A steel strip drilling positioning method, system and device based on monocular vision

Abstract

The application relates to a single-vision-based steel strip drilling positioning method, system and device, relates to the technical field of computer vision and steel strip drilling positioning, and the method comprises the following steps: acquiring an image to be processed; extracting a line laser image feature from the image to be processed, and processing the line laser image feature to obtain a steel strip feature map; performing image enhancement processing on the image to be processed according to the steel strip feature map, and determining a steel strip drilling center data group according to the image to be processed after the image enhancement processing; extracting a light spot image feature from the image to be processed, and pre-processing the light spot image feature to obtain a light spot feature map; extracting a light spot center of the light spot feature map by adopting ellipse fitting to obtain a light spot data group; calculating the pose of drilling on the steel strip according to the steel strip drilling center data and the light spot data group, and then controlling a drilling arm to move to a steel strip drilling position. The application improves the steel strip drilling positioning precision.

Description

Technical Field

[0001] This invention relates to the fields of computer vision and steel strip drilling positioning technology, and in particular to a method, system and device for steel strip drilling positioning based on monocular vision. Background Technology

[0002] With the research and application of various intelligent fully mechanized mining equipment, coal mining capacity and efficiency have been greatly improved. However, because coal mine roadway excavation technology is still in the mechanization stage, it leads to a serious imbalance in mining and excavation efficiency, resulting in a common problem of fast mining but slow excavation in the coal mining industry. Therefore, intelligent excavation and reducing manpower to improve efficiency are the current development trends in the coal mining industry. Based on this, the support work of coal mine roadways places higher demands on the coordination, accuracy, and speed of roadway excavation and drilling and anchoring equipment operations, posing a greater challenge to the technical level and physical fitness of relevant construction personnel. However, this further slows down the efficiency of roadway excavation, resulting in the problem of fast excavation but slow support.

[0003] To address the issue of rapid excavation but slow support in tunneling faces, binocular vision-based steel strip drilling positioning schemes and systems have gradually emerged in the industry, improving support efficiency and reducing the labor intensity of workers. However, the low-light and weak-texture working environment in coal mines has significantly impacted vision-based positioning schemes, and the fixed baseline distance and high computational complexity of binocular vision also affect its positioning accuracy and real-time performance. Summary of the Invention

[0004] The purpose of this invention is to provide a method, system, and device for positioning steel strip boreholes based on monocular vision, to solve the impact of low illumination and weak texture environment in underground coal mines on visual positioning, and to improve the positioning accuracy of steel strip boreholes.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] In a first aspect, the present invention provides a method for positioning steel strip drilling based on monocular vision, comprising:

[0007] Acquire an image to be processed; the image to be processed includes a roadway coal wall, a steel strip, a line laser beam, and multiple light spots; the steel strip is placed on the roadway coal wall, and multiple drill holes are provided on the steel strip;

[0008] Line laser image features are extracted from the image to be processed, and the line laser image features are processed to obtain a steel strip feature map;

[0009] The image to be processed is enhanced based on the steel strip feature map, and the steel strip borehole center data group is determined based on the enhanced image to be processed.

[0010] Extract spot image features from the image to be processed, and preprocess the spot image features to obtain a spot feature map;

[0011] Ellipse fitting is used to extract the center of the light spot feature map to obtain the light spot data set;

[0012] Based on the data set of the center of the drill hole in the steel strip and the data set of the spot, calculate the pose of the drill hole in the steel strip in the camera coordinate system;

[0013] Based on the position of the hole in the steel strip in the camera coordinate system, control the drill arm to move to the position of the hole in the steel strip.

[0014] Optionally, image enhancement processing is performed on the image to be processed based on the steel strip feature map, and the steel strip borehole center data group is determined based on the image to be processed after image enhancement processing, specifically including:

[0015] Based on the steel strip feature map, the image to be processed is subjected to image enhancement processing based on the Laplacian operator to determine the preliminary enhanced image;

[0016] Thresholding is performed on the preliminary enhanced image to determine the outline image of the steel strip drill hole;

[0017] Ellipse fitting is performed on the outline image of the steel strip drill hole to determine the pixel coordinates of the center of the steel strip drill hole in the pixel coordinate system.

[0018] The pixel coordinates of the center of the steel strip drilling in the pixel coordinate system are transformed to determine the image coordinates of the center of the steel strip drilling in the corresponding image coordinate system; the image coordinates of the center of the steel strip drilling in the corresponding image coordinate system constitute the data group of the center of the steel strip drilling.

[0019] Optionally, the light spot image features are preprocessed to obtain a light spot feature map, specifically including:

[0020] The light spot image features are sequentially processed by pyramid downsampling, binarization segmentation, Canny edge detection, and pyramid upsampling to obtain a light spot feature map.

[0021] The step of extracting the center of the light spot from the feature map using ellipse fitting to obtain a light spot data set specifically includes:

[0022] The center of the light spot in the feature map is extracted by ellipse fitting based on Hough transform to determine multiple center points of the light spot;

[0023] The center points of the multiple light spots are arranged in a first order; the first order is clockwise or counterclockwise.

[0024] For two adjacent light spot center points after sorting, a midpoint is determined between the two light spot center points; the midpoint is located on the line connecting the two adjacent light spot center points; multiple light spot center points and multiple midpoints constitute a light spot data group.

