Storage material pile edge contour calculation method based on visual detection

Abstract

The invention discloses a warehouse material pile edge contour calculation method based on visual detection, and relates to the technical field of warehouse management, and the method comprises the steps: obtaining a real-time image of a material pile in a target warehouse through an unmanned plane, carrying out the gray processing of a sorted standard material pile image, and obtaining a material pile gray image; obtaining edge feature points and edge contours of the material pile gray level image based on an edge detection algorithm, integrating the edge contours according to contour tortuosity of the edge contours to obtain a storage material pile contour, establishing a storage space model based on space data, and restoring a storage material pile model according to the storage material pile contour. According to the method and the device, the original storage stockpile model is imported into each corresponding storage space, contour calculation is performed according to the contour line of the original storage stockpile model and the edge line of each space to obtain the contour deviation value, and the stockpile condition is judged according to the preset contour deviation range, so that the storage stockpile accounting accuracy is improved; and the stockpiling can be accurately scheduled according to a calculation result, so that the utilization rate of storage is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of warehousing management, and particularly to a method for calculating the edge contour of a warehousing stockpile based on visual detection. Background Art

[0002] Visual detection is a technology that automatically analyzes and evaluates objects, products, or environments through computer vision technology. It is widely used in many fields such as industrial production, quality control, robot navigation, and medical imaging. The core of visual detection is to use a camera or sensor to obtain image data, and then process these images through algorithms to automatically identify and analyze the features of objects and determine whether they meet the preset standards or requirements.

[0003] During the warehousing management process, inspecting the storage status of the warehouse is an important task. Traditional methods rely on manual inspections at regular intervals, which are time-consuming and inefficient, and problems cannot be detected in a timely manner. Due to the subjectivity and limitations of manual inspections, it is often difficult to accurately judge the status of goods in many cases. Such human factors lead to false alarms and missed reports, resulting in inaccurate inventory information, which in turn affects subsequent replenishment and allocation decisions; moreover, traditional methods cannot achieve real-time monitoring of the warehouse status, and many problems can only be discovered during regular inspections. This lag makes it difficult to solve potential problems in a timely manner and affects the overall efficiency of warehousing management; furthermore, in traditional methods, the cleanliness of the shelves often depends on personal subjective judgment, lacking objective criteria, and for large stockpiles, the manual calculation amount is large and the calculation ability requirements for operators are high;

[0004] In view of the above technical deficiencies, a solution is now proposed. Summary of the Invention

[0005] The purpose of the present invention is to: perform contour calculation based on the contour line of the original warehousing stockpile model and the edge lines of each space to obtain a contour deviation value, and judge the stockpiling situation according to the preset contour deviation range, which improves the accuracy of warehousing stockpile accounting and can accurately schedule the stockpile according to the calculation results, thereby improving the utilization rate of the warehouse.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: A method for calculating the edge contour of a warehousing stockpile based on visual detection, comprising the following steps:

[0007] Step 1: Based on a drone, image acquisition of the stockpile in the target warehouse is performed according to a preset sampling period to obtain a real-time image of the stockpile in the target warehouse, perform preprocessing operations on the real-time image of the stockpile, and sort the standard stockpile images according to the time series collected by the drone to obtain standard stockpile images;

[0008] Step 2: Perform grayscale processing on the sorted standard stockpile image to obtain a stockpile grayscale image. Based on the edge detection algorithm, obtain the edge feature points and edge contour lines of the stockpile grayscale image, and integrate the edge contours according to the contour tortuosity of each edge contour line to obtain the warehouse stockpile contour;

[0009] Step 3: Obtain the spatial data in the target warehouse, establish a warehouse space model based on the spatial data, and obtain the warehouse scheduling information in the target warehouse. Based on the warehouse scheduling information, divide the stored space and the to-be-stored space in the warehouse space model, and obtain the contour distribution lines of each space;

[0010] Step 4: Restore the warehouse stockpiling model according to the warehouse stockpile contour, import the original warehouse stockpiling model into the corresponding storage spaces, and perform contour calculation based on the contour lines of the original warehouse stockpiling model and the edge lines of each space to obtain the contour deviation value;

[0011] Step 5: Judge the stockpiling situation according to the preset contour deviation range. The stockpiling situation includes qualified stockpiling, misaligned stockpiling, and low utilization rate of the stockpiling space, and correspondingly generate a stockpiling scheduling signal and a stockpiling approval signal.

