Machine vision-based methods, systems, and storage media for installing building panels

Abstract

The present application relates to a machine vision-based building board installation method, system and storage medium. The machine vision-based building board installation method includes: obtaining a first working image of a target building board; identifying the first working image to obtain feature points of the target building board; obtaining a flatness detection result of the target building board based on the feature points; and installing the building board based on the flatness detection result and a local optimal algorithm. Through the above means, it is no longer necessary to collaborate with multiple people and install based on visual inspection, but to determine the accurate installation posture through calculation, thereby achieving the effect of high efficiency and high installation accuracy.

Description

Technical Field

[0001] The present application relates to the technical field of machine vision and building board installation, and in particular to a machine vision-based building board installation method, system and storage medium. Background Art

[0002] In the prior art, manual installation is usually used to install building panels (such as galvanized steel panels, aluminum alloy panels, color-coated aluminum panels, metal embossed panels, etc.). In particular, when installing high-altitude building panels (such as color-coated steel panels on ceilings), it is necessary to build scaffolding or use aerial platforms to assist in the installation. When installing larger building panels, it requires the cooperation of multiple people. It has problems such as low efficiency, long construction period, and poor installation accuracy, and seriously slows down the installation efficiency. In addition, when building panels need to be installed over a large area, these building panels are easy to be uneven, and the height and angle of each building panel need to be frequently reviewed to meet the overall requirements in terms of installation height and large surface levelness. Summary of the invention

[0003] Based on this, it is necessary to provide a machine vision-based building panel installation method, system and storage medium to address the above technical problems.

[0004] In a first aspect, a method for installing a building board based on machine vision is provided, the method for installing a building board based on machine vision comprising:

[0005] Acquire a first working image of a target building board;

[0006] Recognize the first working image to obtain characteristic points of the target building board;

[0007] Obtaining a flatness detection result of the target building board based on the feature points;

[0008] Based on the flatness detection result and the local optimal algorithm, the building board is installed.

[0009] In a second aspect, a machine vision-based building board installation system is provided, the building board installation system comprising:

[0010] An acquisition module, used for acquiring a first operation image of a target building board;

[0011] A recognition module, used for recognizing the first operation image to obtain characteristic points of the target building board;

[0012] A flatness detection module, used to obtain a flatness detection result of the target building board based on the feature points;

[0013] The installation module is used to install the building board based on the flatness detection result and the local optimal algorithm.

[0014] According to a third aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the machine vision-based building board installation method according to the first aspect are implemented.

[0015] The above-mentioned machine vision-based building board installation method, system and storage medium, the method includes: obtaining a first working image of the target building board; identifying the first working image to obtain the characteristic points of the target building board; obtaining the flatness detection result of the target building board based on the characteristic points; installing the building board based on the flatness detection result and the local optimal algorithm. Through the above-mentioned means, it is no longer necessary for multiple people to collaborate and install based on visual inspection, but the accurate installation posture is determined by calculation, thereby achieving the effect of high efficiency and high installation accuracy. In addition, the flatness detection result of the target building board is obtained based on the characteristic points, and the building board is installed based on the flatness detection result and the local optimal algorithm, so that there is no need to frequently review the height and angle of each building board, so that the installation of the building board is easy to level, and the overall requirements in terms of installation height and large surface levelness are met. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a building board installation based on machine vision in one embodiment;

[0017] Figure 2 is a flow chart of a method for installing building panels based on machine vision in one embodiment;

[0018] Figure 3 A schematic diagram of Queen adjacency in one embodiment;

[0019] Figure 4 is a schematic diagram of a local reference building board plane in one embodiment;

[0020] Figure 5 A structural diagram of a machine vision-based building board installation system in one embodiment;

[0021] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0023] In the embodiment of the present invention, Figure 1 As shown, a building board 102 is placed above the robot chassis moving module 101. A building board 102-1 to be installed is clamped on the robot installation module (such as a robot arm) 103. The control box 104 is connected to the robot installation module 103 in communication, and has functions such as image recognition, data processing, flatness detection, and control. The camera device 105 captures and obtains the working image of the building board, and the control box 104 recognizes, extracts features, and analyzes the working images of the currently newly installed building board 102-2 and other installed building boards 102-3 (excluding 102-2), and can determine whether the currently newly installed building board 102-2 is installed flat. The control box 104 can also recognize, extract features, and analyze the working images of the currently newly installed building board 102-2 and other installed building boards 102-3 (excluding 102-2) to determine the installation position of the building board 102-1 to be installed. The control box 104 sends installation instructions to the robot installation module 103, and the robot installation module 103 installs the building board 102-1 to be installed.

