Wave trough area identification method and system of corrugated plate

Abstract

The invention provides a wave trough area identification method and system for a corrugated plate, and the method comprises the steps: S1, collecting a first corrugated plate image and a second corrugated plate image through a left camera and a right camera of a binocular camera respectively, and S2, dividing the first corrugated plate image and the second corrugated plate image into a plurality of areas, s3, generating a disparity map of the corrugated plate based on the first corrugated plate image and the second corrugated plate image, S4, traversing a disparity value of each pixel point in the disparity map, calculating a distance value through the disparity value, judging whether the distance value of the pixel points in each region continuously changes or not, if not, dividing the pixel points into two groups according to the distance value of the pixel points in each region, and if yes, determining whether the distance value of the pixel points in each region changes continuously; and determining the region where the group with the larger distance value is located as a trough region. According to the wave trough area identification method for the corrugated plate, the wave trough area can be rapidly and accurately identified, and the situation that when spot welding is conducted on the wave trough area of the corrugated plate later, the spot welding position is wrong, and the spot welding efficiency is affected is avoided.

Description

Technical Field

[0001] The invention belongs to the technical field of corrugated plate welding, and in particular, relates to a method and system for identifying a trough area of ​​a corrugated plate. Background Art

[0002] The subway's chassis and ceiling are formed by welding multiple corrugated plates in parallel. The corrugated plate is a wavy metal plate, usually made of steel, aluminum or stainless steel, with a series of wavy or concave-convex shapes on the surface. It performs well in bearing heavy pressure, weathering resistance and adapting to complex environments, so it is used in the processing of subway ceilings and chassis.

[0003] During the processing of stainless steel subways, corrugated plates need to be connected to curved beams or other components through robot spot welding. Usually, when spot welding corrugated plates, the spot welding position is required to be at the trough of the corrugated plates, and the spot welding position cannot be placed at the crest or slope of the corrugated plates. Due to the short distance between adjacent crests, the spot welding position is often incorrect during the spot welding process, which reduces the accuracy of spot welding and further reduces the efficiency of spot welding.

[0004] In view of this, the present invention is proposed. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art. The first purpose is to provide a method for identifying the trough area of ​​a corrugated plate. The above method can quickly and accurately identify the trough area, thereby avoiding the occurrence of incorrect spot welding position when spot welding the trough area of ​​the corrugated plate, thereby affecting the efficiency of spot welding.

[0006] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0007] A method for identifying a trough region of a corrugated plate, comprising:

[0008] S1, a first corrugated plate image and a second corrugated plate image captured by the left and right cameras of the binocular camera respectively;

[0009] S2, dividing the first corrugated board image and the second corrugated board image into a plurality of regions;

[0010] S3, generating a disparity map of the corrugated board based on the first corrugated board image and the second corrugated board image;

[0011] S4, traversing the disparity values ​​of each pixel point in the disparity map, calculating the distance value through the disparity value, and determining whether the distance value of the pixel points in each area changes continuously. If not, the pixels in each area are divided into two groups according to the distance values, and the area where the group with the larger distance value is located is determined as the trough area.

[0012] Preferably, step S4 comprises,

[0013] Generate preset straight lines located in various areas on the disparity map;

[0014] Traverse the disparity values ​​of the pixel points on the preset straight line and calculate the distance value based on the disparity values;

[0015] If it is determined that the distance values ​​of the pixels on the preset straight line do not change continuously in each area, the pixels on the preset straight line are divided into two groups according to their distance values, and the area where the coordinates of the pixels in the group with a larger distance value are located is determined as the trough area.

[0016] Preferably, the preset straight line is a first straight line and a second straight line that are perpendicular to each other;

[0017] Obtain the distance values ​​of the pixel points on the first straight line and the second straight line respectively;

[0018] When it is determined that the distance value of the pixel points on the first straight line is greater than the threshold and the distance value of the pixel points on the second straight line is less than the threshold, the distance value of the pixel points on the first straight line is retained; otherwise, the distance value of the pixel points on the second straight line is retained.

[0019] Preferably, determining that the distance value of the pixel points on the first straight line is greater than a threshold and the distance value of the pixel points on the second straight line is less than a threshold includes:

[0020] Calculate a first difference between one of the pixel points on the first straight line and the other pixel points on the first straight line;

[0021] Calculate a second difference between one of the pixel points on the second straight line and the other pixel points on the second straight line;

[0022] When it is determined that the first difference is greater than the threshold and the second difference is less than the threshold, the distance value of the pixel points on the first straight line is retained; otherwise, the distance value of the pixel points on the second straight line is retained.