[0012] Further, the specific process of obtaining the standard stockpile image is as follows:

[0013] S101: According to the distribution state of the warehouse stockpile, use a drone to start collecting the top corner image of the warehouse stockpile and then collect the overall images of the four side elevations to obtain the real-time image of the stockpile in the target warehouse;

[0014] S102: Process the collected real-time image of the stockpile by the mean filtering method, replace the gray mean value around each pixel point in the original image with the gray value of the corresponding pixel point, and obtain the standard stockpile image by traversing all pixel points and performing filtering processing;

[0015] S103: Sort the standard stockpile images according to the time series collected by the drone to obtain the standard stockpile images, including the top image of the stockpile, the first side image of the stockpile, the second side image of the stockpile, the third side image of the stockpile, and the fourth side image of the stockpile.

[0016] Further, the specific process of obtaining the warehouse stockpile contour is as follows:

[0017] S201: Based on the edge tracking algorithm, obtain the edge feature points and edge contour lines of the grayscale image, and obtain the gray value corresponding to each edge feature point and the node gray value of each edge contour line;

[0018] S202. Based on a preset grayscale judgment threshold GR1 and a preset grayscale judgment threshold GR2, if the grayscale value corresponding to each edge feature point is greater than or equal to the preset grayscale judgment threshold GR1, then mark this edge feature point as a valid feature point;

[0019] If the grayscale value of the node of each edge contour line is greater than or equal to the preset grayscale judgment threshold GR2, then mark the node of each edge contour line as a valid node;

[0020] S203. After integrating all the valid nodes, connect them with a smooth curve to obtain a corrected edge contour line. Mark the valid feature points one by one on the corrected edge contour line. Based on the gradient features of the valid feature points and the differences between the gradient features, obtain the contour tortuosity of each edge contour line;

[0021] S204. Based on a preset contour tortuosity threshold, select the corner feature points of each edge contour line, and perform straight-line detection according to adjacent corner feature points to obtain the storage stockpile contour.

[0022] Further, the spatial data includes the actual size of the target warehouse, the storage space division data of the target warehouse, and the storage capacity of each storage space. The warehousing scheduling information includes the stacking scheduling quantity, the stacking scheduling target site, and the stacking unit scheduling volume. Mark the storage space as the stored space according to the stacking scheduling target site, and obtain the actual storage quantity of the stored space according to the cumulative value of the stacking scheduling quantity. At the same time, the standard volume of the stacked material can be known according to the cumulative value of the stacking unit scheduling volume.

[0023] Further, the specific process of restoring the warehousing stacking model is as follows:

[0024] S301. Obtain the target line drawing of the warehousing stockpile contour, convert the grayscale value corresponding to the line in the target line drawing into a line color, and generate the warehousing stacking model corresponding to the target line drawing according to the target shape combined by the lines and the storage space corresponding to the warehousing space model;

[0025] Among them, the color of the warehousing stacking model matches the grayscale value of the stockpile grayscale image, the contour of the warehousing stacking model matches the warehousing stockpile contour, and the model type to which the warehousing stacking model belongs matches the scene type in the space warehousing model;

[0026] S302. Replace the target line drawing displayed in the space warehousing model with the warehousing stacking model.

[0027] Further, the specific process of obtaining the contour deviation value is as follows:

[0028] S401. Obtain the contour line of the storage stockpiling model. Select any side of the storage stockpiling model as the image to be evaluated. Obtain the edge line of the stored space where the storage stockpiling model is located. Select the coincident side with the storage stockpiling model as the reference image. Merge the intersecting contours in the image to be evaluated and merge the intersecting contours in the reference image.

[0029] S402. Calculate the intersection-over-union ratio between each contour in the merged image to be evaluated and each contour in the merged reference image. According to the preset intersection-over-union ratio judgment threshold, mark the contour pairs with an intersection-over-union ratio greater than or equal to the intersection-over-union ratio judgment threshold as deviation graphics.