[0024] In one embodiment, Figure 2 As shown, a method for installing a building board based on machine vision is provided, and the method for installing a building board based on machine vision comprises:

[0025] Step S201, obtaining a first operation image of a target building board.

[0026] Among them, the target building panels are the newly installed building panels.

[0027] In an embodiment of the present invention, a first operation image of a target building board is captured by a camera device, wherein the camera device needs to be calibrated in advance before installation, and the first operation image of the target building board is captured using the calibrated camera device.

[0028] The specific process of camera device calibration is as follows: install the camera device at a fixed position in an "eye-to-hand" form, so that the position relationship between the camera device and the base of the building board installation robot (hereinafter referred to as the robot) is fixed. The camera device calibration is completed by the robot hand-eye calibration method, so as to obtain the conversion relationship matrix between the image coordinate system of the camera device and the coordinate system of the robot end actuator, thereby completing the camera device calibration.

[0029] The camera device may be a depth camera.

[0030] Among them, the "eye-to-hand" mentioned above, in robotics, refers to the coordination between the robot's visual system (such as a camera or sensor) and its manipulator (robotic arm or hand), enabling the robot to perform precise operations based on visual information.

[0031] Among them, the hand-eye calibration method can be Zhang Zhengyou calibration method.

[0032] Step S202: Recognize the first working image to obtain feature points of the target building board.

[0033] The first operating image is processed to obtain an RGB-D image of the first operating image. After that, the RGB-D image is preprocessed, such as image denoising, image enhancement, image segmentation, etc., to complete the image preprocessing so that the building board can be better identified and detected later.

[0034] Optionally, the feature point may be a feature point Q of a building board. 1,2,…,n . Wherein, n is the total number of feature points. In this embodiment, n=4 is preferred, and the four feature points are the four corner points of the target building board. Optionally, the feature points can be the coordinates of the four corner points Q1, Q2, Q3, and Q4 of the building board, or the feature points can also be the coordinates of the center point O of the building board.

[0035] Step S203: obtaining a flatness detection result of the target building board based on the feature points.

[0036] The flatness detection results include flatness results and non-flatness results.

[0037] In the embodiment of the present invention, by analyzing the distance and / or angle of the target building board based on the feature points, it can be determined whether the target building board is flat or uneven.

[0038] Step S204: installing the building board based on the flatness detection result and the local optimal algorithm.

[0039] Wherein, based on the flatness detection result and the local optimal algorithm, the posture or height of the target building board can be adjusted, and / or the next building board can be installed.

[0040] In the embodiment of the present invention, when the target building board is flat, it means that the installation of the target building board meets the installation requirements and the installation is accurate, and the next building board can be installed based on the local optimal algorithm. When the target building board is not flat, it means that the installation of the target building board does not meet the installation requirements and the installation is inaccurate, and the posture of the target building board needs to be adjusted and optimized based on the local optimal algorithm and reinstalled.

[0041] The above-mentioned machine vision-based building board installation method includes: acquiring a first working image of a target building board; identifying the first working image to obtain feature points of the target building board; obtaining a flatness detection result of the target building board based on the feature points; and installing the building board based on the flatness detection result and a local optimal algorithm. Through the above-mentioned means, it is no longer necessary for multiple people to collaborate and install based on visual inspection, but the accurate installation posture is determined by calculation, thereby achieving the effect of high efficiency and high installation accuracy. In addition, the flatness detection result of the target building board is obtained based on the feature points, and the building board is installed based on the flatness detection result and a local optimal algorithm, so that there is no need to frequently review the height and angle of each building board, so that the installation of the building board is easy to level, and the overall requirements in terms of installation height and large surface levelness are met.

