A method, apparatus and system for calibrating a line structured light device

Abstract

This application relates to a calibration method, apparatus, and system for line structured light devices. The method includes: acquiring a calibration image captured by a camera onto a calibration board; the calibration image is an image containing positioning markers captured by the camera onto the calibration board; the calibration board coincides with the laser surface of the line structured light device and is provided with an array of markers, including at least one set of positioning markers with identification information; performing image recognition on the calibration image to obtain the image coordinates of the markers; determining a mapping relationship model between the image coordinates and the spatial coordinates based on the obtained image coordinates and spatial coordinates of the positioning markers; and matching the image coordinates and spatial coordinates of the markers according to the image coordinates of each marker and the mapping relationship model. This method only requires capturing a calibration image with positioning markers using a camera, reducing the requirements for capturing the calibration board, simplifying operation, and improving calibration convenience.

Description

Technical Field

[0001] This application relates to the field of visual measurement system calibration technology, and in particular to a calibration method, apparatus and system for line structured light devices. Background Technology

[0002] 3D vision is the identification and perception of spatial targets and is currently a crucial solution for "lights-out" factories. The foundation of 3D vision is three-dimensional reconstruction, and line structured light is an important method for 3D reconstruction, primarily used for the 3D reconstruction of refined and miniaturized targets. Line structured light equipment has wide applications in measurement and defect detection. It also has significant value in the steel smelting industry, capable of scanning and diagnosing faults in rotating equipment, such as misalignment of trolley wheels and ring cooler wheels; it can also monitor faults in assembly line equipment, such as belt tear detection and trolley sideboard tilt detection. Line structured light calibration is the core of 3D scanning. Traditional line structured light equipment calibration typically uses a checkerboard calibration board. The disadvantage of this method is that the entire calibration board must be photographed during calibration, and the tilt angle of the calibration board cannot be too large, otherwise it will hinder checkerboard detection. This results in low calibration convenience, especially in harsh environments such as those in the metallurgical industry, where a convenient calibration solution is crucial. Summary of the Invention

[0003] Therefore, it is necessary to provide a calibration method, apparatus, and system for line structured light devices that can improve calibration convenience in response to the above problems.

[0004] A calibration method for a line structured light device includes:

[0005] Acquire a calibration image captured by a camera on a calibration board; the calibration image is an image containing positioning markers captured by the camera on the calibration board; the calibration board coincides with the laser surface of the line structured light device and is provided with an array of markers, the markers including at least one set of positioning markers with identification information;

[0006] Perform image recognition on the calibration image to obtain the image coordinates of the marker points;

[0007] Based on the obtained image coordinates of the positioning markers and the spatial coordinates of the positioning markers, a mapping relationship model between image coordinates and spatial coordinates is determined.

[0008] The image coordinates and spatial coordinates of the marker points are matched based on the image coordinates of each marker point and the mapping relationship model.

[0009] In one embodiment, the step of performing image recognition on the calibration image to obtain the image coordinates of the marker points includes:

[0010] The calibration image is sequentially filtered and binarized, and the outline of the marker points in the binarized image is determined.

[0011] The marker points are filtered based on their outlines, and the center positions of the filtered marker points are obtained to obtain the image coordinates.

[0012] In one embodiment, the marker point is a concentric circle composed of a black outer circle and a white inner circle. Determining the contour of the marker point in the binarized image includes: performing black-and-white inversion processing on the binarized image, and then selecting the contour of all marker points through a contour search algorithm.

[0013] In one embodiment, the step of filtering the marker points based on their contours and obtaining the center positions of the filtered marker points to obtain image coordinates includes: fitting an ellipse based on the contours of the marker points, calculating the sum of distance deviations between the contour points and the corresponding fitted ellipse points, removing marker points whose sum of distance deviations is greater than a set threshold or whose sub-contours are not present, and obtaining the center positions of the filtered marker points to obtain image coordinates.

[0014] In one embodiment, determining the mapping model between image coordinates and spatial coordinates based on the acquired image coordinates of the positioning markers and the spatial coordinates of the positioning markers includes:

[0015] Information is identified based on the outline of the marker points, and the positioning marker points are selected.

[0016] Based on the image coordinates and spatial coordinates of the positioning markers, the parameter values ​​of the mapping relationship between the image coordinates and spatial coordinates are determined, thus obtaining the mapping relationship model between the image coordinates and spatial coordinates.

