Negative plate perpendicularity detection method and terminal

Abstract

The invention provides a cathode plate verticality detection method and a terminal, and the method comprises the steps: obtaining a cathode plate image corresponding to a cathode plate when the cathode plate enters a detection region, and carrying out the edge extraction to obtain edge pixel points; the slopes of the adjacent edge pixel points relative to the detection area are solved, and the average value of the slopes is calculated to obtain the average slope; the perpendicularity corresponding to the negative plate is obtained according to the average slope; according to the method, the influence of factors such as noisy points brought in the image acquisition and processing process on the final perpendicularity test is avoided, the slope of the edge is obtained by taking the preset detection area as a reference, special detection equipment does not need to be arranged, and the detection cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of quality inspection, and in particular to a cathode plate verticality inspection method and terminal. Background Art

[0002] With the progress and development of today's society, people's demand for copper is increasing year by year. At present, in terms of the output and usage of all non-ferrous metals, aluminum ranks first, and copper ranks second only to aluminum. Copper is most widely used in electrical appliances, power transmission and electronics industries. According to statistics, nearly half of the copper produced in the world is consumed in the electrical industry.

[0003] The market demand for copper is huge, but copper smelting is very complicated, and the production process is also very complicated. From copper ore to the final usable copper product, it must first be roughly refined before it can be further processed and made into products needed in industrial production. The specific production process is to first pre-make the crude copper (containing 99% copper) into a thick plate as the anode, and make a thin sheet of pure copper as the cathode, which is placed in the electrolyte. After the power is turned on, the copper will dissolve from the anode into copper ions (Cu 2+ )Then it moves to the cathode. After the copper ions reach the cathode, they gain electrons and precipitate pure copper at the cathode. The copper produced in this way is called "electrolytic copper".

[0004] In the process of electrolytic copper, when the cathode stripping unit strips cathode copper, the surface of the stainless steel cathode plate is repeatedly squeezed and bent by mechanical force, and the operator may have incorrect operating habits, which causes the cathode plate surface to have spring plates (repeated concave and convex plates), concave and convex stretching on the edge of the plate, wrinkle deformation (plastic deformation), etc., thus affecting production. Therefore, cathode plate verticality detection is required.

[0005] In the actual detection process, the traditional method relies on the inspection personnel to observe with the naked eye with the help of simple tools, which is inefficient and has large deviations in the detection results. The introduction of image analysis algorithms has improved this problem. However, after the edge of the object is extracted, due to the complex detection environment, it is difficult to ensure that the camera or the photographed object remains completely still. Therefore, when extracting the edge, the edge extraction result will have slight fluctuations between different frames. Although this fluctuation is not obvious in the global image field of view, it will cause relatively large errors when using the least squares method for straight line fitting. In addition, most of the existing verticality detection methods use related special equipment for verticality detection, which is expensive; some solutions are cumbersome to operate, have high requirements for detection equipment and detection environment, and are not robust. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a cathode plate verticality detection method and terminal to achieve low-cost cathode plate verticality detection.

[0007] In order to solve the above technical problems, a technical solution adopted by the present invention is:

[0008] A cathode plate verticality detection method comprises the steps of:

[0009] When the cathode plate enters the detection area, a cathode plate image corresponding to the cathode plate is acquired and edge extraction is performed to obtain edge pixel points;

[0010] Obtaining the slopes of the adjacent edge pixels compared to the detection area, and calculating the average of all the slopes to obtain an average slope;

[0011] The verticality corresponding to the cathode plate is obtained according to the average slope.

[0012] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0013] A terminal comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0014] When the cathode plate enters the detection area, a cathode plate image corresponding to the cathode plate is acquired and edge extraction is performed to obtain edge pixel points;

[0015] Obtaining the slopes of the adjacent edge pixels compared to the detection area, and calculating the average of all the slopes to obtain an average slope;

[0016] The verticality corresponding to the cathode plate is obtained according to the average slope.