[0042] In an optional embodiment, step S203, obtaining a flatness detection result of the target building board based on the feature points, includes:

[0043] Step S2031: determining the overall height difference evaluation parameter between the target building board and the ideal installation plane based on the feature points.

[0044] In an optional embodiment, the step of obtaining the ideal installation plane includes:

[0045] Step a1, determining a calibration plate; wherein the calibration plate is the first building board to be installed.

[0046] Step a2, obtaining a second working image of the area where the calibration plate is located; recognizing the second working image to obtain feature points of the calibration plate.

[0047] Step a3: determining the ideal installation plane based on the feature points of the calibration plate.

[0048] In step a1, the first building board to be installed is defined as a calibration board, and the calibration board is installed by a robot, and the posture of the calibration board is adjusted so that the installation flatness meets the design requirements. Optionally, the first building board is required to accurately meet the installation height requirements and horizontal requirements, and can also be installed manually.

[0049] As mentioned above, the feature points can be the coordinates of the four corner points Q1, Q2, Q3, and Q4 of the building board. Based on the three-dimensional coordinates of the four corner points, the least squares plane P is fitted using the least squares method. i , which is regarded as the current plane of the calibration plate and the ideal installation plane of the building board. Set P i The expression is A i x+Bi y+C i z+D i =0, its unit normal vector is The subscript i in this paragraph indicates the i-th building board currently being installed. The least squares plane P of each building board during the installation process i The expression of and the unit normal vector n i All are recorded in the system (such as control box 104).

[0050] In step a2, a second working image of the area where the calibration plate is located is obtained, and image denoising, image enhancement, image segmentation, etc. are performed on the second working image to complete image preprocessing so as to better identify and detect the building board. The contour recognition of the building board is completed through the edge detection algorithm and the contour approximation method, and the three-dimensional coordinates of as many feature points on the building board as possible are extracted. Optionally, the three-dimensional coordinates of the extracted feature points include the three-dimensional coordinates of the four corner points of the building board.

[0051] In step a3, based on the three-dimensional coordinates of the four corner points Q1, Q2, Q3, and Q4 of the building board extracted in step a2, the least squares plane P is fitted using the least squares method. i , which is regarded as the current plane of the calibration plate and the ideal installation plane of the building board. Set P i The expression is A i x+B i y+C i z+D i =0.

[0052] In an optional embodiment, the overall height difference evaluation parameter between the target building board and the ideal installation plane can be calculated by the following method. Specifically: based on the feature point, determine the distance between the target building board and the ideal installation plane, and use the distance as the overall height difference evaluation parameter; wherein the feature point includes the coordinates of the center point of the target building board. For ease of understanding, an example is given here: based on the feature point Q described above 1,2,…,n (In this embodiment, the three-dimensional coordinates of the four corner points Q1, Q2, Q3, and Q4 are preferably used) and the distance between them and the ideal installation plane P is determined according to the point-to-plane distance calculation formula. i The distance d 1,2,…,n If the maximum allowable height difference of the building board installation specified in the construction technical documents is d max >d 1,2,…,n , it is deemed that the evaluation parameters of the overall height difference between the target building board and the ideal installation plane meet the requirements of step S2031, and the next step of judgment can be performed; otherwise, it is deemed that the evaluation parameters of the overall height difference between the target building board and the ideal installation plane do not meet the requirements of step S2031, and the target building material board needs to be adjusted.

[0053] Step S2032: determining local flatness evaluation parameters of the target building board and the local reference building board plane based on the feature points.

[0054] In an optional embodiment, based on the feature point, the angle between the target building board and the local reference building board plane (also referred to as the adjacent installed building board plane) is determined. The included angle is used as the local flatness evaluation parameter.

[0055] In the embodiment of the present invention, the angle between the target building board and the ideal installation plane and the local reference building board plane is calculated. As a judgment parameter of the local flatness of the target building board.

[0056] The local reference building board plane (also called the adjacent installed building board plane) includes: the plane formed by the building boards installed earlier than the target building board and connected to the target building board with the same vertices or common edges. The subscript i represents the i-th building board, which here represents the target building board, and nj represents the building board adjacent to the target building board. That is, the angle between the target building board and the building board with the serial number nj. After calculating the angle between the target building board and each serial number of the building board, the maximum angle is determined. The maximum angle is used as the local flatness evaluation parameter.