[0025] Optionally, based on the data set of the drilled center on the steel strip and the data set of the light spot, the pose of the drilled hole on the steel strip in the camera coordinate system is calculated, specifically including:

[0026] Calculate the first translation vector based on the center point of the light spot in the light spot data group; the first translation vector is the translation vector from the camera coordinate system to the roadway coal wall coordinate system;

[0027] Calculate the first rotation matrix based on the midpoint of the light spot data set; the first rotation matrix is ​​the rotation matrix from the camera coordinate system to the roadway coal wall coordinate system;

[0028] Calculate the second translation vector based on the data set of the center of the steel strip borehole; the second translation vector is the translation vector from the steel strip borehole coordinate system to the roadway coal wall coordinate system;

[0029] The pose of the drill hole on the steel strip in the camera coordinate system is calculated based on the first translation vector, the first rotation matrix, and the second translation vector.

[0030] Optionally, the pose of the drill hole on the steel strip in the camera coordinate system is calculated based on the first translation vector, the first rotation matrix, and the second translation vector, specifically including:

[0031] According to the formula

[0032] T HC =T HW ·T WC

[0033] Calculate the pose of the drilled hole on the steel strip in the camera coordinate system;

[0034] Among them, T HC T represents the pose of the hole drilled on the steel strip in the camera coordinate system. WC T represents the pose of the coal face coordinate system in the camera coordinate system. HW This indicates the pose of the steel strip drilling coordinate system within the roadway coal wall coordinate system. R WC Let t represent the first rotation matrix. WC Let t represent the first translation vector. HW Denotes the second translation vector.

[0035] Optionally, the first translation vector is calculated based on the center point of the light spot in the light spot data group, specifically including:

[0036] According to the formula

[0037]

[0038] Calculate the first translation vector;

[0039] Among them, t WC Denotes the first translation vector. a represents half the distance between the center points of two adjacent light spots after sorting, f represents the focal length of the camera that took the image to be processed, OA′1 represents the distance between point O and point A′1, OC′1 represents the distance between point O and point C′1, point O is the origin of the light spot feature map in the camera coordinate system, point A′1 represents the center point of one light spot in the light spot data group, and point C′1 represents the center point of another light spot in the light spot data group that is not adjacent to point A′1.

[0040] Optionally, the first rotation matrix is ​​calculated based on the midpoint of the light spot data group, specifically including:

[0041] According to the formula

[0042]

[0043] Calculate the first rotation matrix;

[0044] Among them, R WC Denotes the first rotation matrix. This represents the X-coordinate of point E1 in the roadway coal wall coordinate system. Point E1 represents the mapping point of any intermediate point in the spot data set in the roadway coal wall coordinate system. This represents the X-coordinate of point E2 in the roadway coal wall coordinate system. Point E2 also represents the mapping point of another intermediate point in the spot data set in the roadway coal wall coordinate system. This represents the Y-coordinate of point F1 in the roadway coal wall coordinate system. Point F1 represents the mapping point of another intermediate point in the spot data set in the roadway coal wall coordinate system. This represents the Z-coordinate of point E1 in the camera coordinate system. This represents the Z-coordinate of point E2 in the camera coordinate system. This represents the Z coordinate of point F1 in the camera coordinate system, and 'a' represents half the distance between the center points of two adjacent light spots after sorting.

[0045] Optionally, based on the steel strip borehole center data set, a second translation vector is calculated, specifically including:

[0046] According to the formula

[0047]

[0048] Calculate the second translation vector;

[0049] in, This represents the X-coordinate of the origin of the steel strip drilling coordinate system in the roadway coal wall coordinate system. H represents the Y-coordinate of the origin of the steel strip drilling coordinate system in the roadway coal wall coordinate system, and H represents the thickness of the steel strip.

[0050] Secondly, the present invention provides a steel strip drilling positioning system based on monocular vision, comprising:

[0051] An image acquisition module is used to acquire an image to be processed; the image to be processed includes a roadway coal wall, a steel strip, a line laser beam, and multiple light spots; the steel strip is placed on the roadway coal wall, and multiple drill holes are provided on the steel strip;

[0052] The steel strip feature extraction module is used to extract line laser image features from the image to be processed, and to process the line laser image features to obtain a steel strip feature map.

[0053] The steel strip data calculation module is used to perform image enhancement processing on the image to be processed based on the steel strip feature map, and to determine the steel strip borehole center data group based on the image to be processed after image enhancement processing.

[0054] A spot feature extraction module is used to extract spot image features from the image to be processed and to preprocess the spot image features to obtain a spot feature map.

[0055] The spot data calculation module is used to extract the spot center of the spot feature map by ellipse fitting to obtain the spot data set;

[0056] The drilling pose calculation module is used to calculate the pose of the drill hole on the steel strip in the camera coordinate system based on the drilling center data group of the steel strip and the spot data group.

[0057] The drill arm motion control module is used to control the drill arm to move to the drilling position of the steel strip according to the pose of the hole on the steel strip in the camera coordinate system.

[0058] Thirdly, the present invention provides a steel strip drilling positioning device based on monocular vision. The steel strip drilling positioning device is installed on a drilling and anchoring robot. The steel strip drilling positioning device includes a circulating line laser emitting component, a point laser emitting component, a monocular image acquisition component, and a data processing and motion control component.