[0017] The beneficial effects of the present invention are as follows: a detection area is set, edge extraction is performed on the cathode plate image within the detection area to obtain edge pixel points corresponding to the cathode plate image, and the slopes of adjacent edge pixel points are obtained to calculate the average value of all slopes to obtain the verticality corresponding to the cathode plate, thereby avoiding the influence of factors such as noise brought about during the image acquisition and processing process on the final verticality test, and the slope of the edge is obtained by using a pre-set detection area as a reference, without the need to set up special detection equipment, thereby reducing the detection cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flowchart of a cathode plate verticality detection method according to an embodiment of the present invention;

[0019] Figure 2 A schematic diagram of a detection area in a cathode plate verticality detection method according to an embodiment of the present invention;

[0020] Figure 3A schematic diagram of a counting area in a cathode plate verticality detection method according to an embodiment of the present invention;

[0021] Figure 4 A schematic diagram of a verticality detection process of a cathode plate verticality detection method according to an embodiment of the present invention;

[0022] Figure 5 A schematic diagram of the division of verticality calculation areas in a cathode plate verticality detection method according to an embodiment of the present invention;

[0023] Figure 6 This is a schematic structural diagram of a cathode plate verticality detection terminal according to an embodiment of the present invention;

[0024] Description of labels:

[0025] 1. A cathode plate verticality detection terminal; 2. A processor; 3. A memory. DETAILED DESCRIPTION

[0026] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0027] Please refer to Figure 1 , a cathode plate verticality detection method, comprising the steps of:

[0028] When the cathode plate enters the detection area, a cathode plate image corresponding to the cathode plate is acquired and edge extraction is performed to obtain edge pixel points;

[0029] Obtaining the slopes of the adjacent edge pixels compared to the detection area, and calculating the average of all the slopes to obtain an average slope;

[0030] The verticality corresponding to the cathode plate is obtained according to the average slope.

[0031] From the above description, it can be seen that the beneficial effects of the present invention are: setting a detection area, performing edge extraction on the cathode plate image within the detection area to obtain edge pixel points corresponding to the cathode plate image, and obtaining the slopes of adjacent edge pixel points to calculate the average of all slopes to obtain the verticality corresponding to the cathode plate, thereby avoiding the influence of factors such as noise brought about during the image acquisition and processing process on the final verticality test, and by using a pre-set detection area as a reference to obtain the slope of the edge, there is no need to set up special detection equipment, thereby reducing the detection cost.

[0032] Furthermore, when the cathode plate enters the detection area, obtaining the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points includes:

[0033] Setting a detection area, wherein the detection area includes a horizontal reference line and a vertical reference line;

[0034] The camera is debugged until the shooting picture of the camera includes the horizontal reference line and the vertical reference line.

[0035] From the above description, it can be seen that setting horizontal reference lines and vertical reference lines in the detection area can serve as reference lines for cathode plate image correction after the cathode plate image is acquired, and can serve as reference lines for calculating the coordinates of the slope, playing a calibration role in the entire image processing process, and has low requirements for equipment and the surrounding environment, thereby reducing the detection cost.

[0036] Furthermore, when the cathode plate enters the detection area, acquiring the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points includes:

[0037] Determining whether a contour other than the horizontal reference line and the vertical reference line appears in the captured image of the camera;

[0038] If so, it is determined that the cathode plate enters the detection area, a cathode plate image corresponding to the cathode plate is acquired, and edge extraction is performed to obtain edge pixel points.

[0039] From the above description, it can be seen that the camera continuously acquires the shooting picture, and when it is determined that a contour outside the reference line appears, the cathode plate picture is captured and stored to obtain a cathode plate image, and the edge of the cathode plate image is extracted, which can ensure that the reference line in the detection area is captured when the cathode plate image is acquired, thereby ensuring that the subsequent cathode plate image can be processed correctly.

[0040] Furthermore, the obtaining of the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points comprises:

[0041] Acquire a plurality of preset edge extraction algorithms, and determine a target edge extraction algorithm according to the accuracy of the plurality of edge extraction algorithms;

[0042] The edge of the cathode plate image is extracted according to the target edge extraction algorithm to obtain edge pixel points.

