Anchored net supporting hole center point positioning method, system and device based on depth image

Abstract

The application discloses an anchor net supporting hole center point positioning method, system and equipment based on a depth image, relates to the technical field of mine safety, and comprises the following steps: carrying out local region division on a pretreated depth image to obtain a plurality of grid regions; carrying out adaptive threshold segmentation and then merging on each grid region to obtain a binary image with the same size as the depth image; carrying out a contour searching operation on the binary image with the same size as the depth image to obtain a plurality of contours, fitting a minimum circumscribed rectangle of each contour, and obtaining a plurality of minimum circumscribed rectangles; excluding non-target minimum circumscribed rectangles in the plurality of minimum circumscribed rectangles through a preset threshold to obtain target minimum circumscribed rectangles; and finally determining the position of the anchor net supporting hole center point according to the target minimum circumscribed rectangles. The anchor net supporting hole center point positioning method provided by the application can effectively improve the robustness and working efficiency of an automatic anchor supporting robot in positioning the anchor net supporting center point under any illumination conditions in a mine.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mine safety, in particular to an anchor net supporting hole center point positioning method, system and equipment based on a depth image. BACKGROUND

[0002] After the coal mine tunnel is excavated, the rock wall needs to be supported to ensure the safety of the staff. The most commonly used supporting technology is anchor net supporting, and when the automatic anchor supporting robot performs anchor net supporting operation, the anchor rod needs to be inserted into the rock wall through the anchor net hole center point to fix the anchor net. Therefore, in the complex and changeable lighting environment in the mine, accurately and quickly positioning the center point of the anchor net hole is of great significance to the improvement of the excavation efficiency and the safety of the support.

[0003] The existing anchor net hole center point positioning method is based on a color image. However, due to the complex lighting conditions in the mine, such as insufficient lighting, uneven lighting, and light obstruction, the collected color image may be unclear, have insufficient contrast, and unable to display detailed information, which seriously affects the observation effect and information extraction of the image in the mine, and further leads to poor robustness of the anchor net hole center point positioning method based on the color image, which seriously affects the excavation efficiency and the safety of the support. SUMMARY

[0004] The purpose of the present application is to provide an anchor net supporting hole center point positioning method, system and equipment based on a depth image, which can effectively improve the robustness and work efficiency of the automatic anchor supporting robot in positioning the anchor net supporting center point under any lighting conditions in the mine.

[0005] To achieve the above purpose, the present application provides the following scheme:

[0006] An anchor net supporting hole center point positioning method based on a depth image, comprising:

[0007] obtaining a depth image of the anchor net and performing preprocessing to obtain a preprocessed depth image;

[0008] dividing the preprocessed depth image into local regions to obtain a plurality of grid regions;

[0009] performing adaptive threshold segmentation and merging on each grid region to obtain a binary image with the same size as the depth image;

[0010] performing a contour search operation on the binary image with the same size as the depth image to obtain a plurality of contours, and fitting a minimum bounding rectangle of each contour to obtain a plurality of minimum bounding rectangles;

[0011] excluding non-target minimum bounding rectangles from the plurality of minimum bounding rectangles by a preset threshold to obtain a target minimum bounding rectangle;

[0012] Determine the position of the anchor net support hole center point according to the target minimum circumscribed rectangle.

[0013] Optionally, the depth image of the anchor net is acquired and preprocessed to obtain a preprocessed depth image, specifically including:

[0014] The depth image of the anchor net is acquired, and the depth image is subjected to bilateral filtering processing to obtain a preprocessed depth image.

[0015] Optionally, the preprocessed depth image is subjected to local region division to obtain a plurality of grid regions, specifically including:

[0016] The preprocessed depth image is equally divided into n grid regions with k*k pixel points, and let r i represent the i-th divided grid region, i=1, 2, …n.

[0017] Optionally, each grid region is subjected to adaptive threshold segmentation and merging to obtain a binary image with the same size as the depth image, specifically including:

[0018] The average gray value mean i of all pixel points in each grid region r i is calculated.

[0019] The best inter-class segmentation threshold otsu i of each grid region r i is calculated using the OpenCV maximum inter-class variance method.

[0020] According to the average gray value mean i and the best inter-class segmentation threshold otsu i , the binary image t i (x, y) of each grid region r i after adaptive threshold segmentation is determined based on the formula i ; wherein threshold1 is a segmentation threshold; o i (x, y) is the binary image of the grid region r i after maximum inter-class variance method segmentation.