[0059] The circulating linear laser emitting assembly is positioned above the drilling and anchoring robot. The circulating linear laser emitting assembly is used to circulate and project linear laser beams onto the roadway coal wall and the steel strip. The steel strip is placed on the roadway coal wall and has multiple drill holes.

[0060] Both the point laser emitting component and the monocular image acquisition component are mounted on the drill arm of the drilling and anchoring robot; the point laser emitting component is used to project multiple light spots onto the roadway coal wall and the steel strip;

[0061] The monocular image acquisition component is used to acquire an image to be processed; the image to be processed includes the roadway coal wall, the steel strip, the line laser beam, and multiple light spots;

[0062] The data processing and motion control component is connected to the monocular image acquisition component, and the data processing and motion control component is used to execute the monocular vision-based steel strip drilling positioning method described in the first aspect.

[0063] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0064] This invention provides a method, system, and apparatus for locating steel strip boreholes based on monocular vision. First, an image to be processed is acquired, containing the coal face of the roadway, the steel strip, a line laser beam, and multiple light spots. Line laser image features are extracted from the image to be processed, and these features are processed to obtain a steel strip feature map. Image enhancement processing is then performed on the image to be processed based on the steel strip feature map, and the center data group of the steel strip borehole is determined based on the enhanced image. By performing the above-mentioned line laser image feature extraction and image enhancement processing on the image to be processed, the influence of weak texture and low illumination conditions in underground coal mines on visual positioning can be overcome, and the identification of steel strips and boreholes becomes faster and more accurate. The laser spot image features are extracted from the image to be processed and preprocessed to obtain a laser spot feature map. Ellipse fitting is used to extract the center of the laser spot from the feature map, resulting in a laser spot data set. Then, based on the determined center data set of the steel strip borehole and the laser spot data set, the pose of the borehole in the steel strip in the camera coordinate system is calculated. This fuses the line laser features and laser spot features in the image to be processed, enabling them to work together to locate the borehole in the steel strip, improving positioning accuracy and reducing computational complexity. Finally, based on the obtained pose of the borehole in the steel strip in the camera coordinate system, the drill arm is controlled to move to the borehole position in the steel strip, completing the drilling and anchoring operation. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 This is a flowchart illustrating the steel strip drilling positioning method based on monocular vision according to the present invention.

[0067] Figure 2 This is a schematic diagram of the coordinate system of the steel strip drilling pose calculation model in this invention;

[0068] Figure 3 The first translation vector t of this invention WC Solution model diagram;

[0069] Figure 4 R is the first rotation matrix of this invention. WC Solution model illustration Figure 1 ;

[0070] Figure 5 R is the first rotation matrix of this invention. WC Solution model illustration Figure 2 ;

[0071] Figure 6 The second translation vector t of this invention HW Solution model diagram;

[0072] Figure 7 This is a flowchart illustrating the working process of the steel strip drilling positioning system based on monocular vision of the present invention.

[0073] Figure 8 This is a schematic diagram of the steel strip drilling positioning device based on monocular vision according to the present invention;

[0074] Figure 9 This is a schematic diagram of the structure of the cyclic line laser emitting assembly in this invention;

[0075] Figure 10 This is a schematic diagram of the structure of the point laser emission component and the monocular image acquisition component.

[0076] Symbol explanation:

[0077] 1-Light spot, 2-Laser emission and acquisition component, 21-Camera explosion-proof glass, 22-Camera front explosion-proof shell, 23-Point laser emitter explosion-proof glass, 24-High-definition industrial camera, 25-Switching power supply, 26-Point laser emitter, 27-Camera rear explosion-proof shell, 3-Drill arm, 4-Drilling and anchoring robot, 5-Circulating line laser emission component, 51-Upper explosion-proof shell, 52-Explosion-proof glass, 53-Lower explosion-proof shell, 54-Drive motor, 55-Line laser emitter, 56-Double-sided reflector, 6-Data processing and motion control components, 7-Line laser beam, 8-Steel strip, 9-Drill hole. Detailed Implementation

[0078] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0079] This invention provides a method, system, and device for positioning steel strip drilling based on monocular vision. It performs well in the low-light and weak-texture working environment of underground coal mines, and has a simple structure and accurate positioning, making it of great potential for widespread application.

[0080] To make the objectives, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0081] Example 1

[0082] like Figure 1 As shown, this embodiment provides a method for positioning steel strip drilling based on monocular vision, including:

[0083] Step 100: Obtain the image to be processed; the image to be processed includes the roadway coal wall, steel strip, line laser beam, and multiple laser spots; the steel strip is placed on the roadway coal wall, and multiple drill holes are provided on the steel strip. Specifically, a green line laser beam is projected onto the roadway coal wall by the line laser emitter in the circulating line laser emitting assembly, and a red dot laser spot is projected onto the roadway coal wall by the dot laser emitter in the dot laser emitting assembly. Correspondingly, the part of the image to be processed covered by the line laser beam is green, and the part of the image to be processed covered by the dot laser spot is red.

[0084] Step 200: Extract line laser image features from the image to be processed, and process the line laser image features to obtain a steel strip feature map.

[0085] Preferably, step 200 specifically includes:

[0086] (1) Extract line laser image features from the image to be processed:

[0087] The G channel image is separated from the image to be processed (RGB image) to highlight the line laser features, reduce the interference of other features, and lay the foundation for subsequent steps.