[0043] From the above description, it can be seen that before determining the edge extraction algorithm to be finally used, the effects of multiple preset edge extraction algorithms are tested first. By testing the effect of each preset edge extraction algorithm before each use, the most suitable edge extraction algorithm can be identified according to changes in the environment, thereby ensuring the accuracy of the edge extraction algorithm and thus ensuring the accuracy of subsequent verticality calculations; and, when the environment changes, the most suitable target edge extraction algorithm is reselected, without the need to optimize the algorithm trained to adapt to different environments, thereby reducing the cost of algorithm training.

[0044] Further, the step of extracting the edge of the cathode plate image according to the target edge extraction algorithm to obtain edge pixel points comprises:

[0045] The edge pixels are stored in an array in a preset position order.

[0046] It can be seen from the above description that by using an array to store edge pixels, the order of edge pixels can be marked by utilizing the characteristic of array sequence marking, and the calling of edge pixels is convenient.

[0047] Further, obtaining the slopes of the adjacent edge pixels compared to the detection area and calculating the average of all the slopes to obtain an average slope includes:

[0048] Calculate the slope of the line connecting every two adjacent edge pixels to obtain a slope set;

[0049] De-noising the slope set to obtain a de-noised slope set;

[0050] The average value of all slopes in the denoised slope set is calculated to obtain an average slope.

[0051] From the above description, it can be seen that after obtaining the slope of the line between every two adjacent edge pixels, denoising is performed to delete the discontinuous parts of the edge, that is, the noise points that appear during the image processing, thereby avoiding the influence of the noise points on the average value and ensuring that the finally calculated average slope is close to the true value.

[0052] Further, obtaining the slopes of the adjacent edge pixels compared to the detection area and calculating the average of all the slopes to obtain an average slope includes:

[0053] Obtaining a preset number of samples, and taking the edge pixel points of the preset number of samples from a first preset direction and a second preset direction respectively, to obtain sample pixel points; the first preset direction is opposite to the second preset direction;

[0054] The slopes of the adjacent sample pixels compared to the detection area are obtained, and the average slope of the slopes of all the sample pixels is calculated to obtain an average slope.

[0055] From the above description, it can be seen that since edge contours are usually line segments, especially in cathode plate images, the edges of cathode plates are usually straight lines. For example, the endpoints AB of the line segment, from A to B and from B to A are two opposite directions. Therefore, the two directions of the edge are used as preset directions, and a preset number of edge pixel points are taken respectively. While achieving the effect of reducing the amount of calculation and improving the calculation efficiency, it can maximize the guarantee of the sample's summary of the entire edge situation, while ensuring the accuracy of the calculation results.

[0056] Furthermore, the obtaining of the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points comprises:

[0057] An interesting edge extraction region is set, and edge extraction is performed on the cathode plate image through the interesting edge extraction region to obtain edge pixel points.

[0058] From the above description, it can be seen that by setting an edge extraction area of ​​interest, edge pixels are obtained by edge extraction of the cathode plate image within the edge extraction area of ​​interest, which reduces the calculation area during edge extraction, reduces the overall calculation amount, and improves the calculation efficiency during edge extraction.

[0059] Furthermore, the step of obtaining the slopes of the adjacent edge pixels compared to the detection area and calculating the average of all the slopes to obtain the average slope includes:

[0060] Dividing more than one verticality calculation area, each of the verticality calculation areas includes more than one edge pixel point;

[0061] Determine the edge direction of edge pixels in each verticality calculation area, and record the verticality calculation area corresponding to each edge direction;

[0062] The verticality calculation areas corresponding to the edge direction with the least number of verticality calculation areas are eliminated.

[0063] From the above description, it can be seen that the verticality calculation area is divided, and the verticality calculation area with an edge direction different from that of other areas is eliminated to achieve preliminary noise reduction. Due to the characteristics of the cathode plate edge itself, the normal edge must be continuous. If the edge direction is opposite to that of other areas, it means that a deviation occurs in the image processing process, resulting in noise in the final image. Elimination can improve the accuracy of the final calculation result.