[0017] In one embodiment, matching the image coordinates and spatial coordinates of the marker points based on their image coordinates and the mapping model includes:

[0018] Substitute the image coordinates of each marker point into the mapping relationship model to calculate the predicted coordinates of each marker point;

[0019] Find the coordinates among all the spatial coordinates of the marker points that are closest to the predicted coordinates of the marker point, and use them as the spatial coordinates corresponding to the marker point.

[0020] In one embodiment, after finding the coordinate closest to the predicted coordinate of the marker among all spatial coordinates of the marker, the method further includes: removing markers whose predicted coordinates are greater than a distance threshold from their spatial coordinates.

[0021] In one embodiment, the calibration image is a single calibration image captured by the camera of the entire calibration board; or, the calibration image includes a block calibration image obtained by dividing the captured single image into blocks according to a set interval, wherein the interval is determined according to the spacing between each marker point on the calibration board.

[0022] A calibration device for a line structured light device, comprising:

[0023] The image acquisition module is used to acquire a calibration image captured by the camera on the calibration board. The calibration image is an image containing positioning markers captured by the camera on the calibration board. The calibration board coincides with the laser surface of the line structured light device and is provided with an array of markers. The markers include at least one set of positioning markers with identification information.

[0024] The image recognition module is used to perform image recognition on the calibration image and obtain the image coordinates of the marker points;

[0025] The data processing module is used to determine the mapping relationship model between image coordinates and spatial coordinates based on the acquired image coordinates of the positioning markers and the spatial coordinates of the positioning markers.

[0026] The coordinate matching module is used to match the image coordinates and spatial coordinates of the marker points according to the image coordinates of each marker point and the mapping relationship model.

[0027] A calibration system for a line structured light device includes a camera, a calibration board, and a processor. The calibration board coincides with the laser surface of the line structured light device and is provided with an array of marker points, including at least one set of positioning marker points with identification information. The processor calibrates the line structured light device according to the method described above.

[0028] The above-mentioned calibration method, device, and system for line structured light equipment only require the use of a camera to capture calibration images with positioning markers, the acquisition of image coordinates of the markers through image recognition, the determination of a mapping relationship model between image coordinates and spatial coordinates by combining the acquired image coordinates and spatial coordinates of the positioning markers, and then matching the image coordinates and spatial coordinates of the markers with the image coordinates of other markers and the mapping relationship model to complete the calibration of the line structured light equipment. This reduces the requirements for capturing calibration boards, facilitates operation by staff, and improves the convenience of calibration. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating a calibration method for a linear structured light device in one embodiment.

[0030] Figure 2 This is a schematic diagram of a calibration board in one embodiment;

[0031] Figure 3 This is a flowchart illustrating the process of performing image recognition on a calibration image and obtaining the image coordinates of marker points in one embodiment.

[0032] Figure 4 This is a flowchart illustrating a model for determining the mapping relationship between image coordinates and spatial coordinates based on the obtained image coordinates and spatial coordinates of the positioning markers in one embodiment.

[0033] Figure 5 This is a schematic diagram of the process of matching the image coordinates and spatial coordinates of a marker point according to the image coordinates and mapping relationship model of each marker point in one embodiment;

[0034] Figure 6 This is a structural block diagram of a linear structured light device calibration device in one embodiment. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0036] 3D reconstruction is a key technology in computer vision, capable of being integrated with numerous industry applications. For instance, most traditional energy companies in sectors like coal, steel, and mining have long and extremely complex production processes that are virtually invisible and intangible. Problems are often handled primarily based on worker experience. However, by employing 3D reconstruction technology, combined with IoT, big data, visualization, and digital twin technologies, the true production processes can be reconstructed, accelerating the transformation and upgrading of the entire industry. In recent years, information technology applications have become dominant. Increased national environmental protection efforts and optimized energy consumption structures have further compelled companies to pursue green and intelligent transformation. Therefore, vigorously promoting the application of 3D real-scene reconstruction technology is crucial for driving production towards less-manned and unmanned operations. Intelligent management empowers high-quality enterprise development, seizing new opportunities in digital transformation and promoting green, low-carbon, and high-quality development in the steel industry.

[0037] Currently, the calibration method for line structured light typically uses a checkerboard calibration board. The disadvantage of this method is that the entire calibration board must be photographed during calibration, and the tilt angle of the calibration board cannot be too large, which is not conducive to the detection of the checkerboard pattern. The calibration method also uses multi-frame calibration line acquisition and fitting the laser plane. This calibration method not only requires multiple frames of images, but also has complex calculations, and errors in the process can cause calibration errors.