[0057] Step S2033, based on the overall height difference judgment parameter and the local flatness judgment parameter, obtain the flatness detection result of the target building board.

[0058] In an optional embodiment, if the overall height difference judgment parameter is less than a preset distance threshold, and the local flatness judgment parameter is less than a preset angle threshold, the flatness detection result is determined to be a flat result; if the overall height difference judgment parameter is greater than or equal to the preset distance threshold, and / or the local flatness judgment parameter is greater than or equal to the preset angle threshold, the flatness detection result is determined to be a non-flat result.

[0059] In an optional embodiment, when the flatness detection result is the non-flat result, the posture correction amount of the target building board is determined according to the local optimal algorithm described later; based on the posture correction amount and the height correction amount, the target building board is subjected to posture correction and height correction, and the step of acquiring the first working image of the target building board is returned to execute; when the flatness detection result is the flatness result, the installation posture and installation height of the next building board are determined using the local optimal algorithm based on the inverse distance weighted interpolation method; and the next building board is installed according to the installation posture and installation height.

[0060] In an optional embodiment, using a local optimal algorithm based on inverse distance weighted interpolation, before determining the installation posture and installation height of the next building board, it is necessary to first determine the number of covering layers, and determine the local reference building board plane based on the number of covering layers. Specifically:

[0061] like Figure 3 As shown in the figure, with the center point A of the target building board (essentially the target building board) as the center, based on the adjacency spatial dependency, the relationship between A and C1, C2, C3, C4 is called a common vertex connection, the relationship between A and B1, B2, D1, D2 is called a common edge connection, and the relationship between A and C1, C2, C3, C4, B1, B2, D1, D2 is called a common vertex and common edge connection (i.e., Queen connection). Figure 4 As shown in the figure, A is regarded as the target building board. Based on the Queen connection theory of spatial weight matrix, the size of a single building board is considered. With A as the center, a certain area around A is used as the building board distribution area that needs to be referenced for local flatness detection (i.e., the local reference building board plane).

[0062] The specific implementation method is as follows: Assume that the evaluation basis of local flatness is based on A as the center and length L C , Width W C The length and width of a single building board are L b , W b Building panels are generally rectangular. Assume that σL and σW are L C and W C The number of coating layers in the direction is: σL=(L C -L b ) / 2L b (Rounded to the nearest integer), σW=(W C -W b ) / 2W b (Rounded to the nearest integer). The number of cladding layers in the horizontal direction and the number of cladding layers in the vertical direction of the local reference building board plane are thereby determined, and the local reference building board plane is further determined.

[0063] In an optional embodiment, a local optimal algorithm based on an inverse distance weighted interpolation method is used to determine the installation posture and installation height of the next building board, including: determining the posture and height of a local reference building board in a local reference building board plane; determining a weight coefficient of each local reference building board based on the distance between the center point of the target building board and the center point of the next building board; determining the installation posture of the next building board based on the posture and height of the local reference building board and the weight coefficient of each local reference building board. Specifically:

[0064] According to the calculated number of covering layers, the local reference building board plane is determined; the weight coefficient of each local reference building board in the local reference building board plane is determined; the height and posture data of the above-mentioned local reference building board are read (or re-measured), and the installation posture and height (also called theoretical installation position) of the next building board are calculated according to the local optimal algorithm.

[0065] Among them, in addition to obtaining the installation posture and height of the next building board (also known as the theoretical installation position), the center point coordinate representation of the theoretical installation position and the normal vector representation of the theoretical installation position can also be obtained. Assume that the center point coordinates of the theoretical installation position are A = (xa, ya, za), and the normal vector of the theoretical installation position is nA = (Aa, Ba, Ca). The theoretical installation position of the next building board is regarded as the center, and according to the method in step S1032, the building board distribution area that needs to be referenced near the theoretical installation position of the next building board (i.e., the local reference building board plane) is determined, and the installed building boards in the local reference building board plane are identified. The three-dimensional coordinates Xi = (xi, yi, zi) and the normal vector nXi = (Ai, Bi, Ci) of the center point of each building board are extracted.