[0021] The binary image t i (x, y) of each grid region r j after adaptive threshold segmentation is merged according to the original position to obtain a binary image T(x, y) with the same size as the depth image.

[0022] Optionally, a contour search operation is performed on the binary image of the same size as the depth image to obtain a plurality of contours, and a minimum circumscribed rectangle of each contour is fitted to obtain a plurality of minimum circumscribed rectangles, specifically comprising:

[0023] A contour search operation is performed on the binary image T(x, y) of the same size as the depth image using the cv: :findContours method in OpenCV to obtain a plurality of contours contour j ; j = 1, 2, … m; m is the number of all contours found;

[0024] All contours are traversed, and a minimum circumscribed rectangle of each contour contour j is fitted to obtain a plurality of minimum circumscribed rectangles rect j ; The feature structure of the minimum circumscribed rectangle rect j includes the minimum circumscribed rectangle center (x j , y j ), width width j , height height j and rotation angle θ j .

[0025] Optionally, the non-target minimum circumscribed rectangle in the plurality of minimum circumscribed rectangles is excluded by a preset threshold to obtain a target minimum circumscribed rectangle, specifically comprising:

[0026] The aspect ratio ratio j of each minimum circumscribed rectangle rect j is calculated;

[0027] According to a preset threshold threshold2, a non-target minimum circumscribed rectangle in a plurality of minimum circumscribed rectangles rect j is excluded to obtain a target minimum circumscribed rectangle rect p ; p = 1, 2, … s; s is the number of target minimum circumscribed rectangles.

[0028] Optionally, the position of the anchor net support hole center point is determined according to the target minimum circumscribed rectangle, specifically comprising:

[0029] The center point (x p , y p ) of the target minimum circumscribed rectangle rect p is determined as the position of the anchor net support hole center point.

[0030] A depth image-based anchor net support hole center point positioning system, comprising:

[0031] A preprocessing module for obtaining a depth image of an anchor net and performing preprocessing to obtain a preprocessed depth image;

[0032] The local region segmentation module is used to segment the preprocessed depth image into local regions to obtain multiple grid regions.

[0033] The binary image determination module is used to perform adaptive threshold segmentation and merging on each of the grid regions to obtain a binary image with the same size as the depth image;

[0034] The minimum bounding rectangle determination module is used to perform contour search operation on the binary image with the same size as the depth image to obtain multiple contours, and fit the minimum bounding rectangle of each contour to obtain multiple minimum bounding rectangles.

[0035] The target minimum bounding rectangle determination module is used to exclude non-target minimum bounding rectangles from multiple minimum bounding rectangles by using a preset threshold, and obtain the target minimum bounding rectangle.

[0036] The center point location determination module is used to determine the location of the center point of the anchor mesh support hole based on the minimum bounding rectangle of the target.

[0037] An electronic device includes a memory and a processor, the memory storing a computer program, and the processor running the computer program to enable the electronic device to perform the method for locating the center point of anchor mesh support holes based on depth images.

[0038] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for locating the center point of anchor mesh support holes based on depth images.

[0039] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0040] This invention provides a method, system, and device for locating the center point of anchor mesh support holes based on depth images. The method involves dividing a preprocessed depth image into multiple grid regions, then performing adaptive threshold segmentation on each grid region and merging them to obtain a binary image with the same size as the depth image. A contour search operation is then performed on the binary image to obtain multiple minimum bounding rectangles. Non-target minimum bounding rectangles are excluded from these rectangles using a preset threshold to obtain the target minimum bounding rectangle. Finally, the position of the anchor mesh support hole center point is determined based on the target minimum bounding rectangle. This effectively improves the robustness and efficiency of anchor mesh support hole center point location in mines, enabling automated anchoring robots to locate the center point of anchor mesh support holes more accurately and quickly under arbitrary lighting conditions in mines. Attached Figure Description

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0042] Figure 1 The flow chart of the anchor net supporting hole center point positioning method based on the depth image provided by the embodiment of the present application is shown in the following table.

[0043] Figure 2 The depth image schematic diagram of the anchor net provided by the embodiment of the present application is shown in the following table.

[0044] Figure 3 The depth image schematic diagram of the pre-processed provided by the embodiment of the present application is shown in the following table.