[0038] Based on this, this application provides a calibration method for line structured light devices. The method involves acquiring calibration images captured by a camera on a calibration board, performing image recognition on the calibration images to obtain the image coordinates of the marker points, determining a mapping model between the image coordinates and spatial coordinates based on the acquired image coordinates and spatial coordinates of the marker points, and matching the image coordinates and spatial coordinates of each marker point according to the mapping model. A calibration board with positioning marker points is used, enabling recognition even without capturing the entire image. The positioning marker points are located at the center of the calibration board. Furthermore, this calibration board offers the advantage of more complete coverage of the image area, achieving more accurate calibration. The calibration method employs a single-frame calibration mode, suitable for applications with less stringent scanning accuracy requirements. Additionally, it supports a global transformation parameter plus local precision block calibration method to ensure calibration accuracy in key areas.

[0039] In one embodiment, such as Figure 1 As shown, a calibration method for a line structured light device is provided, including:

[0040] Step S110: Obtain the calibration image captured by the camera on the calibration board. The calibration image is an image containing positioning markers captured by the camera on the calibration board; the calibration board coincides with the laser surface of the line structured light device and is provided with an array of markers, including at least one set of positioning markers with identification information.

[0041] Specifically, a processor can connect to a camera to receive calibration images captured by the camera. The camera can be an industrial camera, with a line structured light device as the light source. A line laser can be used to emit laser lines, and the camera collects the line stripe data. The camera can capture a complete image of the calibration board or a partial image; it only needs to capture the positioning markers with identification information. The shape of the markers on the calibration board, and the type of identification information carried by the positioning markers, are not unique. The positioning markers can be placed at the center of the calibration board for the camera to capture. Figure 2 As shown, the markers can be designed as concentric circles composed of a black outer circle and a white inner circle. The identification information carried by the positioning markers can be letters, or other arbitrary identifiers, shapes, etc., such as numbers: 1, 2, 3, 4, Chinese characters, squares, etc. In this embodiment, small concentric black and white circles are used as markers, and the markers are arranged in a rectangle to form a calibration plate. The number of markers can be selected according to the camera's field of view. The four circles in the center of the calibration plate are positioning markers, and each positioning marker has a letter set in its center: A, B, C, and D. These positioning markers are used to locate the entire calibration point. The four central positioning markers must be photographed every time calibration is performed, while the other markers do not need to be photographed completely.

[0042] Step S120: Perform image recognition on the calibration image to obtain the image coordinates of the marker points. Correspondingly, after acquiring the calibration image captured by the camera on the calibration board, the processor performs image recognition on the calibration image and analyzes the image coordinates of each marker point, including the positioning marker points, in the calibration image. It is understood that the method of performing image recognition on the calibration image is not unique; in one embodiment, such as... Figure 3 As shown, step S120 includes steps S122 and S124.

[0043] Step S122: The calibration image is sequentially filtered and binarized, and the contours of the marker points in the binarized image are determined. Specifically, the calibration image is first filtered using a Gaussian algorithm, and then binarized based on a threshold value, which is the average brightness value of the filtered image. This means that white areas are brighter than the average brightness, and black areas are less bright than the average brightness, thus directly segmenting the black and white areas. Then, a contour-finding algorithm is used to determine the contours of the marker points in the binarized image.

[0044] Further, step S122, determining the contours of the marker points in the binarized image, includes: inverting the black and white values ​​of the binarized image, and then selecting the contours of all marker points using a contour search algorithm. Inverting the black and white values ​​of the binarized image, i.e., making the black portion of the marker point the maximum value, facilitates contour search. After inverting the black and white values ​​of the image, the contours of all marker points are selected using a contour search algorithm.

[0045] Step S124: Filter the marker points based on their outlines and obtain the center positions of the filtered marker points to get their image coordinates. After determining the outlines of the marker points in the image, the processor can filter the marker points based on their outlines, removing those that do not meet the requirements. It is understood that the specific method for filtering marker points is not unique and can be set according to actual needs.

[0046] In this embodiment, step S124 includes: fitting an ellipse based on the contour of the marker points; calculating the sum of distance deviations between the contour points and the corresponding fitted ellipse points; removing marker points whose sum of distance deviations is greater than a set threshold or who do not have sub-contours; and obtaining the center positions of the filtered marker points to obtain image coordinates. Specifically, after fitting an ellipse based on the contour points, the sum of distance deviations between the contour points and the corresponding fitted ellipse is calculated. If there are marker points with large sums of distance deviations, it indicates that the marker point is incomplete or is an abnormal detection point, and it is removed. At the same time, marker points without sub-contours (i.e., inner contours) are also removed. Finally, the center P of all remaining valid marker points, i.e., the center of the fitted ellipse, is calculated to determine the image coordinates of the valid marker points.