[0023] Preferably, step S3 includes:

[0024] Find corresponding points in the second corrugated plate image taken by the right camera of the binocular camera that match the pixel points in the first corrugated plate image taken by the left camera of the binocular camera;

[0025] Calculate the gray value difference between the pixel and the corresponding point, and generate a disparity map based on the principle of the binocular camera.

[0026] Preferably, step S2 includes:

[0027] Dividing the first corrugated plate image and the second corrugated plate image into a plurality of grids;

[0028] Get the gray value of each grid center point;

[0029] Calculate the difference between the grayscale value of one of the grid center points and the grayscale value of the center point adjacent to the grid center point to form a grayscale value difference;

[0030] If it is determined that the gray value difference is less than the preset gray value, the grids where the two center points are located are merged into one area until all grids cannot be merged to form an area.

[0031] Preferably, in step S1, the left and right cameras of the binocular camera respectively capture the first corrugated plate image and the second corrugated plate image, and the left and right cameras of the binocular camera are arranged directly above the corrugated plate.

[0032] Preferably, step S1 includes preprocessing the first corrugated board image and the second corrugated board image;

[0033] Preferably, removing noise from the first corrugated board image and the second corrugated board image;

[0034] The first corrugated plate image and the second corrugated plate image are equally divided into three images of RGB channels;

[0035] Recombining the three images of the RGB channels divided by the first corrugated plate image after equalization processing and recombining the three images of the RGB channels divided by the second corrugated plate image after equalization processing;

[0036] Preferably, a split function is used to split the first corrugated plate image into three images of RGB channels and to split the second corrugated plate image into three images of RGB channels;

[0037] Preferably, a merge function is used to recombine the three images of RGB channels into which the first corrugated board image is divided and to recombine the three images of RGB channels into which the second corrugated board image is divided after processing.

[0038] Preferably, finding corresponding points in the second corrugated plate image taken by the right camera that match pixel points in the first corrugated plate image taken by the left camera includes:

[0039] Obtain pixel points in the first corrugated plate image taken by the left camera;

[0040] Obtain corresponding points in the second corrugated plate image taken by the right camera;

[0041] Calculate the variance of pixels and corresponding points;

[0042] If the variance is within the preset variance range, the pixel point matches the corresponding point.

[0043] A second object of the present invention is to provide a system for identifying the valley area of ​​a corrugated board.

[0044] After adopting the above technical scheme, the present invention has the following beneficial effects compared with the prior art.

[0045] In the present invention, the first corrugated plate image and the second corrugated plate image taken by the left and right cameras of the binocular camera are divided into multiple areas, and the disparity map of the corrugated plate is generated according to the first corrugated plate image and the second corrugated plate image, and the disparity values ​​of each pixel point in the disparity map are traversed, and the distance value is calculated by the disparity value, and it is determined whether the distance value of the pixel points in each area changes continuously. If not, it is divided into two groups according to the distance value of the pixel points in each area, and the area where the distance value is larger is determined as the trough area. The above-mentioned recognition method can accurately identify the trough area, ensure the accuracy of the spot welding position, and further improve the overall quality and efficiency of spot welding. In addition, the entire spot welding process does not require manual intervention, which improves the intelligent level of spot welding.

[0046] In the present invention, the distance values ​​of the pixel points on the first straight line and the second straight line are respectively obtained. When it is determined that the distance value of the pixel points on the first straight line is greater than a threshold value and the distance value of the pixel points on the second straight line is less than the threshold value, the distance value of the pixel points on the first straight line is retained, otherwise, the distance value of the pixel points on the second straight line is retained. The above method can be used to find a straight line perpendicular to the texture direction of the corrugated board, and the trough area can be further found according to the area where the pixel points on the straight line are located, thereby further improving the efficiency of determining the trough area.

[0047] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the drawings described below are only some embodiments. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the accompanying drawings:

[0049] Figure 1 is a flow chart of a method for identifying a trough region of a corrugated plate according to the present invention;

[0050] Figure 2 It is an overall block diagram of the trough area recognition system of the corrugated plate of the present invention.