[0045] Figure 4 The depth image schematic diagram of the local area division provided by the embodiment of the present application is shown in the following table.

[0046] Figure 5 The depth image schematic diagram of the adaptive threshold segmentation provided by the embodiment of the present application is shown in the following table.

[0047] Figure 6 The depth image fitting after the fitting of the depth image provided by the embodiment of the present application is shown in the following table.

[0048] Figure 7 The schematic diagram of the final determined anchor net hole center point provided by the embodiment of the present application is shown in the following table. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments only constitute some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0050] The purpose of the present application is to provide a deep image-based anchor net supporting hole center positioning method, system and device, which can improve the robustness and work efficiency of the automatic anchor net supporting robot in the mine under any lighting conditions. The depth image is generally obtained by a depth camera and does not contain color information, so the depth information in the depth image is used for anchor net supporting hole center detection, which can overcome the defects of the color image detection method which is easily affected by color, light and other factors, and because the color image information is not used, the accurate and rapid detection and positioning of the mine anchor net supporting hole center can be carried out in any light or even no light environment.

[0051] In order to make the above-mentioned purpose, characteristics and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0052] Embodiment 1

[0053] Figure 1 The flow chart of the deep image-based anchor net supporting hole center positioning method provided by the embodiment of the present application is shown in Figure 1 The embodiment of the present application provides a deep image-based anchor net supporting hole center positioning method, which comprises the following steps:

[0054] Step 1: Obtain the depth image of the anchor net and perform preprocessing to obtain the preprocessed depth image.

[0055] Figure 2 The deep image of the anchor net provided by the embodiment of the present application is shown in Figure 2 In the obtained depth image of a certain area of the anchor net, all straight lines are anchor nets, and the remaining parts are backgrounds (rock walls); since the anchor net is a support, it is naturally closer to the depth camera than the background, which is reflected in the depth image that the gray value of the anchor net is smaller (darker) than that of the background in each area.

[0056] The preprocessing of the obtained depth image can preserve the edge features of the anchor net while reducing noise. As a specific embodiment, the obtained anchor net depth image is subjected to bilateral filtering processing, which can remove noise and also preserve the edge information of the depth image. Specifically, the cv: :bilateralFilter method in OpenCV is used to perform bilateral filtering processing on the original depth image to obtain the preprocessed depth image, as shown in Figure 3

[0057] Step 2: Divide the preprocessed depth image into local regions to obtain a plurality of grid regions.

[0058] ​Local region segmentation of the preprocessed depth image refers to dividing the preprocessed depth image obtained in step 1 into n grid regions of size k*k pixels each, such as... Figure 4 As shown, and define r i (i = 1, 2, ..., n) represents the i-th grid region.

[0059] Step 3: Perform adaptive thresholding and then merge each grid region to obtain a binary image with the same size as the depth image.

[0060] Step 3 specifically includes:

[0061] Step 3.1: Calculate r for each of the grid regions. i The mean gray value of all pixels within. i .

[0062] Specifically, for each grid region r i (i = 1, 2, ..., n), calculate the average gray value of all pixels within this grid region, denoted as mean. i (i=1, 2,...n), then Where v i (x, y) is the value of each grid region r. i The gray value of (i = 1, 2, ..., n) at point (x, y) is between [0, 255].

[0063] Step 3.2: Calculate the r of each mesh region using OpenCV's Otsu's method. i Optimal inter-class splitting threshold otsu i .

[0064] Specifically, for each grid region r i (i = 1, 2, ..., n), the optimal inter-class segmentation threshold for each grid region is calculated using the OpenCV Otsu's method (Otsu's method), denoted as otsu. i (i = 1, 2, ..., n), whose values ​​are between [0, 255].

[0065] Step 3.3: To accurately separate the anchor mesh portion from the background (rock wall) portion, based on the mean gray value... i and the optimal inter-class segmentation threshold otsu i Based on the formula Determine each of the grid regions r i Binary image t after adaptive thresholding i (x, y); where threshold1 is the segmentation threshold; o i (x, y) represents the grid region r.i the binary image segmented by the maximum inter-class variance method.