[0088] (2) Process the features of the line laser image to obtain a steel strip feature map:

[0089] 1) The linear laser image features are sequentially processed by pyramid downsampling, low-pass filtering, median filtering, binarization segmentation, morphological opening, and pyramid upsampling to obtain a preliminary preprocessed image.

[0090] Specifically, the steps of pyramid downsampling include:

[0091] The G-channel image in step (1) is Gaussian smoothed using a convolution kernel of size K, and then resampled to remove even-numbered rows and columns, resulting in a new image with half the original size. Pyramid downsampling reduces the image size by half, decreasing the number of pixel traversals in subsequent steps, reducing processing time, and improving real-time performance. Gaussian smoothing reduces the stitching effect after pyramid downsampling, making the resampled image more accurate and natural.

[0092] The steps of low-pass filtering include:

[0093] A trapezoidal low-pass filter with a cutoff frequency of D0 is used to filter out high-frequency noise and remove stray light interference from the coal face in the roadway. This step is performed because the steel strip surface is relatively flat, resulting in a smaller rate of change in image grayscale, while the coal face surface is rough, leading to a larger rate of change in image grayscale; therefore, low-pass filtering is necessary.

[0094] The steps of median filtering include:

[0095] The median gray value of the pixels within the K0×K0 region (where K0 is an odd number) is used as the gray value of the region's center point to initially filter out noise in the image and enhance image quality.

[0096] The steps of binarization segmentation include:

[0097] The algorithm iterates through all pixels in the image after median filtering. Pixels with a gray level greater than or equal to a set threshold (Threshold) are initially identified as steel strip regions, and their gray level is set to 255. Otherwise, their gray level is set to 0, and a binarized image is output. Specifically, considering that the surface of the steel strip is smooth and has less diffuse reflection of light, resulting in a higher image gray level, while the surface of the coal wall in the roadway is rough and has severe diffuse reflection, resulting in a lower image gray level, binarization segmentation is performed.

[0098] Morphological opening operations include:

[0099] The image is first eroded and then dilated, followed by morphological opening operations to further filter out edge noise in the image.

[0100] Pyramid upsampling processing includes:

[0101] Pyramid upsampling is performed on the image after morphological opening to restore the original size of the image, resulting in a preliminary preprocessed image.

[0102] 2) Perform image region connectivity processing and image contour detection processing on the preliminary preprocessed image in sequence to determine the steel strip feature map.

[0103] Specifically, a window of size M0×N0 is used to move and search across the pre-processed image. If the grayscale value of any pixel in the window region is not zero, the entire window region is set to 255 to connect the image regions. Then, contour detection is performed on the connected image regions to obtain the coordinates (X0, Y0) of the upper left corner of the contour boundary rectangle and the width and height W0, H0 of the rectangle, in order to determine the steel strip feature map. The steel strip feature map records the position and size of the steel strip region in the image.

[0104] Step 300: Perform image enhancement processing on the image to be processed based on the steel strip feature map, and determine the drilling center data of the steel strip based on the image to be processed after image enhancement processing.

[0105] Step 300 specifically includes:

[0106] (1) The image to be processed is subjected to Laplacian operator-based image enhancement processing based on the steel strip feature map to determine the preliminary enhanced image. Specifically, the steel strip region determined by the steel strip feature map in the original image (image to be processed) is subjected to Laplacian operator-based image enhancement processing, and all pixels outside the steel strip region are set to zero.

[0107] (2) The preliminary enhanced image is binarized and segmented to determine the outline image of the steel strip drill hole.

[0108] (3) Ellipse fitting is performed on the outline image of the steel strip drill hole to determine the pixel coordinates of the center of the steel strip drill hole in the pixel coordinate system. Specifically, ellipse fitting based on Hough transform is used to extract the pixel coordinates of the center of the steel strip drill hole.

[0109] (4) Perform coordinate system transformation on the coordinates of the steel strip drilling center in the pixel coordinate system to determine the coordinates of the steel strip drilling center in the corresponding image coordinate system; the coordinates of the steel strip drilling center in the corresponding image coordinate system constitute the steel strip drilling center data set. Specifically, through the formula...

[0110]

[0111] The coordinates of the steel strip drill hole center in the pixel coordinate system are transformed to the corresponding image coordinate system and stored in the steel strip drill hole center data group vectorA. Here, u and v are the coordinates of the steel strip drill hole center in the pixel coordinate system, x and y are the coordinates of the steel strip drill hole center in the image coordinate system, and K is the camera's intrinsic parameter matrix.

[0112] Step 400: Extract the light spot image features from the image to be processed, and preprocess the light spot image features to obtain a light spot feature map.

[0113] Preferably, step 400 specifically includes:

[0114] (1) Extracting spot image features from the image to be processed:

[0115] The R channel image is separated from the original RGB image (the image to be processed), highlighting the characteristics of the red laser spot and reducing the interference of other features, thus laying the foundation for subsequent steps.

[0116] (2) The light spot image features are preprocessed to obtain a light spot feature map:

[0117] The light spot image features are sequentially processed by pyramid downsampling, binarization segmentation, Canny edge detection, and pyramid upsampling to obtain the light spot feature map.