[0064] The cathode plate verticality detection method and terminal of the present invention can be applied to the detection of the verticality of the cathode plate in the electrolytic copper production process, which is described below through a specific implementation method.

[0065] Please refer to Figure 1 , Embodiment 1 of the present invention is:

[0066] A cathode plate verticality detection method specifically comprises:

[0067] S0. Setting a detection area, wherein the detection area includes a horizontal reference line and a vertical reference line; debugging the camera until a captured image of the camera includes the horizontal reference line and the vertical reference line;

[0068] In an optional implementation, the detection area can be set on a preset background plate, that is, a horizontal reference line and a vertical reference line are drawn on the preset background plate, and the shooting direction of the camera is set perpendicular to the surface where the detection area is located, so as to reduce the distortion of the cathode plate image;

[0069] In an optional embodiment, the horizontal reference line and the vertical reference line each include two, the area enclosed by the horizontal reference line and the vertical reference line is used as the detection area, and the intersection of the horizontal reference line and the vertical reference line is marked to obtain a marking point, for example, the marking point is positioned using an Aruco identification code; this facilitates extraction in the subsequent image processing process, cuts off irrelevant image edges, reduces the amount of data processing, and improves data processing efficiency;

[0070] In an optional embodiment, a counting area is set in the detection area, and the position of the counting area matches the conveying direction of the cathode plate test conveyor belt, that is, the counting area is set on the side of the detection area that the cathode plate first enters; please refer to Figure 3 , when the cathode plate is Figure 3 The transmission direction in is from right to left, and the dotted area on the right is the counting area; the width of the counting area along the forward direction of the cathode plate is greater than the thickness of the cathode plate facing the camera;

[0071] S1. When the cathode plate enters the detection area, a cathode plate image corresponding to the cathode plate is obtained and edge extraction is performed to obtain edge pixel points, including:

[0072] S11, determining whether a contour other than the horizontal reference line and the vertical reference line appears in the image captured by the camera; if so, executing S12;

[0073] S12, determining that the cathode plate enters the detection area, obtaining a cathode plate image corresponding to the cathode plate, and performing edge extraction to obtain edge pixel points;

[0074] In an optional embodiment, the method further includes: when each edge passes through the counting area, edge counting is performed, and if there are an odd number of edges, a cathode plate counter (an electronic counter may be used) is incremented by 1, and the initial value of the cathode plate counter is 0;

[0075] For example, there are 3 cathode plates with 6 edges on the sides. When the 1 / 3 / 5 edge enters the counting area, the electronic counter will increase by 1 and get the number 3. This can count the cathode plates that have been tested.

[0076] Wherein, obtaining the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points includes: setting an edge extraction region of interest (ROI, Region of Interest), performing edge extraction on the cathode plate image through the edge extraction region of interest to obtain edge pixel points;

[0077] S121, obtaining a plurality of preset edge extraction algorithms, and determining a target edge extraction algorithm according to the accuracy of the plurality of edge extraction algorithms;

[0078] In an optional implementation, S221 may be executed only once, that is, if the environment has not changed, when calculating the verticality of the second cathode plate, there is no need to re-select the target edge extraction algorithm, and the selected target edge extraction algorithm can be directly obtained; until the environment changes, such as position changes, light changes, cathode plate type changes, etc., the target edge extraction algorithm is selected again;

[0079] In an optional implementation, the edge extraction algorithm includes two types: pixel-level edge extraction and sub-pixel-level edge extraction. Pixel-level edge extraction mainly includes the following methods:

[0080] (1) Sobel edge detection operator: The Sobel edge detection operator has a good smoothing effect on the noise in the digital image and can remove the influence of the noise. As a type of discrete differential operator, the Sobel edge detection operator belongs to the first-order derivative edge detection operator in principle. It is mainly used to approximate the gradient of the brightness function of the target image. It calculates the influence of the position of the image pixel and performs weighted processing on it.