[0030] S403. Mark a number of sampling points Mk on the image to be evaluated in the deviation graphics, and obtain the position coordinates Mk(xk, yk) of the sampling points from the spatial storage model. Mark reference points Mg(xg, yg) in the reference image according to the interval of the sampling points. Calculate the contour deviation value S between the sampling points and the reference points according to the following formula: where i = 1, 2, 3,..., n, and n is the number of sampling points;

[0031] The contour deviation value S is used to reflect the edge placement error corresponding to each sampling point. The larger the contour deviation value S, the greater the deviation between the contour line of the original storage stockpiling model and the edge lines of each space. On the contrary, the smaller the contour deviation value S, the smaller the deviation between the contour line of the original storage stockpiling model and the edge lines of each space.

[0032] Further, the specific process of judging the stockpiling situation is as follows:

[0033] Obtain the preset contour deviation range (Smin, Smax). If the contour deviation value S is less than or equal to Smin, the stockpiling situation corresponds to qualified stockpiling;

[0034] If the contour deviation value S is less than Smax and greater than Smin, the stockpiling situation corresponds to misaligned stockpiling, and a stockpiling approval signal is generated accordingly;

[0035] If the contour deviation value S is greater than or equal to Smax, the stockpiling situation corresponds to low utilization rate of the stockpiling space, and a stockpiling scheduling signal is generated accordingly.

[0036] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0037] The method for calculating the edge contour of a storage pile based on visual detection obtains real-time images of the pile in the target warehouse through a drone, performs gray-scale processing on the sorted standard pile images to obtain gray-scale pile images, obtains the edge feature points and edge contour lines of the gray-scale pile images based on an edge detection algorithm, integrates the edge contours according to the contour tortuosity of each edge contour line to obtain the storage pile contour, establishes a storage space model based on spatial data, restores the storage pile model according to the storage pile contour, and imports the original storage pile model into the corresponding storage spaces. The contour deviation value is obtained by performing contour calculation based on the contour lines of the original storage pile model and the edge lines of each space, and the stacking situation is judged according to a preset contour deviation range, which improves the accuracy of storage pile accounting and can accurately schedule the stacking according to the calculation results, improving the utilization rate of the storage. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The overall method flow diagram of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0040] Embodiment 1:

[0041] As Figure 1 shown, the method for calculating the edge contour of a storage pile based on visual detection includes the following steps:

[0042] Step 1: Based on the drone, image acquisition of the pile in the target warehouse is performed according to a preset sampling period to obtain real-time images of the pile in the target warehouse, preprocessing operations are performed on the real-time images of the pile, and the standard pile images are sorted according to the time series collected by the drone to obtain standard pile images;

[0043] The specific process of obtaining the standard pile images is as follows:

[0044] S101: According to the distribution state of the storage pile, the drone is used to start collecting the top image of the piled material from the top corner of the storage pile, and then the overall images of the four side facades are collected to obtain real-time images of the pile in the target warehouse;

[0045] S102. Process the real-time image of the stockpile collected by means of mean filtering, replace the gray-scale mean value around each pixel point in the original image with the gray scale of the corresponding pixel point, and obtain the standard stockpile image by traversing all pixel points and performing filtering processing;

[0046] S103. Sort the standard stockpile images according to the time series collected by the unmanned aerial vehicle to obtain standard stockpile images, including the top image of the stockpile, the first side image of the stockpile, the second side image of the stockpile, the third side image of the stockpile, and the fourth side image of the stockpile.

[0047] Step 2. Perform gray-scale processing on the sorted standard stockpile images to obtain the stockpile gray-scale image, obtain each edge feature point and the edge contour line of the stockpile gray-scale image based on the edge detection algorithm, and integrate the edge contours according to the contour tortuosity of each edge contour line to obtain the storage stockpile contour;

[0048] The specific process of obtaining the storage stockpile contour is as follows:

[0049] S201. Obtain each edge feature point and each edge contour line of the gray-scale image based on the edge tracking algorithm, and obtain the gray-scale value corresponding to each edge feature point and the node gray-scale value of each edge contour line;

[0050] S202. Based on the preset gray-scale judgment threshold GR1 and the preset gray-scale judgment threshold GR2, if the gray-scale value corresponding to each edge feature point is greater than or equal to the preset gray-scale judgment threshold GR1, then mark this edge feature point as a valid feature point;