[0118] Specifically, the steps of pyramid downsampling include:

[0119] The R-channel image in step (1) is Gaussian smoothed using a convolution kernel of size K, and then the image is resampled to remove even-numbered rows and columns, resulting in a new image with half the original size. This reduces the processing time of subsequent algorithms and improves the real-time performance of the algorithm.

[0120] The steps of binarization segmentation include:

[0121] Iterate through all pixels in the image after the second pyramid downsampling process, and initially identify the pixels with a gray value greater than or equal to the set threshold Threshold as the red laser spot area, and set their gray value to 255; otherwise, set their gray value to 0, and output the binarized image.

[0122] The steps of Canny edge detection processing include:

[0123] Gaussian filtering is applied to the binarized image obtained after the second binarization segmentation using a convolution kernel of size K1. Then, the gradient magnitude and direction of each pixel are calculated using the Sobel operator. Finally, the edge contours of the image are detected through non-maximum suppression and double thresholding.

[0124] The steps of pyramid upsampling include:

[0125] Pyramid upsampling is performed on the contour image after Canny edge detection to restore the original size of the image and obtain the spot feature map.

[0126] Step 500: Extract the center of the light spot from the light spot feature map using ellipse fitting to obtain a light spot data set. Preferably, step 500 specifically includes:

[0127] (1) The center of the light spot in the feature map is extracted by ellipse fitting based on Hough transform to determine multiple center points of the light spot.

[0128] (2) Sort the multiple light spot center points in a first order; the first order is clockwise or counterclockwise. Specifically, ellipse fitting based on Hough transform is used to extract the four light spot centers in the image to be processed and they are respectively denoted as A′1, B′1, C′1 and D′1 in a clockwise direction from the upper left corner, and the coordinates of the four light spot centers in the pixel coordinate system are determined.

[0129] (3) For two adjacent light spot center points after sorting, determine the midpoint between the two light spot center points; the midpoint is located on the line connecting the two adjacent light spot center points; multiple light spot center points and multiple midpoints constitute a light spot data group. Specifically, the midpoint of line segment B′1C′1 is selected as the midpoint and denoted as E′1; the midpoint of line segment C′1D′1 is selected as the midpoint and denoted as E′2; the midpoint of line segment B′1C′1 is selected as the midpoint and denoted as F′1. In a specific embodiment, the midpoint can be any point in the line segment. The pixel coordinates of A′1, B′1, C′1, D′1, E′1, E′2, and F′1 are converted into image coordinates and stored in the light spot data group vectorB. For the specific steps of converting pixel coordinates into image coordinates, please refer to the steps of converting the center coordinates of the steel strip drilling above.

[0130] Step 600: Calculate the pose of the drill hole on the steel strip in the camera coordinate system based on the data set of the drill hole center on the steel strip and the data set of the light spot.

[0131] like Figure 2 As shown, a coordinate system for the steel strip drilling pose calculation model is first established. Specifically, for ease of analysis and description, a camera coordinate system O is established at the camera's optical center position. C X C Y C Z CThe origin is the camera's optical center, the X-axis is to the right, the Y-axis is downward, and the Z-axis is forward along the optical axis. A roadway coal wall coordinate system O is established at the center of the red laser spot quadrilateral (i.e., the quadrilateral formed by the four actual laser spots A1B1C1D1). W X W Y W Z W The origin of the roadway coal wall coordinate system is the center of the quadrilateral of the red laser spot, with the X-axis pointing to the right, the Y-axis downwards, and the Z-axis pointing inwards from the coal wall. A steel strip drilling coordinate system O is established at the actual center position of the steel strip borehole. H X H Y H Z H The origin of the steel strip drilling coordinate system is the center of the steel strip drilling, the X-axis is to the right, the Y-axis is downward, and the Z-axis points to the inside of the coal wall.

[0132] according to Figure 2 It is obvious that the camera coordinate system O C X C Y C Z C To the coal wall coordinate system O W X W Y W Z W The first rotation matrix R WC Rotational component γ about the Z-axis WC =0, first translation vector t WC Translational components X along the X and Y directions WC =0, Y WC =0. Assuming the local flatness of the coal face in the roadway is relatively good, then the coal face coordinate system O... W X W Y W Z w To the steel strip drilling coordinate system O H X H Y H Z H The second rotation matrix R HW All angular components are 0, therefore the second rotation matrix is ​​an identity matrix, and the second rotation matrix is ​​the rotation matrix from the steel strip borehole coordinate system to the roadway coal wall coordinate system; the second translation vector t HW The translational components along the X, Y, and Z directions are respectively And -H. Where H represents the thickness of the steel strip.

[0133] Further, step 600 specifically includes:

[0134] 1) Calculate the first translation vector t based on the center point of the light spot in the light spot data group. WCThe first translation vector is the translation vector from the camera coordinate system to the roadway coal wall coordinate system.

[0135] Read the coordinates of the center points A′1 and C′1 of the light spot data vectorB in the image coordinate system, and calculate the distances of each point to the origin O (the origin in the camera coordinate system):

[0136]

[0137]

[0138] like Figure 3 As shown, based on the coordinate system of the steel strip drilling pose calculation model established above, in triangle O C In triangle A1C1, due to the similarity of triangles, we have...

[0139]

[0140]

[0141] and

[0142]

[0143] Where A1, B1, C1, D1, E1, E2, and F1 are the mapping points of A′1, B′1, C′1, D′1, E′1, E′2, and F′1 in the roadway coal wall coordinate system, respectively, and are the actual light spots illuminating the roadway coal wall.