[0081] (2) Roberts edge detection operator; The Roberts edge detection operator is more accurate in determining the position of image edges. It is more accurate in detecting vertical edges than diagonal edges, but the Roberts edge detection operator is very sensitive to various types of noise in the image. The Roberts edge detection operator locates the edge of an image by performing differential processing on the local part of the image. When performing differential processing on the local part of the image, the two pixels on the adjacent diagonal lines are subtracted, and then the gradient amplitude of the image is approximately estimated based on the difference, thereby achieving edge detection; therefore, the Roberts edge detection operator is usually used in situations where the edges of the image are relatively clear and there is not much noise in the image; its disadvantage is that the edges of the image may appear intermittent after processing, which is not smooth enough and cannot effectively suppress the impact of noise removal;

[0082] (3) Prewitt edge detection operator: The Prewitt edge detection operator also has the function of smoothing the image. It subtracts the grayscale of a certain pixel point of the desired image and its four neighboring points. In order to remove the erroneous edges of the image, edge detection is performed when the grayscale reaches the extreme value at the desired edge. The principle of the Prewitt edge detection operator is to use two templates in the horizontal and vertical directions of the desired edge to perform convolution operations with the desired image in the neighborhood, thereby achieving the purpose of edge detection.

[0083] (4)0Canny edge detection algorithm; Canny edge detection algorithm is actually a solution to the problem of finding the maximum value; the high threshold is used to judge the edge point, that is, if the gradient value of a pixel is greater than the high threshold, the point is considered to be an edge point, if it is less than the low threshold, the point is considered not to be an edge point and will be discarded. When the gradient value of a pixel point is between the high threshold and the low threshold, the pixel point set in the neighborhood of the point is judged. If one or more points in the set are edge points, the point will also be judged as an edge point and retained, otherwise it will be deleted;

[0084] Sub-pixel edge extraction methods mainly include the following methods:

[0085] (1) Interpolation method: The interpolation method is to interpolate the grayscale value or its derivative of the pixel point in the digital image to increase the image information. However, this method is only applicable to images with strong contrast between the target and the background. Moreover, if the noise in the image is too large, it will have a great impact on the detection result and cannot accurately locate. Therefore, this method is not suitable for industrial environments with complex detection environments.

[0086] (2) Fitting method: Least squares regression straight line fitting method, Chebyshev polynomial fitting method, these methods are to obtain the sub-pixel positioning of edge contour by fitting the gray value of edge model; Least squares linear regression method realizes fitting through straight line edge vector, this method is mainly used for straight line edge fitting; its detection accuracy depends on the number of fitting points, the more fitting points, the higher the accuracy, but the longer the time required to fit the data. Chebyshev polynomial fitting method has high repeatability and positioning accuracy, but because this method uses low-order polynomial edge function to obtain the continuous distribution function near the edge, and then uses the continuous distribution function near the edge for sub-pixel positioning, this method needs to process the edge of the image one by one, so that this method is time-consuming and inefficient.

[0087] (3) Moment method; The sub-pixel step edge detection model based on orthogonal moments was proposed by Ghosal et al. The detection algorithm based on moments is simple to calculate and has a strong ability to suppress noise in the image. The detection methods based on grayscale moments, spatial moments, and Zernike moments are all sub-pixel detection methods based on orthogonal moments. Among them, the detection methods based on grayscale moments and spatial moments both solve the edge position of the grayscale space of the image edge, and the running speed of this method is relatively long. The detection method based on Zernike orthogonal moments only needs to calculate 3 templates, and the amount of calculation is much smaller than the detection method based on spatial moments, so the processing time is short;

[0088] S122, extracting the edge of the cathode plate image according to the target edge extraction algorithm to obtain edge pixel points;

[0089] In an optional implementation, the edge pixel points are stored in an array according to a preset position order; for example, if a point is PointsArr[n], the second point is PointsArr[len(PointsArr)-1-n], and when n=0, the first point is PointsArr[0], that is, the first point in the array;

[0090] S3, obtaining the slope of the adjacent edge pixel points compared to the detection area, and calculating the average of all the slopes to obtain an average slope, including:

[0091] S30, performing perspective change on the cathode plate image according to the horizontal reference line and the vertical reference line to correct image distortion; the specific implementation method is relatively mature in the existing image processing and will not be described in detail here;