[0047] Step S130: Based on the obtained image coordinates and spatial coordinates of the positioning markers, determine the mapping relationship model between image coordinates and spatial coordinates. After determining the image coordinates of the calibration points on the calibration image, the processor selects the image coordinates of the positioning markers in the calibration image, combines them with the spatial coordinates of the positioning markers on the calibration board, analyzes the mapping relationship between image coordinates and spatial coordinates, and determines the mapping relationship model between image coordinates and spatial coordinates.

[0048] In one embodiment, such as Figure 4 As shown, step S130 includes steps S132 and S134.

[0049] Step S132: Based on the outline of the marker points, information is identified and the positioning marker points are selected. Specifically, all outline data can be selected and combined with a character recognition algorithm to detect the positioning marker points, that is, to identify the four letters A, B, C, and D.

[0050] Step S134: Based on the image coordinates and spatial coordinates of the positioning markers, determine the parameter values ​​of the mapping relationship between the image coordinates and spatial coordinates, and obtain the mapping relationship model between the image coordinates and spatial coordinates.

[0051] Once the calibration board is designed, the spatial coordinates of each marker point are determined. Assuming the horizontal distance between marker points is *l* and the vertical distance is *h*, with the top-left corner as the coordinate origin, the x-axis extending horizontally to the right, the y-axis extending vertically downwards, and the center of the first top-left corner as the origin, then the coordinates of the point in the *i*th column and the *j*th row are (i... lj h), this coordinate is the design coordinate, i.e., the spatial coordinate, such as Figure 1 The coordinates of point A are (3 l, h), similarly the coordinates of point B are (4 l,h).

[0052] Specifically, the image coordinates can be mapped to the spatial coordinates of the calibration plate according to the perspective projection algorithm, such as Equation (1). That is, the relationship established by special marker points (i.e., positioning marker points) is applied to the matching of non-special marker points.

[0053]

[0054] (1)

[0055] Where (u, v) are image coordinates, (x', y') are the transformed spatial coordinates; a11, a12, a21, a22, a31, a32 are rotations, and a13, a23, a33 are translations. Because perspective transformation is non-linear, it cannot be represented homogeneously; the perspective transformation matrix is ​​3. 3. Substitute the image coordinates and spatial coordinates of points A, B, C, and D into equation (1) to solve for 9 parameters a, and obtain the mapping relationship between the special marker points.

[0056] Step S140: Match the image coordinates and spatial coordinates of the marker points according to the image coordinates and mapping relationship model of each marker point. Correspondingly, after determining the mapping relationship model between image coordinates and spatial coordinates based on special marker points, the processor matches the image coordinates and spatial coordinates of other non-special marker points. In one embodiment, such as... Figure 5 As shown, step S140 includes steps S142 and S144.

[0057] Step S142: Substitute the image coordinates of each marker point into the mapping relationship model to calculate the predicted coordinates of each marker point.

[0058] Step S144: Find the coordinates among all the spatial coordinates of the marker points that are closest to the predicted coordinates of the marker point, and use them as the spatial coordinates of the marker point.

[0059] Furthermore, after step S144, step S140 also includes: removing markers whose distance between the predicted coordinates and spatial coordinates is greater than a distance threshold.

[0060] Specifically, by substituting the image coordinates of all detected marker points into equation (1), the mapping parameters adopt the mapping relationship between special marker points to calculate a predicted value of a spatial coordinate. Then, the nearest point between the predicted value of the spatial coordinate and all real spatial coordinates is found as the spatial coordinate point corresponding to the marker point. The spatial coordinate points of all marker points are found in turn to complete the matching of marker points. Among them, a distance threshold t is set for the nearest point. For individual marker points whose predicted value of spatial coordinates is greater than the distance threshold t from the nearest real spatial coordinate point, they are removed, that is, it is considered that such marker points have not matched the corresponding spatial coordinate points. In addition, the image coordinates and spatial coordinates of the matched marker points can be substituted into equation (1) again to optimize the parameters in the mapping relationship. In order to improve the fitting accuracy of the mapping relationship and eliminate the weight error caused by the unevenness of coordinate points, the matched image marker points can be uniformly sampled, and then the mapping relationship between the uniformly sampled image marker points and the corresponding spatial points can be calculated.