[0144] By combining the above three equations, we can determine the Z-component of the first translation vector in the Z direction. WC for:

[0145]

[0146] Based on this, according to the formula

[0147]

[0148] The camera coordinate system O is calculated. C X C Y C Z C To the coal face coordinate system O in the roadway W X W Y W Z W The first translation vector t WC .

[0149] Where a represents half the distance between the center points of two adjacent light spots after sorting, f represents the focal length of the camera that took the image to be processed, OA′1 represents the distance between the origin O and point A′1, OC′1 represents the distance between the origin O and point C′1, point O is the origin of the light spot feature map in the camera coordinate system, point A′1 represents the center point of one light spot in the light spot data group, and point C′1 represents the center point of another light spot in the light spot data group that is not adjacent to point A′1.

[0150] 2) Calculate the first rotation matrix R based on the midpoint of the light spot data set. WC The first rotation matrix is ​​the rotation matrix from the camera coordinate system to the roadway coal wall coordinate system.

[0151] Read the coordinates of the intermediate points E′1, E′2, and F′1 stored in the light spot data group vectorB, and calculate the distances from E′1 to the y-axis, E′2 to the y-axis, and F′1 to the x-axis, respectively:

[0152]

[0153]

[0154]

[0155] like Figure 4 As shown, based on the coordinate system of the steel strip drilling pose calculation model established above, in triangle O C O W In E1, there is

[0156]

[0157]

[0158] Therefore, point E1 is in the roadway coal wall coordinate system O W X W Y W Z W The coordinates below are

[0159]

[0160] Point E1 is in the camera coordinate system O C X C Y C Z C The coordinates below are

[0161]

[0162] In triangle O C O W In E2, there is

[0163]

[0164]

[0165] Therefore, point E2 is in the roadway coal wall coordinate system O W X W Y W Z W The coordinates below are

[0166]

[0167] Point E2 is in the camera coordinate system O. C X C Y C Z C The coordinates below are

[0168]

[0169] like Figure 5 As shown, in triangle O C O W In F1, there are

[0170]

[0171]

[0172] Therefore, point F1 is in the roadway coal wall coordinate system O W X W Y W Z W The coordinates below are

[0173]

[0174] Point F1 is in camera coordinate system O C X C Y C Z C The coordinates below are

[0175]

[0176] In summary, from the three sets of corresponding coordinates of E1, E2, and F1, we can obtain...

[0177]

[0178] Therefore, the first rotation matrix R can be obtained. WC for:

[0179]

[0180] Among them, R WC Denotes the first rotation matrix. This represents the X-coordinate of point E1 in the roadway coal wall coordinate system. Point E1 represents the mapping point of any intermediate point in the spot data set in the roadway coal wall coordinate system. This represents the X-coordinate of E2 in the roadway coal wall coordinate system. Point E2 represents the mapping point of another intermediate point in the spot data set in the roadway coal wall coordinate system. This represents the Y-coordinate of point F1 in the roadway coal wall coordinate system. Point F1 represents the mapping point of another intermediate point in the spot data set in the roadway coal wall coordinate system. This represents the Z-coordinate of point E1 in the camera coordinate system. This represents the Z-coordinate of point E2 in the camera coordinate system. This represents the Z coordinate of point F1 in the camera coordinate system, and 'a' represents half the distance between the center points of two adjacent light spots after sorting.

[0181] In the pose calculation model mentioned in the monocular vision-based steel strip drilling positioning method of this invention, 'a' is also used to limit the X coordinates of E1 and E2 in the camera coordinate system and the Y coordinate of F1 in the camera coordinate system. Furthermore, points E1 and E2 are different intermediate points from point F1.

[0182] 3) Calculate the second translation vector based on the data set of the center of the steel strip borehole; the second translation vector is the translation vector from the coordinate system of the steel strip borehole to the coordinate system of the roadway coal wall.

[0183] Read the coordinates of the center data of the steel strip drill hole in the image coordinate system stored in vectorA. like Figure 6 As shown, in In the steel strip drilling coordinate system O H X H Y H Z H The origin is in the coordinate system O of the coal face in the roadway. W X W Y W Z W The X-coordinate below is:

[0184]

[0185] in:

[0186]

[0187]

[0188] From O H To X W O W Y W Draw a perpendicular line from the coordinate plane, and then from the foot of the perpendicular to the Y-axis.W Draw a perpendicular line with the foot of the perpendicular at H1. In the middle, there is a steel strip drilling coordinate system O H X H Y H Z H The origin is in the coordinate system O of the coal face in the roadway. W X W Y W Z W The Y-coordinate below is:

[0189]

[0190] in:

[0191]

[0192]

[0193] Therefore, the steel strip drilling coordinate system O H X H Y H Z H Relative to the rock wall coordinate system O W X W Y W Z W The second translation vector is:

[0194]

[0195] 4) Based on the first translation vector, the first rotation matrix, and the second translation vector, calculate the pose of the hole in the steel strip in the camera coordinate system. Specifically,

[0196] According to the formula

[0197] T HC =T HW ·T WC

[0198] Calculate the pose of the drilled hole on the steel strip in the camera coordinate system.