[0092] In an optional implementation, perspective changes may be performed based on the marker points to correct image distortion;

[0093] In an optional implementation, step S30 may be executed after the cathode plate image is acquired, that is, in step S12, the cathode plate image corresponding to the cathode plate is acquired, and after executing S30, edge extraction is performed to obtain edge pixel points;

[0094] S31, calculating the slope of the line connecting every two adjacent edge pixels to obtain a slope set;

[0095] In an optional implementation, a preset number of samples is obtained, and the edge pixel points of the preset number of samples are taken from a first preset direction and a second preset direction respectively to obtain sample pixel points; the first preset direction is opposite to the second preset direction, and the slopes of the adjacent sample pixel points compared to the detection area are obtained to obtain a slope set;

[0096] S32, denoising the slope set to obtain a denoised slope set;

[0097] In an optional implementation, S22 is optional, that is, in order to strike a balance between processing efficiency and processing accuracy, it can be determined whether to perform the denoising step according to actual needs;

[0098] S33, calculating the average of all slopes in the denoised slope set to obtain an average slope, including:

[0099] In an optional implementation, the average slope is obtained by calculating the average of the slopes of all sample pixel points;

[0100] S4, obtaining the verticality corresponding to the cathode plate according to the average slope;

[0101] In an optional implementation, the verticality f=arctan(x), wherein x is the average slope.

[0102] Embodiment 2 of the present invention is:

[0103] A cathode plate verticality detection method, which is different from the first embodiment in that:

[0104] In an optional implementation, step S2 is included between steps S1 and S3:

[0105] S21, dividing into more than one verticality calculation area, each of the verticality calculation areas including more than one edge pixel point;

[0106] S22, determining the edge direction of edge pixels in each verticality calculation area, and recording the verticality calculation area corresponding to each edge direction;

[0107] S23, eliminating the verticality calculation areas corresponding to the edge directions with the least number of verticality calculation areas, to obtain a target verticality calculation area set;

[0108] After eliminating the verticality calculation areas with different edge directions from other verticality calculation areas, the average value of the slope is calculated to avoid the influence of values ​​with excessive deviations on the calculation results. That is, higher-dimensional denoising is performed first, and then finer-grained denoising can be performed (step S32), which can further improve data processing efficiency.

[0109] In an optional implementation, S3 includes:

[0110] S31, calculating the slope of the line connecting every two adjacent edge pixels to obtain a slope set;

[0111] S32, calculating the partition average slope of the slope corresponding to the line connecting the edge pixels in each verticality calculation region (ROI); and the total average slope of the slope corresponding to the line connecting the edge pixels after all verticality calculation regions are combined;

[0112] In summary, please refer to Figure 5 , where the box is the verticality calculation area, which can be generated randomly or by receiving the input quantity and position. The slope of the straight line in each verticality calculation area is the average slope of the partition;

[0113] That is, in this embodiment, steps S31-S32 can replace steps S31-S33 in the first embodiment; for example, a verticality calculation method can be selected according to the received user selection information, and the scheme of comprehensive accuracy and efficiency can be convenient for users to choose according to their own different needs;

[0114] Then S4 includes:

[0115] S41, calculating a first verticality corresponding to each verticality calculation area according to the average slope of the partitions; and calculating a second verticality according to the total average slope;

[0116] S42, calculating the average of each first verticality and the second verticality to obtain the verticality corresponding to the cathode plate.

[0117] Please refer to Figure 2 , Embodiment 3 of the present invention is:

[0118] A cathode plate verticality detection terminal 1 includes a processor 2, a memory 3, and a computer program stored in the memory 3 and executable on the processor 2. When the processor 2 executes the computer program, each step in the first or second embodiment is implemented.