[0066] Since the installation flatness of the building board is mainly related to the height and the angle with the roof surface, it is assumed that the coordinate value error between the center point of the building board and the theoretical installation position on the x and y axes has little effect on the installation quality. That is, for the center point coordinate A = (xa, ya, za) of the theoretical installation position, it is assumed that the error between the center coordinates xa, ya of the actual installation position and the theoretical installation position does not affect the installation quality.

[0067] Therefore, the coordinates of the center point of each building board and the normal vector nXi are used as the attribute values ​​of the building board (see below for related discussion). The distance between the center point of each building board in the local reference building board plane and the center point of the theoretical installation position of the next building board is calculated, and the weight influence factor is obtained based on the distance conversion according to the inverse distance weighted interpolation method described below, so as to calculate the installation posture and height of the next building board (i.e., the guiding installation posture and height).

[0068] In the embodiment of the present invention, the number of transverse and longitudinal covering layers is 1, that is, for the next building board A to be installed, Figure 3 The installed panels in B1, B2, D1, D2, C1, C2, C3, and C4 are used as local reference building panels (those panels that have not been installed will no longer participate in the algorithm calculation, and building panel A will no longer be adjusted, even if they are installed later).

[0069] The local optimal algorithm proposed in the present invention can be used to calculate the installation height and posture of the next building board. The posture of the building board is represented by its unit normal vector. The calculation formula is as follows:

[0070] In the formula, Z A For the next building board (i.e. Figure 3 The installation height of the plate A) shown in FIG; j is the number of the local reference building plate; r is the number of the local reference building plate; Z r is the installation height of the rth local reference building board recorded previously; k r is the weight coefficient of the rth local reference building board.

[0071] It should be noted that the local optimal algorithm proposed in the present invention does not calculate the installation position of the next building board in the horizontal plane (i.e. the coordinates in the horizontal XOY plane). Instead, each building board is installed in a manner of vertical and horizontal alignment and close splicing, thereby determining the X and Y coordinate values.

[0072]

[0073] Where n A For the next building board (i.e. Figure 3 The installation unit normal vector of the plate A) shown in FIG; j is the number of the local reference building plate; r is the number of the local reference building plate; n r is the unit normal vector of the rth local reference building board recorded previously; k r is the weight coefficient of the rth local reference building board.

[0074] In this embodiment, the weight coefficient of each local reference building board is determined by using the inverse distance weight difference method, that is, for a local reference building board, the inverse of the distance between its center point and the center point of the installation position of the next building board is used as its weight coefficient.

[0075] The calculation formula is as follows: In the formula, k r is the weight coefficient of the rth local reference building board, d Or-OA Point O r With point O A Distance in the horizontal XOY plane. r is the center point of the rth local reference building board, O A For the next building board (i.e. Figure 3 The center point of the building board A) shown in FIG. A and O rBoth can be obtained by calculating the average value of the coordinates of the four corner points Q1, Q2, Q3, and Q4 extracted in step a2 of step S2031 of the building board.

[0076] The installation height and posture of the next building board can be calculated by the above formula. The building board is installed based on the installation height and posture of the next building board.

[0077] After the next building board is installed, the center point and plane normal vector of the installed building board are extracted, and the center point and plane normal vector of the installed building board are subtracted from the guiding installation posture to obtain the posture correction amount and height correction amount. The posture correction amount and height correction amount are used as input and converted into the control signal of the robot to control the robot to complete the installation correction of the installed building board.

[0078] Among them, when the flatness detection result is the non-flat result, the posture correction amount and the height correction amount of the target building board are determined according to the local optimal algorithm, and based on the posture correction amount and the height correction amount, the relevant content of the posture correction and the height correction of the target building board are the same means as the correction above, which will not be repeated here.