[0061] In one embodiment, the calibration image is a single image captured by a camera of the entire calibration board. In this embodiment, a single-image calibration method is used. The calibration board is placed on the laser plane, and the camera captures an image of the calibration board, capturing as much of the board as possible. By identifying calibration points and matching them with the coordinate points on the calibration board plane, a transformation relationship (perspective transformation of the plane) between the image coordinates and the calibration board plane is established.

[0062] Substituting the image coordinates and matching spatial coordinates of all detected marker points into equation (1), and optimizing nine parameters (DLT, a11, a12, a13, a21, a22, a23, a31, a32, a33) using least squares corresponding points, calibration parameters are generated, thus completing the system calibration part of the line structured light. In application, it is only necessary to transform the laser line points detected on the image onto the light plane through this mapping to complete the data scanning of that section.

[0063] In another embodiment, the calibration image includes a block calibration image obtained by sequentially dividing a single captured image into blocks at set intervals, the intervals being determined by the spacing between the marker points on the calibration board. In this embodiment, a block calibration image calibration method is used. Block calibration is essentially local calibration, which can overcome the influence of camera distortion to some extent, making the calculated results more accurate after calibration. That is, a mapping relationship is established locally, and then multiple mapping relationships are established for the entire region to achieve optimal local mapping.

[0064] The calibration plate is divided into multiple blocks based on its calibration points. Each block has a length, width, and number of marker points (m, n), where m and n >= 2, forming a window block. Starting from the starting point of the calibration plate, the sliding interval is based on the spacing between the calibration points. For example, the block is moved horizontally by one calibration point interval and vertically by one calibration point interval each time to divide the captured image into multiple block calibration images. Specifically, the window block can be moved horizontally by one calibration point interval each time to obtain the relevant block calibration images. Then, the window block is moved vertically down by one calibration point interval, and then horizontally by one calibration point interval each time to obtain the block calibration images. This process is repeated to divide a single captured image into blocks to obtain block calibration images.

[0065] Based on the calculation of local perspective transformation parameters in blocks according to the calibration plate, the total number of DLT parameters k that need to be calculated is:

[0066] K=(w-(m+1) / 2) (h-(n+1) / 2)

[0067] Where: w and h are the number of calibration points in the horizontal and vertical directions, respectively.

[0068] After calibration, a block region parameter selection mapping table can be established. The size of the mapping table is the same as the image size, and the local region DLT parameter is used in the block region of the image.

[0069] The above-described calibration method for line structured light equipment uses a labeled calibration board that can be automatically identified. The calibration method employs a single-frame calibration mode and supports a global transformation parameter calibration mode, suitable for applications where scanning accuracy requirements are not too high. It also supports a global transformation parameter plus localized precise block calibration method to ensure calibration accuracy in critical areas. The equipment is relatively simple to install, has low maintenance costs, requires few supporting devices, is stable and reliable, and can be calibrated on-site, enabling rapid application in the steel industry. It can achieve applications such as defect diagnosis and 3D scanning of critical equipment.

[0070] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0071] Based on the same inventive concept, this application also provides a line structured light device calibration apparatus for implementing the line structured light device calibration method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more line structured light device calibration apparatus embodiments provided below can be found in the limitations of the line structured light device calibration method described above, and will not be repeated here.

[0072] In one embodiment, such as Figure 6 As shown, a calibration device for a line structured light device is provided, including an image acquisition module 110, an image recognition module 120, a data processing module 130, and a coordinate matching module 140, wherein:

[0073] Image acquisition module 110 is used to acquire calibration images captured by the camera on the calibration board; the calibration image is an image containing positioning markers captured by the camera on the calibration board; the calibration board coincides with the laser surface of the line structured light device and is provided with an array of markers, the markers including at least one set of positioning markers with identification information.

[0074] The image recognition module 120 is used to perform image recognition on the calibration image and obtain the image coordinates of the marker points.

[0075] The data processing module 130 is used to determine the mapping relationship model between image coordinates and spatial coordinates based on the acquired image coordinates and spatial coordinates of the positioning markers.

[0076] The coordinate matching module 140 is used to match the image coordinates and spatial coordinates of the marker points according to the image coordinates and mapping relationship model of each marker point.

[0077] Each module in the aforementioned line structured light equipment calibration device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0078] In one embodiment, a line structured light device calibration system is also provided, including a camera, a calibration board, and a processor. The calibration board coincides with the laser surface of the line structured light device and is provided with an array of marker points, including at least one set of positioning marker points with identification information. The processor calibrates the line structured light device according to the method described above.

