquisition device, characterized in that, Comprising: A control device according to claim 8; A binocular vision device configured to acquire binocular vision images.

10. A machine-readable storage medium, characterized in that, Instructions are stored on the machine-readable storage medium, and the instructions are used to cause the machine to execute the binocular vision verification method according to any one of claims 1 to 7.

Images