[0199] Among them, T HC T represents the pose of the hole drilled on the steel strip in the camera coordinate system. WC T represents the pose of the coal face coordinate system in the camera coordinate system. HW This indicates the pose of the borehole on the steel strip in the coal wall coordinate system, and R WC Let t represent the first rotation matrix. WC Let t represent the first translation vector. HW Let R represent the second translation vector. HW Let represent the second rotation matrix, and

[0200] Step 700: Based on the pose of the hole in the steel strip in the camera coordinate system, control the drill arm to move to the drilling position in the steel strip. Specifically, based on the pose transformation matrix from the camera to the drilling hole in the steel strip, the motion of each joint on the drill arm of the drilling and anchoring robot is solved, and then the drill arm is controlled to move to the drilling position in the steel strip to complete the drilling and anchoring operation.

[0201] This invention uses the coal face of the tunnel as a bridge. First, the positional relationship between the coal face coordinate system and the camera coordinate system is determined by the red laser spot image projected onto the coal face. Then, the pose of the steel strip borehole in the camera coordinate system is accurately determined by the relative position of the red laser spot in the image and the center of the borehole. Subsequently, the drill arm is controlled to move to the position of the steel strip borehole to complete the drilling and anchoring operation. The whole process promotes the automation and intelligence of coal mining.

[0202] Example 2

[0203] like Figure 7 As shown, this embodiment provides a steel strip drilling positioning system based on monocular vision, including:

[0204] Image acquisition module 101 is used to acquire an image to be processed; the image to be processed includes a roadway coal wall, a steel strip, a line laser beam and multiple light spots; the steel strip is placed on the roadway coal wall and has multiple drill holes.

[0205] The steel strip feature extraction module 201 is used to extract line laser image features from the image to be processed, and process the line laser image features to obtain a steel strip feature map.

[0206] The steel strip data calculation module 301 is used to perform image enhancement processing on the image to be processed based on the steel strip feature map, and to determine the steel strip borehole center data group based on the image to be processed after image enhancement processing.

[0207] The light spot feature extraction module 401 is used to extract light spot image features from the image to be processed and to preprocess the light spot image features to obtain a light spot feature map.

[0208] The spot data calculation module 501 is used to extract the spot center of the spot feature map by ellipse fitting to obtain a spot data set.

[0209] The drilling pose calculation module 601 is used to calculate the pose of the drill hole on the steel strip in the camera coordinate system based on the drilling center data group of the steel strip and the spot data group.

[0210] The drill arm motion control module 701 is used to control the drill arm to move to the drilling position on the steel strip according to the pose of the hole in the camera coordinate system. Specifically, it controls the movement of the drill arm to the drilling position on the steel strip according to the pose transformation matrix T from the camera to the drilling position on the steel strip. HC The motion of each joint on the drill arm of the drilling and anchoring robot is calculated by reverse engineering, and then the drill arm is controlled to move to the drilling position of the steel strip to complete the drilling and anchoring operation.

[0211] Example 3

[0212] like Figure 8 As shown, this embodiment provides a steel strip drilling positioning device based on monocular vision. The steel strip drilling positioning device is installed on the drilling and anchoring robot 4. The steel strip drilling positioning device includes a circulating line laser emitting component 5, a point laser emitting component, a monocular image acquisition component, and a data processing and motion control component 6.

[0213] The circulating linear laser emitting component 5 is positioned above the drilling and anchoring robot 4. The circulating linear laser emitting component 5 is used to circulate and project linear laser beams 7 onto the roadway coal wall and the steel strip 8. The steel strip 8 is placed on the roadway coal wall and has multiple drill holes 9.

[0214] Both the point laser emitting component and the monocular image acquisition component are mounted on the drill arm of the drilling and anchoring robot 4. In one specific embodiment, the point laser emitting component and the monocular image acquisition component are integrated into a laser emitting and acquisition component 2. The point laser emitting component is used to project multiple light spots 1 onto the roadway coal wall and the steel strip 8. The monocular image acquisition component is used to acquire an image to be processed. The image to be processed includes the roadway coal wall, the steel strip 8, the line laser beam 7, and the multiple light spots 1.

[0215] The data processing and motion control component 6 is connected to the monocular image acquisition component, and the data processing program in the data processing and motion control component 6 is used to execute the steel strip drilling positioning method based on monocular vision as described in Embodiment 1.

[0216] Specifically, such as Figure 9As shown, the circulating line laser emitting assembly 5 includes a line laser emitter 55, a double-sided reflector 56, a drive motor 54, and an explosion-proof housing for the line laser emitting module. The explosion-proof housing for the line laser emitting module includes an upper explosion-proof housing 51, explosion-proof glass 52, and a lower explosion-proof housing 53. The explosion-proof glass 52 is embedded in the upper part of the upper explosion-proof housing 51, and the upper explosion-proof housing 51 and the lower explosion-proof housing 53 are fixedly connected by bolts. Two line laser emitters 55 are installed facing each other on both sides inside the lower explosion-proof housing 53, emitting a green line laser beam 7 towards the double-sided reflector 56. The double-sided reflector 56 is mounted on the upper side inside the lower explosion-proof housing 53 via bearings, and is at the same height as the line laser emitters 55. The drive motor 54 is connected to the shaft of the double-sided reflector 56 via a coupling, driving the double-sided reflector 56 to rotate around the shaft. Each rotation of the double-sided reflector 56 reflects the line laser beam 7 and scans the tunnel four times.