[0051] If the node gray-scale value of each edge contour line is greater than or equal to the preset gray-scale judgment threshold GR2, then mark the node of each edge contour line as a valid node;

[0052] S203. Integrate all the valid nodes and connect them with a smooth curve to obtain the corrected edge contour line, mark each valid feature point on the corrected edge contour line one by one, and obtain the contour tortuosity of each edge contour line based on the gradient feature of the valid feature points and the difference between the gradient features;

[0053] S204. Select the corner feature points of each edge contour line based on the preset contour tortuosity threshold, perform straight line detection according to the adjacent corner feature points, and obtain the storage stockpile contour;

[0054] Step 3. Obtain the spatial data in the target warehouse, establish a storage space model based on the spatial data, and obtain the storage scheduling information in the target warehouse. Divide the stored space and the to-be-stored space in the storage space model based on the storage scheduling information, and obtain the contour distribution lines of each space;

[0055] Spatial data includes the actual dimensions of the target warehouse, the storage space division data of the target warehouse, and the storage capacity of each storage space. Warehouse scheduling information includes the stacking scheduling quantity, the stacking scheduling target site, and the stacking unit scheduling volume. The storage space is marked as the stored space according to the stacking scheduling target site, and the actual storage quantity of the stored space is obtained according to the cumulative value of the stacking scheduling quantity. At the same time, the standard volume of the stacked materials can be known according to the cumulative value of the stacking unit scheduling volume.

[0056] Step 4: Restore the warehouse stacking model according to the warehouse stack contour, import the original warehouse stacking model into the corresponding storage spaces, and perform contour calculation based on the contour line of the original warehouse stacking model and the edge lines of each space to obtain the contour deviation value.

[0057] The specific process of restoring the warehouse stacking model is as follows:

[0058] S301: Obtain the target line drawing of the warehouse stack contour, convert the gray value corresponding to the line in the target line drawing into the line color, and generate the warehouse stacking model corresponding to the target line drawing according to the target shape combined by the lines and the storage space corresponding to the warehouse space model.

[0059] Among them, the color of the warehouse stacking model matches the gray value of the stack gray image, the contour of the warehouse stacking model matches the warehouse stack contour, and the model type to which the warehouse stacking model belongs matches the scene type in the space warehouse model.

[0060] S302: Replace the target line drawing displayed in the space warehouse model with the warehouse stacking model.

[0061] The specific process of obtaining the contour deviation value is as follows:

[0062] S401: Obtain the contour line of the warehouse stacking model, select any side of the warehouse stacking model as the image to be evaluated, obtain the edge line of the stored space where the warehouse stacking model is located, select the coincident side with the warehouse stacking model as the reference image, and merge the intersecting contours in the image to be evaluated and the intersecting contours in the reference image.

[0063] S402: Calculate the intersection-over-union ratio between each contour in the merged image to be evaluated and each contour in the merged reference image, and mark the contour pairs with the intersection-over-union ratio greater than or equal to the preset intersection-over-union ratio judgment threshold as deviation graphics according to the preset intersection-over-union ratio judgment threshold.

[0064] S403: Mark a number of sampling points Mk on the image to be evaluated in the deviation graphics, obtain the position coordinates Mk(xk, yk) of the sampling points from the space warehouse model, and also mark the reference points Mg(xg, yg) in the reference image according to the interval of the sampling points. Calculate the contour deviation value S between the sampling points and the reference points according to the following formula: where \(i = 1, 2, 3, \cdots, n\), and \(n\) is the number of sampling points;

[0065] The contour deviation value \(S\) is used to reflect the edge placement error corresponding to each sampling point. The larger the contour deviation value \(S\), the greater the deviation between the contour line of the original storage stockpiling model and the edge lines of each space. Conversely, the smaller the contour deviation value \(S\), the smaller the deviation between the contour line of the original storage stockpiling model and the edge lines of each space.

[0066] Step Five: Determine the stockpiling situation according to the preset contour deviation range. The stockpiling situation includes qualified stockpiling, misaligned stockpiling, and low utilization rate of the stockpiling space, and correspondingly generate a stockpiling scheduling signal and a stockpiling approval signal.