[0066] Specifically, for each grid region r i (i = 1, 2, … n), given a segmentation threshold threshold1, according to mean i (i = 1, 2, … n) and otsu i (i = 1, 2, … n), the grid region is threshold binarized, and the specific judgment method is:

[0067] If |mean i -otsu i | < threshold1, it indicates that the local grid region depth value difference is not large, that is, the grid region does not contain anchor nets, but only contains background, so no segmentation operation is performed to avoid false division, and the gray value of all pixel points of the grid region r i (i = 1, 2, … n) is assigned as 255.

[0068] If |mean i -otsu i | ≥ threshold1, it indicates that the local grid region depth value difference is large, that is, the grid region contains not only anchor nets, but also background, so segmentation operation is performed; the binary image segmented by the maximum inter-class variance method (Otsu method) is calculated using OpenCV maximum inter-class variance method, denoted as o i (x, y) (i = 1, 2, … n), wherein o i (x, y) is the gray value of the i-th grid region at (x, y).

[0069] In summary, for the grid region r i (i = 1, 2, … n), set t i (x, y) (i = 1, 2, … n) as the threshold processed gray value of the i-th grid region at (x, y), then:

[0070]

[0071] Step 3.4: Merge each of the grid regions r i the binary image segmented by the adaptive threshold t i (x, y) according to the original position to obtain a binary image T(x, y) with the same size as the depth image.

[0072] Specifically, for each grid region r i (i = 1, 2, … n), the binary image t i(x, y) are merged according to the original position to obtain a binary image T(x, y) with the same size as the original depth image, and the part with a gray value of 255 (white part) in T(x, y) is the background and the part with a gray value of 0 (black part) is the anchor net, as shown in Figure 5 .

[0073] Step 4: performing a contour search operation on the binary image with the same size as the depth image to obtain a plurality of contours, and fitting a minimum circumscribed rectangle of each contour to obtain a plurality of minimum circumscribed rectangles.

[0074] The step 4 specifically comprises:

[0075] Step 4.1: performing a contour search operation on the binary image T(x, y) with the same size as the depth image by using the cv: :findContours method in OpenCV to obtain a plurality of contours contour j ; j = 1, 2, … m; m is the number of all contours found, and each contour is composed of a plurality of points.

[0076] Step 4.2: traversing all contours, fitting a minimum circumscribed rectangle of each contour contour j to obtain a plurality of minimum circumscribed rectangles rect j ; the feature structure of the minimum circumscribed rectangle rect j includes a minimum circumscribed rectangle center (x j , y j ), a width width j , a height height j and a rotation angle θ j .

[0077] Specifically, for each contour contour j (j = 1, 2, … m), a minimum circumscribed rectangle fitting operation is performed on each contour by using the cv: :minAreaRect method in OpenCV to obtain a set of minimum circumscribed rectangle feature information, denoted as rect j (j = 1, 2, … m).

[0078] Step 5: excluding non-target minimum circumscribed rectangles in the plurality of minimum circumscribed rectangles by a preset threshold to obtain a target minimum circumscribed rectangle.

[0079] The step 5 specifically comprises:

[0080] Step 5.1: calculating an aspect ratio ratio j of each minimum circumscribed rectangle rect j .

[0081] AsFigure 6 As shown in the acquired depth image, there are always some circumscribed rectangles that are wrongly fitted around the edge area, therefore, the present application introduces the aspect ratio of the minimum circumscribed rectangle to exclude the non-target minimum circumscribed rectangle according to the feature that the height and width of the target minimum circumscribed rectangle are not much different, and the height and width of the non-target minimum circumscribed rectangle are greatly different. Specifically, let ratio j (j=1,2,…m) be the aspect ratio of the jth minimum circumscribed rectangle, then ratio j =height j / width j .

[0082] Step 5.2: excluding the non-target minimum circumscribed rectangle from the plurality of minimum circumscribed rectangles rect j according to a preset threshold threshold2, to obtain the target minimum circumscribed rectangle rect p ; p=1,2,…s; s is the number of target minimum circumscribed rectangles.

[0083] Specifically, given a threshold threshold2 (threshold2>1), if ratio j >threshold2 or 1 / ratio j >threshold2, the minimum circumscribed rectangle rect j is excluded; otherwise, the minimum circumscribed rectangle rect j is retained, denoted as rect p (p=1,2,…s, s≤m), which has the same feature structure as rect j .

[0084] Step 6: determining the position of the anchor net supporting hole center point according to the target minimum circumscribed rectangle.