[0079] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0080] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described above.

[0081] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0082] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0083] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A calibration method for a line structured light device, characterized in that, include: Acquire calibration images of the calibration board captured by the camera; The line structured light device serves as a light-emitting device, and the camera is used to collect line stripe data. The calibration image is an image containing positioning markers taken by the camera onto the calibration plate. The markers are concentric circles composed of a black outer circle and a white inner circle. The calibration plate coincides with the laser surface of the line structured light device and is provided with an array of markers. The markers include at least one set of positioning markers with identification information. The calibration image is sequentially filtered and binarized. After the binarized image is reversed, the contours of all marker points are selected by a contour search algorithm. The marker points are then filtered based on their contours, and the center positions of the filtered marker points are obtained to obtain the image coordinates. Information is identified based on the contours of the markers, and the positioning markers are selected. Based on the image coordinates and spatial coordinates of the positioning markers, the parameter values ​​of the mapping relationship between the image coordinates and spatial coordinates are determined, and the mapping relationship model between the image coordinates and spatial coordinates is obtained. Substitute the image coordinates of each marker point into the mapping relationship model to calculate the predicted coordinates of each marker point; find the coordinates among all the spatial coordinates of the marker points that are closest to the predicted coordinates of the marker point, and use them as the spatial coordinates corresponding to the marker point; optimize the parameters in the mapping relationship by combining the image coordinates and spatial coordinates of the matched marker points, uniformly sample the matched image marker points, and then calculate the mapping relationship between the uniformly sampled image marker points and the corresponding spatial points.

2. The method according to claim 1, characterized in that, The step of filtering marker points based on their contours and obtaining the center positions of the filtered marker points to obtain image coordinates includes: fitting an ellipse based on the contours of the marker points, calculating the sum of distance deviations between the contour points and the corresponding fitted ellipse points, removing marker points whose sum of distance deviations is greater than a set threshold or whose sub-contours are not present, and obtaining the center positions of the filtered marker points to obtain image coordinates.

3. The method according to claim 1, characterized in that, After finding the coordinates closest to the predicted coordinates of all the marker points in their spatial coordinates, and using these coordinates as the spatial coordinates of the marker point, the process further includes: removing marker points whose predicted coordinates are more than a distance threshold from their spatial coordinates.

4. The method according to any one of claims 1-3, characterized in that, The calibration image is a single calibration image obtained by the camera taking a picture of the entire calibration board; or, the calibration image includes a block calibration image obtained by dividing the single image taken into blocks according to a set interval, wherein the interval is determined according to the spacing between the markers on the calibration board.

5. A calibration device for a line structured light equipment, characterized in that, include: The image acquisition module is used to acquire calibration images captured by the camera on the calibration board; The line structured light device serves as a light-emitting device, and the camera is used to collect line stripe data. The calibration image is an image containing positioning markers taken by the camera onto the calibration plate. The markers are concentric circles composed of a black outer circle and a white inner circle. The calibration plate coincides with the laser surface of the line structured light device and is provided with an array of markers. The markers include at least one set of positioning markers with identification information. The image recognition module is used to sequentially filter and binarize the calibration image, and after the binarized image is reversed, the contours of all the marker points are selected by the contour search algorithm; the marker points are filtered according to the contours of the marker points, and the center position of the filtered marker points is obtained to obtain the image coordinates. The data processing module is used to identify information based on the contour of the marker points and filter out the positioning marker points; based on the image coordinates and spatial coordinates of the positioning marker points, it determines the parameter values ​​of the mapping relationship between the image coordinates and spatial coordinates to obtain the mapping relationship model between the image coordinates and spatial coordinates. The coordinate matching module is used to substitute the image coordinates of each marker point into the mapping relationship model to calculate the predicted coordinates of each marker point; find the coordinates among all the spatial coordinates of the marker points that are closest to the predicted coordinates of the marker point, and use them as the spatial coordinates corresponding to the marker point; optimize the parameters in the mapping relationship by combining the image coordinates and spatial coordinates of the matched marker points; uniformly sample the matched image marker points; and then calculate the mapping relationship between the uniformly sampled image marker points and the corresponding spatial points.

6. A calibration system for a line structured light device, characterized in that, The device includes a camera, a calibration board, and a processor. The calibration board coincides with the laser surface of the line structured light device and is provided with an array of marker points, including at least one set of positioning marker points with identification information. The processor performs line structured light device calibration according to any one of claims 1-4.

Images