[0079] Among them, the inverse distance weighted interpolation method is a commonly used spatial interpolation method, which can infer unknown points through the attribute values ​​of known points. The attribute value of the unknown point is equal to the weighted average of the attribute value of the known point and the inverse of the distance. The closer the distance, the greater the corresponding weight, and the farther the distance, the smaller the corresponding weight. Figure 3 As shown, assuming that ABCD are the centers of the grid, the attribute values ​​at BCD are x(Xi), d(A-Xi) is the distance from A to Xi, where i=1, 2, 3...X=B, C, D. The inverse distance weighted interpolation method uses the inverse of the distance between Xi and A 1 / d(A-Xi) as the weight coefficient of Xi to A. The attribute value x(A) at A can be expressed as (the sum of the product of the weight coefficient at BCD and the attribute value divided by the sum of the weight coefficients at BCD).

[0080] In an embodiment of the present invention, the machine vision-based building board installation method can complete the installation flatness (also referred to as installation flatness quality) detection of the building board, and can output the posture correction amount and height correction amount, or output the guiding installation posture and height of the next building board to assist the robot in correction and / or installation. Thus, a quality control closed loop of "installation-visual quality detection-local optimal installation posture and height guidance" is formed in the building board installation process, which can effectively guarantee and improve the installation quality of the roof building board. In addition, the inverse distance weighted interpolation method is adopted, and the reciprocal of the distance between the center point of the theoretical installation position of the next building board and the center point of the surrounding installed building boards (the center point of each building board in the local reference building board plane) is used as the weight, combined with the posture and height of each local reference building board in the local reference building board plane, so as to obtain the local optimal guiding installation posture and height of the next building board, assist the building board in installation, and ensure the local installation flatness of the building board.

[0081] It should be understood that although Figure 2 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2 At least part of the steps may include multiple steps or multiple stages. These steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0082] In one embodiment, Figure 5 As shown, a machine vision-based building panel installation system is provided, comprising:

[0083] An acquisition module, used for acquiring a first operation image of a target building board;

[0084] A recognition module, used for recognizing the first operation image to obtain characteristic points of the target building board;

[0085] A flatness detection module, used to obtain a flatness detection result of the target building board based on the feature points;

[0086] The installation module is used to install the building board based on the flatness detection result and the local optimal algorithm.

[0087] In an optional embodiment, the flatness detection module is used to:

[0088] Based on the feature points, determining an overall height difference evaluation parameter between the target building board and an ideal installation plane;

[0089] Based on the feature points, determining local flatness evaluation parameters of the target building board and the local reference building board plane;

[0090] Based on the overall height difference judgment parameter and the local flatness judgment parameter, a flatness detection result of the target building board is obtained.

[0091] In an optional embodiment, the building board installation system further includes an ideal installation plane acquisition module, which is used to:

[0092] Determine a calibration plate; wherein the calibration plate is the first building board installed;

[0093] Acquire a second operating image of the area where the calibration plate is located;

[0094] Recognize the second working image to obtain feature points of the calibration plate;

[0095] The ideal installation plane is determined based on the feature points of the calibration plate.

[0096] In an optional embodiment, the flatness detection module is used to:

[0097] Based on the feature points, determine the maximum distance between the target building board and the ideal installation plane, and use the maximum distance as the overall height difference evaluation parameter; wherein the feature points include the coordinates of the corner points of the target building board;

[0098] or,

[0099] Based on the feature points, the distance between the target building board and the ideal installation plane is determined, and the distance is used as the overall height difference evaluation parameter; wherein the feature points include the coordinates of the center point of the target building board.

[0100] In an optional embodiment, the flatness detection module is used to:

[0101] Based on the feature points, the angle between the target building board and the local reference building board plane is determined, and the angle is used as the local flatness evaluation parameter.

[0102] In an optional embodiment, the building board installation system further includes a local reference building board plane acquisition module, which is used to:

[0103] Based on the spatial weight matrix, the local reference building board plane is determined with the target building board as the center.

[0104] In an optional embodiment, the flatness detection module is used to:

[0105] If the overall height difference judgment parameter is less than a preset distance threshold, and the local flatness judgment parameter is less than a preset angle threshold, then the flatness detection result is determined to be a flat result;

[0106] If the overall height difference judgment parameter is greater than or equal to the preset distance threshold, and / or the local flatness judgment parameter is greater than or equal to the preset angle threshold, the flatness detection result is determined to be a non-flat result.