[0067] The specific process of determining the stockpiling situation is as follows:

[0068] Obtain the preset contour deviation range \((S_{min}, S_{max})\). If the contour deviation value \(S\) is less than or equal to \(S_{min}\), the stockpiling situation corresponds to qualified stockpiling;

[0069] If the contour deviation value \(S\) is less than \(S_{max}\) and greater than \(S_{min}\), the stockpiling situation corresponds to misaligned stockpiling, and correspondingly generate a stockpiling approval signal;

[0070] If the contour deviation value \(S\) is greater than or equal to \(S_{max}\), the stockpiling situation corresponds to low utilization rate of the stockpiling space, and correspondingly generate a stockpiling scheduling signal.

[0071] In the present invention, a real-time image of the stockpile in the target warehouse is obtained by a drone, the sorted standard stockpile image is subjected to grayscale processing to obtain a grayscale stockpile image, various edge feature points and edge contour lines of the grayscale stockpile image are obtained based on an edge detection algorithm, the edge contours are integrated according to the contour tortuosity of each edge contour line to obtain a storage stockpile contour, a storage space model is established based on spatial data, the storage stockpiling model is restored according to the storage stockpile contour, and the original storage stockpiling model is imported into the corresponding storage spaces. The contour calculation is performed according to the contour line of the original storage stockpiling model and the edge lines of each space to obtain the contour deviation value. The stockpiling situation is judged according to the preset contour deviation range, which improves the accuracy of storage stockpile accounting and can accurately schedule the stockpiling according to the calculation results, thereby improving the utilization rate of the storage.

[0072] The setting of the size of the interval and threshold is for the convenience of comparison. Regarding the size of the threshold, it depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0073] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0074] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, making equivalent substitutions or changes, should be covered within the protection scope of the present invention.

Claims

1. A method for calculating the edge contour of a storage stockpile based on visual detection, characterized in that, Including the following steps: Step 1: Based on the drone, acquire images of the stockpiles in the target warehouse at a preset sampling period to obtain real-time images of the stockpiles in the target warehouse. Perform preprocessing operations on the real-time images of the stockpiles, and sort the standard stockpile images according to the time series collected by the drone to obtain standard stockpile images; Step 2: Perform gray-scale processing on the sorted standard stockpile images to obtain stockpile gray-scale images. Based on the edge detection algorithm, obtain the edge feature points and edge contour lines of the stockpile gray-scale images, and integrate the edge contours according to the contour tortuosity of each edge contour line to obtain the storage stockpile contour; Step 3: Obtain the spatial data in the target warehouse, establish a storage space model based on the spatial data, and obtain the storage scheduling information in the target warehouse. Based on the storage scheduling information, divide the stored space and the to-be-stored space in the storage space model, and obtain the contour distribution lines of each space; Step 4: Restore the storage stockpiling model according to the storage stockpile contour, and import the original storage stockpiling model into the corresponding storage spaces. Perform contour calculations based on the contour lines of the original storage stockpiling model and the edge lines of each space to obtain the contour deviation value; Step 5: Judge the stockpiling situation according to the preset contour deviation range. The stockpiling situation includes qualified stockpiling, misaligned stockpiling, and low utilization rate of the stockpiling space, and generate corresponding stockpiling scheduling signals and stockpiling approval signals.

2. The method for calculating the edge contour of a storage material pile based on visual detection according to claim 1, wherein The specific process of obtaining the standard stockpile images is as follows: S101: According to the distribution state of the storage stockpiles, select the top corner of the storage stockpiles by the drone to start collecting the top images of the stockpiles, and then collect the overall images of the four side facades to obtain the real-time images of the stockpiles in the target warehouse; S102: Process the collected real-time images of the stockpiles by the mean filtering method, replace the gray-scale mean value around each pixel point in the original image with the gray-scale of the corresponding pixel point, and obtain the standard stockpile images by traversing all pixel points and performing filtering processing; S103: Sort the standard stockpile images according to the time series collected by the drone to obtain standard stockpile images, including the top image of the stockpiles, the first side image of the stockpiles, the second side image of the stockpiles, the third side image of the stockpiles, and the fourth side image of the stockpiles.