[0119] In summary, the present invention provides a cathode plate verticality detection method and terminal, which obtains the slope of each segment by calculating the slope of the edge segment, and finally calculates the average value corresponding to all the slopes, and obtains the verticality corresponding to the cathode plate according to the average slope, thereby avoiding the problem of excessive deviation caused by sampling and fitting a straight line, and can incorporate the characteristics of each segment into the final verticality calculation; at the same time, the concept of samples is introduced, and all edge pixel points in an edge are not calculated, but edge pixel points of a preset number of samples are selected from two directions for straight line fitting, which reduces the amount of calculation while ensuring accuracy; and, ROI is introduced to denoise the edge pixel points from small to large, further improving the accuracy of the final detection result; and, the average verticality of all extracted frames is calculated, so that the verticality detection system is more robust.

[0120] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A cathode plate verticality detection method, characterized in that: Includes steps: When the cathode plate enters the detection area, a cathode plate image corresponding to the cathode plate is acquired and edge extraction is performed to obtain edge pixel points; Obtaining the slopes of the adjacent edge pixels compared to the detection area, and calculating the average of all the slopes to obtain an average slope; The verticality corresponding to the cathode plate is obtained according to the average slope.

2. A cathode plate verticality detection method according to claim 1, characterized in that: When the cathode plate enters the detection area, obtaining the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points includes: Setting a detection area, wherein the detection area includes a horizontal reference line and a vertical reference line; The camera is debugged until the shooting picture of the camera includes the horizontal reference line and the vertical reference line.

3. A cathode plate verticality detection method according to claim 2, characterized in that: When the cathode plate enters the detection area, acquiring the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points comprises: Determining whether a contour other than the horizontal reference line and the vertical reference line appears in the captured image of the camera; If so, it is determined that the cathode plate enters the detection area, a cathode plate image corresponding to the cathode plate is acquired, and edge extraction is performed to obtain edge pixel points.

4. A cathode plate verticality detection method according to any one of claims 1 to 3, characterized in that: The acquiring of the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points comprises: Acquire a plurality of preset edge extraction algorithms, and determine a target edge extraction algorithm according to the accuracy of the plurality of edge extraction algorithms; The edge of the cathode plate image is extracted according to the target edge extraction algorithm to obtain edge pixel points.

5. A cathode plate verticality detection method according to claim 4, characterized in that: The step of extracting the edge of the cathode plate image according to the target edge extraction algorithm to obtain edge pixel points comprises: The edge pixels are stored in an array according to a preset position order.

6. A cathode plate verticality detection method according to claim 1, characterized in that: The obtaining of the slopes of the adjacent edge pixels compared to the detection area and calculating the average of all the slopes to obtain the average slope includes: Calculate the slope of the line connecting every two adjacent edge pixels to obtain a slope set; De-noising the slope set to obtain a de-noised slope set; The average value of all slopes in the denoised slope set is calculated to obtain an average slope.

7. A cathode plate verticality detection method according to claim 1 or 6, characterized in that: The obtaining of the slopes of the adjacent edge pixels compared to the detection area and calculating the average of all the slopes to obtain the average slope includes: Obtaining a preset number of samples, and taking the edge pixel points of the preset number of samples from a first preset direction and a second preset direction respectively, to obtain sample pixel points; the first preset direction is opposite to the second preset direction; The slopes of the adjacent sample pixels compared to the detection area are obtained, and the average slope of the slopes of all the sample pixels is calculated to obtain an average slope.

8. A cathode plate verticality detection method according to claim 1, characterized in that: The acquiring of the cathode plate image corresponding to the cathode plate and performing edge extraction to obtain edge pixel points comprises: An interesting edge extraction region is set, and edge extraction is performed on the cathode plate image through the interesting edge extraction region to obtain edge pixel points.

9. A cathode plate verticality detection method according to claim 1, characterized in that: The step of obtaining the slopes of the adjacent edge pixels compared to the detection area and calculating the average of all the slopes to obtain the average slope includes: Dividing more than one verticality calculation area, each of the verticality calculation areas includes more than one edge pixel point; Determine the edge direction of edge pixels in each verticality calculation area, and record the verticality calculation area corresponding to each edge direction; The verticality calculation areas corresponding to the edge direction with the least number of verticality calculation areas are eliminated.

10. A cathode plate verticality detection terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of a cathode plate verticality detection method as described in any one of claims 1 to 9 when executing the computer program.

Images