[0107] In an optional embodiment, the installation module is used to:

[0108] When the flatness detection result is the non-flatness result, determining the posture correction amount and the height correction amount of the target building board according to the local optimal algorithm;

[0109] Based on the posture correction amount and the height correction amount, the target building board is subjected to posture correction and height correction, and the step of obtaining the first working image of the target building board is returned to be executed;

[0110] When the flatness detection result is the flatness result, the installation posture and installation height of the next building board are determined by using a local optimal algorithm based on an inverse distance weighted interpolation method;

[0111] The next building board is installed according to the installation posture and installation height.

[0112] In an optional embodiment, the installation module is used to:

[0113] determining a pose of a local reference building sheet in a local reference building sheet plane;

[0114] Determining a weight coefficient based on a distance between a center point of the target building board and a center point of the next building board;

[0115] Based on the posture of the local reference building board and the weight coefficient, the installation posture of the next building board is determined.

[0116] The relevant description of the above-mentioned building board installation system is consistent with the relevant description in the method embodiment, and will not be repeated here.

[0117] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a building board installation method based on machine vision is implemented.

[0118] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0119] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the machine vision-based building board installation method described in any of the above embodiments are implemented.

[0120] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0121] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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.

[0122] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for installing building panels based on machine vision, characterized in that: The machine vision-based building board installation method comprises: Acquire a first operation image of a target building board; wherein the target building board is a building board that has been newly installed; Recognize the first working image to obtain characteristic points of the target building board; Obtaining a flatness detection result of the target building board based on the feature points; wherein the flatness detection result includes a flatness result and a non-flatness result; Installing the building board based on the flatness detection result and the local optimal algorithm; The installation of the building board includes adjusting and installing the target building board and installing the next building board; wherein, based on the local optimal algorithm, the installation height and posture of the next building board are calculated; The posture of the building board is expressed by the unit normal vector, and the calculation formula is as follows: In the formula, Z A is the installation height of the next building board; j is the number of local reference building boards; r is the number of local reference building boards; Z r is the installation height of the rth local reference building board recorded previously; k r is the weight coefficient of the rth local reference building board; Where n A is the installation unit normal vector of the next building board; j is the number of local reference building boards; r is the number of local reference building boards; n r is the unit normal vector of the rth local reference building board recorded previously; k r is the weight coefficient of the rth local reference building board; Wherein, obtaining the flatness detection result of the target building board based on the feature points includes: Based on the feature points, determining an overall height difference evaluation parameter between the target building board and an ideal installation plane; Based on the feature points, local flatness evaluation parameters of the target building board and the local reference building board plane are determined; wherein the local reference building board plane includes: a plane formed by building boards installed earlier than the target building board and connected to the target building board with common vertices or common edges; wherein the local reference building board is a building board in the local reference building board plane; Based on the overall height difference judgment parameter and the local flatness judgment parameter, a flatness detection result of the target building board is obtained; The step of acquiring the ideal installation plane includes: determining a calibration plate; wherein the calibration plate is the first building board to be installed; acquiring a second operation image of the area where the calibration plate is located; recognizing the second operation image to obtain feature points of the calibration plate; and determining the ideal installation plane based on the feature points of the calibration plate; Wherein, determining the overall height difference evaluation parameter between the target building board and the ideal installation plane based on the feature points includes: determining the distance between the target building board and the ideal installation plane based on the feature points, and using the distance as the overall height difference evaluation parameter; wherein the feature points include the coordinates of the center point of the target building board; Determining the local flatness evaluation parameter of the target building board and the local reference building board plane based on the feature points includes: determining the angle between the target building board and the local reference building board plane based on the feature points, and using the angle as the local flatness evaluation parameter; Wherein, obtaining the flatness detection result of the target building board based on the overall height difference judgment parameter and the local flatness judgment parameter includes: If the overall height difference judgment parameter is less than a preset distance threshold, and the local flatness judgment parameter is less than a preset angle threshold, then the flatness detection result is determined to be a flat result; If the overall height difference judgment parameter is greater than or equal to the preset distance threshold, and / or the local flatness judgment parameter is greater than or equal to the preset angle threshold, the flatness detection result is determined to be a non-flat result.