[0051] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0053] In the description of the present invention, it should be noted that the directions or positional relationships indicated by terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “inside” and “outside” are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0054] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0055] like Figure 1 As shown, the present invention provides a method for identifying a trough area of ​​a corrugated plate, comprising:

[0056] S1, a first corrugated plate image and a second corrugated plate image captured by the left and right cameras of the binocular camera respectively;

[0057] S2, dividing the first corrugated board image and the second corrugated board image into a plurality of regions;

[0058] S3, generating a disparity map of the corrugated board based on the first corrugated board image and the second corrugated board image;

[0059] S4, traversing the disparity values ​​of each pixel point in the disparity map, calculating the distance value through the disparity value, and determining whether the distance value of the pixel points in each area changes continuously. If not, the pixels in each area are divided into two groups according to the distance values, and the area where the group with the larger distance value is located is determined as the trough area.

[0060] The above-mentioned trough area recognition method can accurately identify the trough area, ensure the accuracy of the spot welding position, and further improve the overall quality and efficiency of spot welding. In addition, the entire spot welding process does not require manual intervention, which improves the intelligence level of spot welding.

[0061] In addition, before the first corrugated plate image and the second corrugated plate image are respectively collected by the left and right cameras of the binocular camera, the binocular camera needs to be calibrated, and the internal parameters, external parameters, radial distortion coefficient and tangential distortion coefficient of the binocular camera are obtained by calibrating the binocular camera. The first corrugated plate image and the second corrugated plate image are corrected by the radial distortion coefficient and the tangential distortion coefficient. The specific calibration and correction process of the binocular camera belong to the existing technology and will not be repeated here.

[0062] Specifically, when the left and right cameras of the binocular camera are arranged directly above the corrugated plate, the first corrugated plate image and the second corrugated plate image are respectively collected, including preprocessing the first corrugated plate image and the second corrugated plate image.

[0063] Corrugated plate is a plate with a wavy surface, generally a metal part made of stainless steel or aluminum alloy. It has the advantages of light weight, high strength and strong corrosion resistance, and is widely used in the construction industry. Since there may be reflection on the surface of the corrugated plate when collecting images of the corrugated plate, in order to reduce the impact of the reflection on the surface of the corrugated plate on the quality of the collected corrugated plate image and enhance the contrast of the corrugated plate image, it is necessary to pre-process the collected first corrugated plate image and the second corrugated plate image.

[0064] Furthermore, Gaussian filtering technology is used to eliminate the noise of the corrugated plate image to ensure that the corrugated plate image is smoother.

[0065] Specifically, a Gaussian kernel is created, which is a two-dimensional Gaussian function matrix. The Gaussian kernel is applied to each pixel of the corrugated plate image, and the smoothed pixel value is obtained by calculating the weighted sum of the Gaussian kernel and the local area of ​​the corrugated plate image.

[0066] Preferably, a split function is used to split the first corrugated plate image into three images of RGB channels and to split the second corrugated plate image into three images of RGB channels;

[0067] Preferably, a merge function is used to divide the first corrugated board image into three images of RGB channels and to recombine the images, and to divide the second corrugated board image into three images of RGB channels and to recombine the images.

[0068] The first corrugated board image and the second corrugated board image are both divided into three images of RGB channels by using the split function, and the three images of the RGB channels of the first corrugated board image and the three images of the RGB channels of the second corrugated board image are subjected to histogram equalization processing, that is, the grayscale values ​​of the images are nonlinearly stretched by the cumulative distribution function to ensure that the pixel values ​​on the three images of the RGB channels of the first corrugated board image and the three images of the RGB channels of the second corrugated board image after processing are more evenly distributed, thereby further enhancing the overall contrast of the image.

[0069] In addition, since the peak areas and trough areas of the corrugated board appear alternately in the corrugated board image, and there will be obvious edges between the peak areas and the trough areas, the grayscale value will change. In order to avoid the influence of the different grayscale values ​​of the peak areas and the trough areas, the first corrugated board image and the second corrugated board image are divided into multiple areas, that is, the grayscale values ​​of the areas formed after the division are approximately the same.

[0070] Specifically, the first corrugated board image and the second corrugated board image are divided into a plurality of regions by using the region growing method.

[0071] Furthermore, the first corrugated board image and the second corrugated board image are evenly divided into a plurality of grids, each of which is usually formed by a regular rectangle with a length of 5 to 9 pixels, in order to avoid the elimination of image edge information due to a grid being too small and the loss of image detail information due to a grid being too large. Each grid is formed by a regular rectangle with a length of 5 to 9 pixels, and each grid is smoothed using a convolution kernel not less than the size of the grid, so as to ensure that the gray value of the center point of each grid after processing can represent the gray value of the grid.