[0085] The center point (x p , y p ) of the target minimum circumscribed rectangle rect p is determined as the position of the anchor net supporting hole center point. Denote the final anchor net hole center point as set P={(x p , y p )|p∈1,2,…,s}, then set P is the final anchor net hole center point, as shown in Figure 7 .

[0086] Embodiment 2

[0087] The present application also provides a depth image-based anchor net supporting hole center point positioning system, comprising:

[0088] A preprocessing module is configured to acquire a depth image of the anchor net and perform preprocessing to obtain a preprocessed depth image.

[0089] A local region division module is configured to divide the preprocessed depth image into a plurality of grid regions.

[0090] A binary image determination module is configured to perform adaptive threshold segmentation and merging on each grid region to obtain a binary image with the same size as the depth image.

[0091] A minimum bounding rectangle determination module is configured to perform a contour search operation on the binary image with the same size as the depth image to obtain a plurality of contours, and fit a minimum bounding rectangle for each contour to obtain a plurality of minimum bounding rectangles.

[0092] A target minimum bounding rectangle determination module is configured to exclude non-target minimum bounding rectangles from the plurality of minimum bounding rectangles by a preset threshold to obtain a target minimum bounding rectangle.

[0093] A center point position determination module is configured to determine the position of the anchor net support hole center point according to the target minimum bounding rectangle.

[0094] The anchor net support hole center point positioning method and system based on a depth image provided by the present application determines the final anchor net hole center point through depth image preprocessing, depth image local region division, calculation of local grid region average gray value, determination of a final segmentation threshold, adaptive threshold segmentation of each local grid region, merging of all local grid regions, contour search on the obtained binary image, and given threshold exclusion of non-target minimum bounding rectangles, so as to achieve the purpose of accurate and rapid detection and positioning of the mine anchor net support hole center point under any illumination condition.

[0095] Embodiment 3

[0096] The present application also provides an electronic device comprising a memory and a processor, the memory being configured to store a computer program, and the processor being configured to run the computer program to enable the electronic device to perform the anchor net support hole center point positioning method based on a depth image.

[0097] Embodiment 4

[0098] In one embodiment, the present application also provides a computer readable storage medium storing a computer program, the computer program being executed by a processor to implement the anchor net support hole center point positioning method based on a depth image.

[0099] The method of the present application has the following advantages over the prior art:

[0100] 1) The method provided by the present application uses a single-channel depth image instead of a color image as a medium for segmenting the anchor net and the background. Since the depth image is almost not affected by light, using the depth image can well solve the problem of poor robustness of the anchor net support hole center point detection method based on color images under complex lighting conditions in mines.

[0101] 2) In the depth image provided by the method, the gray value intensity of the pixel represents the distance. According to the gray intensity difference between the anchor net and the background (that is, the distance difference between the anchor net and the background), the two can be accurately segmented.

[0102] 3) The method provided by the present application considers that directly performing adaptive threshold segmentation on the depth image containing the anchor net and the background will result in poor segmentation effect. Therefore, the depth image is divided into n grid regions of the same size, adaptive threshold segmentation is performed on each grid region, and then the grid regions are merged, which can greatly improve the segmentation effect.

[0103] In the specification, each embodiment is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method part.

[0104] The principles and implementation manners of the present application are described by using specific examples in this paper. The above embodiment description is only used to help understand the method of the present application and its core idea. For those skilled in the art, according to the idea of the present application, the specific implementation manner and application range can be changed. In summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method for locating the center point of an anchor net support hole based on a depth image, characterized in that, The method comprises the following steps: obtaining a depth image of the anchor net and pre-processing to obtain a pre-processed depth image; dividing the pre-processed depth image into local regions to obtain a plurality of grid regions; performing adaptive threshold segmentation and merging on each grid region to obtain a binary image with the same size as the depth image; performing a contour search operation on the binary image with the same size as the depth image to obtain a plurality of contours, and fitting a minimum bounding rectangle for each contour to obtain a plurality of minimum bounding rectangles; excluding non-target minimum bounding rectangles from the plurality of minimum bounding rectangles by a preset threshold to obtain target minimum bounding rectangles; determining the position of the anchor net support hole center point according to the target minimum bounding rectangle.