2. The machine vision-based building board installation method according to claim 1, characterized in that: The installation of building panels based on the flatness detection result and the local optimal algorithm includes: When the flatness detection result is the non-flatness result, determining the posture correction amount and the height correction amount of the target building board according to the local optimal algorithm; Based on the posture correction amount and the height correction amount, the target building board is subjected to posture correction and height correction, and the step of obtaining the first working image of the target building board is returned to be executed; When the flatness detection result is the flatness result, the installation posture and installation height of the next building board are determined by using a local optimal algorithm based on an inverse distance weighted interpolation method; The next building board is installed according to the installation posture and installation height.

3. The machine vision-based building board installation method according to claim 2, characterized in that: The method of using a local optimal algorithm based on an inverse distance weighted interpolation method to determine the installation posture and installation height of the next building board comprises: determining a pose of a local reference building sheet in a local reference building sheet plane; Determining a weight coefficient of the local reference building board based on a distance between a center point of the target building board and a center point of the next building board; Based on the posture of the local reference building board and the weight coefficient, the installation posture of the next building board is determined.

4. A building board installation system based on machine vision, characterized in that: The building board installation system comprises: An acquisition module, used to acquire a first operation image of a target building board; wherein the target building board is a building board that has been newly installed; A recognition module, used for recognizing the first operation image to obtain characteristic points of the target building board; A flatness detection module, used to obtain a flatness detection result of the target building board based on the feature points; wherein the flatness detection result includes a flatness result and a non-flatness result; An installation module, used for installing the building board based on the flatness detection result and the local optimal algorithm; The installation of the building board includes adjusting and installing the target building board and installing the next building board; wherein, based on the local optimal algorithm, the installation height and posture of the next building board are calculated; The posture of the building board is expressed by the unit normal vector, and the calculation formula is as follows: In the formula, Z A is the installation height of the next building board; j is the number of local reference building boards; r is the number of local reference building boards; Z r is the installation height of the rth local reference building board recorded previously; k r is the weight coefficient of the rth local reference building board; Where n A is the installation unit normal vector of the next building board; j is the number of local reference building boards; r is the number of local reference building boards; n r is the unit normal vector of the rth local reference building board recorded previously; k r is the weight coefficient of the rth local reference building board; Wherein, obtaining the flatness detection result of the target building board based on the feature points includes: Based on the feature points, determining an overall height difference evaluation parameter between the target building board and an ideal installation plane; Based on the feature points, local flatness evaluation parameters of the target building board and the local reference building board plane are determined; wherein the local reference building board plane includes: a plane formed by building boards installed earlier than the target building board and connected to the target building board with common vertices or common edges; wherein the local reference building board is a building board in the local reference building board plane; Based on the overall height difference judgment parameter and the local flatness judgment parameter, a flatness detection result of the target building board is obtained; The step of acquiring the ideal installation plane includes: determining a calibration plate; wherein the calibration plate is the first building board to be installed; acquiring a second operation image of the area where the calibration plate is located; recognizing the second operation image to obtain feature points of the calibration plate; and determining the ideal installation plane based on the feature points of the calibration plate; Wherein, determining the overall height difference evaluation parameter between the target building board and the ideal installation plane based on the feature points includes: determining the distance between the target building board and the ideal installation plane based on the feature points, and using the distance as the overall height difference evaluation parameter; wherein the feature points include the coordinates of the center point of the target building board; Determining the local flatness evaluation parameter of the target building board and the local reference building board plane based on the feature points includes: determining the angle between the target building board and the local reference building board plane based on the feature points, and using the angle as the local flatness evaluation parameter; Wherein, obtaining the flatness detection result of the target building board based on the overall height difference judgment parameter and the local flatness judgment parameter includes: If the overall height difference judgment parameter is less than a preset distance threshold, and the local flatness judgment parameter is less than a preset angle threshold, then the flatness detection result is determined to be a flat result; If the overall height difference judgment parameter is greater than or equal to the preset distance threshold, and / or the local flatness judgment parameter is greater than or equal to the preset angle threshold, the flatness detection result is determined to be a non-flat result.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images