[0072] Furthermore, the grayscale value difference between one of the center point pixels and its adjacent center point pixel is calculated, and if it is determined that the grayscale value difference is less than a preset grayscale value, the grids where the two center point pixels are located are merged into one area until the grids where all the center point pixels are located cannot be merged with each other.

[0073] For example, according to the similarity criterion, select the center point of one of the grids, and record the gray value of the center point pixel of the grid as g1, record the center point pixel adjacent to the center point pixel of the grid as g2, calculate the gray value difference between the two, that is, g1-g2, and obtain the difference between the two |g1-g2|, record the preset gray value as Tolerance, and when it is judged that |g1-g2|<Tolerance, merge the two grids where the above two center point pixels are located into one area until all the grids where the center point pixels are located cannot be merged with each other. It can be understood that the value of Tolerance can be adjusted according to the actual situation. Preferably, Tolerance is set to 3.

[0074] Preferably, generating a disparity map of the corrugated plate based on the first corrugated plate image and the second corrugated plate image comprises finding corresponding points in the second corrugated plate image captured by the right camera of the binocular camera that match pixel points in the first corrugated plate image captured by the left camera of the binocular camera;

[0075] Specifically, obtaining pixel points in the first corrugated plate image and corresponding points in the second corrugated plate image;

[0076] Calculate the variance of pixels and corresponding points;

[0077] If the variance is within the preset variance range, the pixel point matches the corresponding point.

[0078] Furthermore, the first corrugated board image taken by the left camera is used as the reference image, and the second corrugated board image taken by the right camera is used as the search image. A rectangular frame of the same size is determined in the search image to search for each pixel point in the reference image, and find the corresponding point matching each pixel point. Preferably, the width of the rectangular frame is set to 2m+1, and the height of the rectangular frame is set to 2n+1. It can be understood that the width and height of the rectangular frame can be adaptively adjusted according to actual conditions.

[0079] In addition, the horizontal and vertical coordinates of each pixel point and the horizontal and vertical coordinates of the corresponding point can be found according to the pixel point in the first corrugated board image taken by the left camera and the corresponding point in the second corrugated board image taken by the right camera. The horizontal coordinate of the above pixel point is marked as r, and the vertical coordinate is marked as c. The horizontal coordinate of the corresponding point is marked as r' and the vertical coordinate is marked as c'. The grayscale value in the first corrugated board image taken by the left camera is recorded as g1, the grayscale value in the second corrugated board image taken by the right camera is recorded as g2, and the grayscale mean is recorded as

[0080] in,

[0081] The above variance formula is, The value of variance S is obtained by calculating the variance formula, and if the variance S is within the preset variance range, the above pixel points are matched with the corresponding points until all the pixel points in the first corrugated board image taken by the left camera can find corresponding points in the second corrugated board image taken by the right camera. Preferably, the preset variance range is between -1.0 and 1.0, including -1.0 and 1.0.

[0082] The grayscale value difference between each pixel of the first corrugated plate image taken by the left camera and the corresponding point matched with each pixel is calculated, and a disparity map is generated according to the principle of a binocular camera. It is understandable that the above method of generating a disparity map belongs to the prior art and will not be described in detail here.

[0083] Preferably, step S4 comprises,

[0084] Generate preset straight lines located in various areas on the disparity map;

[0085] Traverse the disparity values ​​of the pixel points on the preset straight line and calculate the distance value based on the disparity values;

[0086] If it is determined that the distance values ​​of the pixels on the preset straight line do not change continuously in each area, the pixels on the preset straight line are divided into two groups according to their distance values, and the area where the coordinates of the pixels in the group with a larger distance value are located is determined as the trough area.

[0087] Preferably, the preset straight line is a first straight line and a second straight line that are perpendicular to each other;

[0088] Obtain the distance values ​​of the pixel points on the first straight line and the second straight line respectively;

[0089] When it is determined that the distance value of the pixel points on the first straight line is greater than the threshold and the distance value of the pixel points on the second straight line is less than the threshold, the distance value of the pixel points on the first straight line is retained, otherwise, the distance value of the pixel points on the second straight line is retained. The above method can find a straight line perpendicular to the texture direction of the corrugated board, and further find the trough area according to the area where the pixel points on the straight line are located, thereby further improving the efficiency of determining the trough area.