3. The method for calculating the edge contour of a storage stockpile based on visual detection according to claim 1, wherein The specific process of obtaining the storage stockpile contour is as follows: S201: Based on the edge tracking algorithm, obtain the edge feature points and each edge contour line of the gray-scale image, and obtain the gray-scale values corresponding to each edge feature point and the node gray-scale values of each edge contour line; S202: Based on the preset gray-scale judgment threshold GR1 and the preset gray-scale judgment threshold GR2, if the gray-scale value corresponding to each edge feature point is greater than or equal to the preset gray-scale judgment threshold GR1, then mark the edge feature point as a valid feature point; If the node gray-scale value of each edge contour line is greater than or equal to the preset gray-scale judgment threshold GR2, then mark the node of each edge contour line as a valid node; S203. Integrate all valid nodes and connect them with a smooth curve to obtain the corrected edge contour line. Mark each valid feature point on the corrected edge contour line. Based on the gradient features of the valid feature points and the differences between the gradient features, obtain the contour tortuosity of each edge contour line. S204. Based on a preset contour tortuosity threshold, select the corner feature points of each edge contour line. Perform line detection based on adjacent corner feature points to obtain the contour of the storage stockpile.

4. The method for calculating the edge contour of a storage material pile based on visual detection according to claim 1, wherein The spatial data includes the actual size of the target warehouse, the storage space division data of the target warehouse, and the storage capacity of each storage space. The warehousing scheduling information includes the stockpiling scheduling quantity, the target site of the stockpiling scheduling, and the unit scheduling volume of the stockpiling. Mark the storage space as the stored space according to the target site of the stockpiling scheduling, and obtain the actual storage quantity of the stored space according to the cumulative value of the stockpiling scheduling quantity. At the same time, the standard volume of the stockpiling can be known according to the cumulative value of the unit scheduling volume of the stockpiling.

5. The method for calculating the edge contour of a storage stockpile based on visual detection according to claim 1, wherein The specific process of restoring the warehousing stockpile model is as follows: S301. Obtain the target line drawing of the warehousing stockpile contour. Convert the gray value corresponding to the line in the target line drawing into a line color, and generate the warehousing stockpile model corresponding to the target line drawing according to the target shape combined by the lines and the storage space corresponding to the warehousing space model. Among them, the color of the warehousing stockpile model matches the gray value of the stockpile gray image, the contour of the warehousing stockpile model matches the warehousing stockpile contour, and the model type to which the warehousing stockpile model belongs matches the scene type in the spatial warehousing model. S302. Replace the target line drawing displayed in the spatial warehousing model with the warehousing stockpile model.

6. The method for calculating the edge contour of a warehouse stockpile based on visual detection according to claim 1, wherein The specific process of obtaining the contour deviation value is as follows: S401. Obtain the contour line of the warehousing stockpile model. Select any side of the warehousing stockpile model as the image to be evaluated. Obtain the edge line of the stored space where the warehousing stockpile model is located. Select the side that coincides with the warehousing stockpile model as the reference image. Merge the intersecting contours in the image to be evaluated and merge the intersecting contours in the reference image. S402. Calculate the intersection-over-union ratio between each contour in the merged image to be evaluated and each contour in the merged reference image. According to a preset intersection-over-union ratio judgment threshold, mark the contour pairs with an intersection-over-union ratio greater than or equal to the intersection-over-union ratio judgment threshold as deviation figures. S403. Mark a number of sampling points Mk on the graph to be evaluated in the deviation graph, obtain the position coordinates Mk(xk, yk) of the sampling points from the spatial storage model, and also mark reference points Mg(xg, yg) in the reference image according to the interval of the sampling points. Calculate the contour deviation value S between the sampling points and the reference points according to the following formula: where i = 1, 2, 3, …, n, and n is the number of sampling points; The contour deviation value S is used to reflect the edge placement error corresponding to each sampling point.

7. The method for calculating the edge contour of a storage stockpile based on visual detection according to claim 1, wherein The specific process of judging the stockpiling situation is as follows: Obtain a preset contour deviation range (Smin, Smax). If the contour deviation value S is less than or equal to Smin, the stockpiling situation corresponds to qualified stockpiling. If the contour deviation value S is less than Smax and greater than Smin, the stockpiling situation corresponds to misaligned stockpiling, and a stockpiling approval signal is generated accordingly. If the contour deviation value S is greater than or equal to Smax, the stockpiling situation corresponds to low utilization rate of the stockpiling space, and a stockpiling scheduling signal is generated accordingly.

Images