2. The method for locating the center point of the support and protection hole according to claim 1, characterized in that, The step of obtaining a depth image of the anchor net and pre-processing to obtain a pre-processed depth image specifically comprises: obtaining a depth image of the anchor net, and performing bilateral filtering processing on the depth image to obtain a pre-processed depth image.

3. The method for locating the center point of the support and protection hole according to claim 1, characterized in that, The step of dividing the pre-processed depth image into local regions to obtain a plurality of grid regions specifically comprises: The pre-processed depth image is equally divided into n grid regions with k*k pixels, and let r i represents the i-th divided grid region, i = 1, 2, … n.

4. The method for locating the center point of the anchor net support hole according to claim 3, characterized in that, The step of performing adaptive threshold segmentation and merging on each grid region to obtain a binary image with the same size as the depth image specifically comprises: calculating the average gray value mean of all pixel points in each of the grid regions r i i ;​ The optimal inter-class segmentation threshold otsu of each of the grid regions r is calculated using the OpenCV method of maximum variance between classes i i ;​ According to the average gray value mean i And the optimal inter-class segmentation threshold otsu i , based on the formula Determine each of the grid area r i Binary image t after adaptive threshold segmentation i (x,y); wherein threshold1 is the segmentation threshold; o i (x,y) is the grid area r i Binary image after maximum inter-class variance method segmentation; Each of the grid regions r i The binary image t after adaptive threshold segmentation i (x, y) are merged according to the original position to obtain a binary image T(x, y) with the same size as the depth image.

5. The method for locating the center point of the support and protection hole according to claim 4, characterized in that, The step of performing a contour search operation on the binary image with the same size as the depth image to obtain a plurality of contours, and fitting a minimum bounding rectangle for each contour to obtain a plurality of minimum bounding rectangles specifically comprises: The cv: findContours method in OpenCV is used to find contours of the binary image T(x, y) with the same size as the depth image, and a plurality of contours contour are obtained j j = 1, 2, … m; m is the number of all contours found Traverse all contours and fit a contour to each contour. j Find the minimum bounding rectangle, and obtain multiple minimum bounding rectangles rect. j The minimum circumscribed rectangle rect j The characteristic structure includes the center of the minimum bounding rectangle (x j y j ), width j height j and rotation angle θ j .

6. The method for locating the center point of the support and protection hole according to claim 5, characterized in that, The step of excluding non-target minimum bounding rectangles from the plurality of minimum bounding rectangles by a preset threshold to obtain target minimum bounding rectangles specifically comprises: calculating a height-to-width ratio ratio of each of the minimum bounding rectangles rect j ; calculating a height-to-width ratio ratio of each of the minimum bounding rectangles rect j ; excluding non-target minimum bounding rectangles from the plurality of minimum bounding rectangles rect according to a preset threshold threshold2, to obtain target minimum bounding rectangles rect j p p = 1, 2, … s; s is the number of target minimum bounding rectangles.​ 7. The method of locating the center point of a longwall panel entry according to claim 6, wherein, The step of determining the position of the anchor net support hole center point according to the target minimum bounding rectangle specifically comprises: determining the target minimum circumscribed rectangle rect p The center point (x p , y p ) of the target minimum circumscribed rectangle rect is the position of the anchor net support hole center point.

8. A system for locating the center point of an anchor net support hole based on a depth image, characterized by, The method comprises the following steps: a pre-processing module for obtaining a depth image of the anchor net and pre-processing to obtain a pre-processed depth image; a local region division module for dividing the pre-processed depth image into local regions to obtain a plurality of grid regions; a binary image determination module for performing adaptive threshold segmentation and merging on each grid region to obtain a binary image with the same size as the depth image; a minimum bounding rectangle determination module for performing a contour search operation on the binary image with the same size as the depth image to obtain a plurality of contours, and fitting a minimum bounding rectangle for each contour to obtain a plurality of minimum bounding rectangles; a target minimum bounding rectangle determination module for excluding non-target minimum bounding rectangles from the plurality of minimum bounding rectangles by a preset threshold to obtain target minimum bounding rectangles; a center point position determination module for determining the position of the anchor net support hole center point according to the target minimum bounding rectangle.

9. An electronic device, comprising: The electronic device comprises a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the anchor net support hole center point positioning method based on a depth image according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program stored in the memory is executed by the processor to implement the anchor net support hole center point positioning method based on a depth image according to any one of claims 1 to 7.

Images