[0090] Preferably, determining that the distance value of the pixel points on the first straight line is greater than a threshold and the distance value of the pixel points on the second straight line is less than a threshold includes:

[0091] Calculate a first difference between one of the pixel points on the first straight line and the other pixel points on the first straight line;

[0092] Calculate a second difference between one of the pixel points on the second straight line and the other pixel points on the second straight line;

[0093] When it is determined that the first difference is greater than the threshold and the second difference is less than the threshold, the distance value of the pixel points on the first straight line is retained; otherwise, the distance value of the pixel points on the second straight line is retained.

[0094] The method for locating the trough area of ​​the corrugated plate also includes, after determining it as the trough area, obtaining the three-dimensional coordinates of the center point of the trough area, and sending the three-dimensional coordinates of the center point of the trough area to the robot, and controlling the robot to spot weld the trough area of ​​the corrugated plate according to the three-dimensional coordinates of the center point of the trough area.

[0095] Specifically, the center point of the trough area is found through an image processing algorithm, and the horizontal and vertical coordinates of the center point of the trough area are obtained. Then, the distance value is calculated according to the disparity value of the center point of the trough area, and the vertical coordinate of the center point of the trough area is obtained, thereby obtaining the three-dimensional coordinates of the center point of the trough area.

[0096] Furthermore, the vertical coordinate of the center point of the trough area is the distance between the binocular camera and the corrugated plate, that is, the distance value calculated by the parallax value, and the distance d between the camera and the corrugated plate is obtained by the following calculation formula: Among them, d disparity is the disparity value of the center point of the trough area, x or,x ol is the column coordinates of one pixel point of the first corrugated plate image and the corresponding point matching the pixel point, f is the focal length of the binocular camera, and b is the baseline distance of the binocular camera.

[0097] Furthermore, the focal length f and the baseline distance b of the binocular camera can be obtained by calibrating the binocular camera. It can be understood that the calibration of the binocular camera also belongs to the prior art and will not be described in detail here.

[0098] In addition, before sending the three-dimensional coordinates of the center point of the trough area to the robot, the binocular camera coordinate system and the robot coordinate system need to be associated through hand-eye calibration.

[0099] Specifically, through cam H cal = cam H base · base H tool · tool H cal Realize the association between the binocular camera coordinate system and the robot coordinate system. cam H cal is the pose of the correction object in the binocular camera coordinate system, cam H base is the pose of the robot coordinate system in the binocular camera coordinate system, base H tool The pose of the robot tool in the robot coordinate system, tool H cal Correct the pose of the object coordinate system in the robot tool coordinate system. cam H cal and base H tool The calibration of the stereo camera and the robot properties are known, and tool H cal and cam H base Unknown. Through hand-eye calibration, the robot is moved to different positions to capture images of the calibration plate, so as to calculate the unknown transformation relationship of each coordinate system. In addition, the hand-eye calibration technology is an existing technology and will not be elaborated here.

[0100] like Figure 2 As shown, the present invention provides a trough area recognition system for a corrugated plate, comprising a collection module, a division module, a generation module, a judgment module, a grouping module, a determination module and an action module.

[0101] The acquisition module is used to acquire the first corrugated plate image and the second corrugated plate image;

[0102] The division module is used to divide the first corrugated board image and the second corrugated board image into a plurality of regions;

[0103] The generating module is used to generate a disparity map of the corrugated board based on the first corrugated board image and the second corrugated board image;

[0104] The judgment module is used to traverse the disparity values ​​of each pixel point in the disparity map, and calculate the distance value through the disparity value to determine whether the distance value of the pixel points in each area changes continuously. If not, the grouping module is used to divide the pixels into two groups according to the distance values ​​of the pixels in each area. The determination module determines that the area where the group with the larger distance value is located is the trough area. Then, the action module controls the robot to spot weld the trough area of ​​the corrugated plate according to the three-dimensional coordinates of the center point of the trough area.

[0105] The above description is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this patent can make some changes or modifications to equivalent embodiments of equivalent changes by using the technical content suggested above without departing from the scope of the technical solution of the present invention. The implementation scheme in the above embodiment can also be further combined or replaced. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the scope of the solution of the present invention.

Claims

1. A method for identifying the trough area of ​​a corrugated plate, characterized in that: include, S1, a first corrugated plate image and a second corrugated plate image captured by the left and right cameras of the binocular camera respectively; S2, dividing the first corrugated board image and the second corrugated board image into a plurality of regions; S3, generating a disparity map of the corrugated board based on the first corrugated board image and the second corrugated board image; S4, traversing the disparity values ​​of each pixel point in the disparity map, calculating the distance value through the disparity value, and determining whether the distance value of the pixel points in each area changes continuously. If not, the pixels in each area are divided into two groups according to the distance values, and the area where the group with the larger distance value is located is determined as the trough area.

2. The method for identifying the trough area of ​​a corrugated plate according to claim 1, characterized in that: Step S4 comprises, Generate preset straight lines located in various areas on the disparity map; Traverse the disparity values ​​of the pixel points on the preset straight line and calculate the distance value based on the disparity values; If it is determined that the distance values ​​of the pixels on the preset straight line do not change continuously in each area, the pixels on the preset straight line are divided into two groups according to their distance values, and the area where the coordinates of the pixels in the group with a larger distance value are located is determined as the trough area.

3. The method for identifying the trough area of ​​a corrugated plate according to claim 2, characterized in that: The preset straight lines are a first straight line and a second straight line that are perpendicular to each other; Obtain the distance values ​​of the pixel points on the first straight line and the second straight line respectively; When it is determined that the distance value of the pixel points on the first straight line is greater than the threshold and the distance value of the pixel points on the second straight line is less than the threshold, the distance value of the pixel points on the first straight line is retained; otherwise, the distance value of the pixel points on the second straight line is retained.

4. The method for identifying the trough area of ​​a corrugated plate according to claim 3, characterized in that: It is determined that the distance value of the pixel points on the first straight line is greater than a threshold value and the distance value of the pixel points on the second straight line is less than a threshold value, include, Calculate a first difference between one of the pixel points on the first straight line and the other pixel points on the first straight line; Calculate a second difference between one of the pixel points on the second straight line and the other pixel points on the second straight line; When it is determined that the first difference is greater than the threshold and the second difference is less than the threshold, the distance value of the pixel points on the first straight line is retained; otherwise, the distance value of the pixel points on the second straight line is retained.

5. The method for identifying the trough area of ​​a corrugated plate according to any one of claims 1 to 4, characterized in that: Step S3 comprises: Find corresponding points in the second corrugated plate image taken by the right camera of the binocular camera that match the pixel points in the first corrugated plate image taken by the left camera of the binocular camera; Calculate the gray value difference between the pixel and the corresponding point, and generate a disparity map based on the principle of the binocular camera.

6. The method for identifying the trough area of ​​a corrugated plate according to any one of claims 1 to 5, characterized in that: Step S2 comprises: Dividing the first corrugated plate image and the second corrugated plate image into a plurality of grids; Get the gray value of each grid center point; Calculate the difference between the grayscale value of one of the grid center points and the grayscale value of the center point adjacent to the grid center point to form a grayscale value difference; If it is determined that the gray value difference is less than the preset gray value, the grids where the two center points are located are merged into one area until all grids cannot be merged to form an area.

7. The method for identifying the trough area of ​​a corrugated plate according to any one of claims 1 to 6, characterized in that: Step S1, the left camera and the right camera of the binocular camera respectively capture a first corrugated plate image and a second corrugated plate image, and the left camera and the right camera of the binocular camera are set directly above the corrugated plate.

8. The method for identifying the trough area of ​​a corrugated plate according to any one of claims 1 to 7, characterized in that: Step S1, comprising: preprocessing the first corrugated board image and the second corrugated board image; Preferably, removing noise from the first corrugated board image and the second corrugated board image; The first corrugated plate image and the second corrugated plate image are equally divided into three images of RGB channels; Recombining the three images of the RGB channels divided by the first corrugated plate image after equalization processing and recombining the three images of the RGB channels divided by the second corrugated plate image after equalization processing; Preferably, a split function is used to split the first corrugated plate image into three images of RGB channels and to split the second corrugated plate image into three images of RGB channels; Preferably, a merge function is used to recombine the three images of RGB channels into which the first corrugated board image is divided and to recombine the three images of RGB channels into which the second corrugated board image is divided after processing.

9. The method for identifying the trough area of ​​a corrugated plate according to claim 5, characterized in that: Finding corresponding points in the second corrugated plate image taken by the right camera that match the pixel points in the first corrugated plate image taken by the left camera, including, Obtain pixel points in the first corrugated plate image taken by the left camera; Obtain corresponding points in the second corrugated plate image taken by the right camera; Calculate the variance of pixels and corresponding points; If the variance is within a preset variance range, the pixel point matches the corresponding point.

10. A corrugated plate trough area recognition system, characterized in that: A method for identifying a trough area of ​​a corrugated plate as claimed in any one of claims 1 to 9.

Images