[0217] like Figure 10 As shown, the monocular image acquisition component comprises three parts: a camera explosion-proof housing, a high-definition industrial camera 24, and a switching power supply 25. The camera explosion-proof housing includes a camera explosion-proof glass 21, a front camera explosion-proof housing 22, a rear camera explosion-proof housing 27, and a point laser emitter explosion-proof glass 23. The camera explosion-proof glass 21 is embedded in the front end of the front camera explosion-proof housing 22, and the point laser emitter explosion-proof glass 23 is embedded in the rear end connector of the front camera explosion-proof housing 22. The rear camera explosion-proof housing 27 is provided with a camera mounting cantilever beam and a point laser emitter mounting slot. The front camera explosion-proof housing 22 and the rear camera explosion-proof housing 27 are fastened together by bolts on the connector.

[0218] The high-definition industrial camera 24 and the switching power supply 25 are respectively fixedly installed on the front and rear ends of the camera mounting cantilever beam on the explosion-proof housing at the rear of the camera. The high-definition industrial camera 24 mainly completes the acquisition of the image to be processed, and the switching power supply 25 provides a stable power supply for the high-definition industrial camera 24 and the point laser emission assembly.

[0219] Furthermore, the point laser emission assembly includes four point laser emitters 26, which are fixedly mounted in point laser emitter slots provided on the explosion-proof housing 27 on the rear side of the camera. For example... Figure 10 As shown, four point laser emitters 26 are evenly and symmetrically distributed at the four corners of the high-definition industrial camera, and the distance between any two adjacent point laser emitters is 2a.

[0220] Preferably, such as Figure 8As shown, the laser emission and acquisition component 2 is fixedly connected to the front end of the drill arm 3 of the drilling and anchoring robot 4 via bolts at the bottom of the explosion-proof shell 22 on the front side of the camera. The point laser emission component projects a red point laser spot 1 onto the coal wall of the tunnel, which is used to determine the relative position of the camera and the coal wall. The circulating line laser emission component 5 is mounted upwards on the body of the drilling and anchoring robot 4, rapidly and cyclically projecting a green line laser beam 7 onto the coal wall and the steel strip on the coal wall. The data processing and motion control component 6 is mounted on the tail platform of the drilling and anchoring robot 4 to complete the drilling positioning and motion control tasks.

[0221] In one specific embodiment, the working process of the steel strip drilling positioning device based on monocular vision is as follows:

[0222] The point laser emitter 26 in the point laser emission assembly projects a red point laser beam to form a red laser spot on the roadway. This red laser spot helps determine the relative position of the camera and the coal wall. The line laser emitter 55 in the circulating line laser emission assembly cyclically emits a green line laser beam towards the coal wall of the roadway. The monocular image acquisition assembly acquires roadway images containing the steel strip, red point laser, and green line laser beam using a soft-trigger method. Then, the data processing program in the data processing and motion control unit sequentially performs image processing, image enhancement, and borehole positioning as described in Embodiment 1, outputting the pose of the steel strip borehole in the camera coordinate system. Finally, the motion control program in the data processing and motion control unit deduces the motion of each joint on the drill arm of the drilling and anchoring robot based on the pose of the steel strip borehole in the camera coordinate system, thereby controlling the drill arm to move to the position of the steel strip borehole and completing the drilling and anchoring operation.

[0223] Compared with the prior art, the present invention also has the following advantages:

[0224] This invention proposes a method, system, and device for locating steel strip drill holes based on monocular vision, addressing the low-light and weak-texture working environment of underground coal mines. This enables faster and more accurate identification of steel strips and drill holes. Compared with existing steel strip drill hole location schemes based on binocular vision, this invention uses monocular vision to locate steel strip drill holes, greatly increasing the real-time performance of the location while also achieving higher positioning accuracy.

[0225] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0226] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for positioning drilled steel strips based on monocular vision, characterized in that, The monocular vision-based steel strip drilling positioning method includes: Acquire an image to be processed; the image to be processed includes a roadway coal wall, a steel strip, a line laser beam, and multiple light spots; the steel strip is placed on the roadway coal wall, and multiple drill holes are provided on the steel strip; Line laser image features are extracted from the image to be processed, and the line laser image features are processed to obtain a steel strip feature map; The image to be processed is enhanced based on the steel strip feature map, and the steel strip borehole center data group is determined based on the enhanced image to be processed. Extract spot image features from the image to be processed, and preprocess the spot image features to obtain a spot feature map; Ellipse fitting is used to extract the center of the light spot feature map to obtain the light spot data set; Based on the data set of the drilling center of the steel strip and the data set of the light spot, calculate the pose of the drilling coordinate system on the steel strip in the camera coordinate system; the pose is a transformation matrix. Based on the pose of the drilling coordinate system on the steel strip in the camera coordinate system, the drill arm is controlled to move to the drilling position on the steel strip.

2. The method for positioning steel strip drilling based on monocular vision according to claim 1, characterized in that, The image to be processed is enhanced based on the steel strip feature map, and the steel strip borehole center data group is determined based on the enhanced image, specifically including: Based on the steel strip feature map, the image to be processed is subjected to image enhancement processing based on the Laplacian operator to determine the preliminary enhanced image; Thresholding is performed on the preliminary enhanced image to determine the outline image of the steel strip drill hole; Ellipse fitting is performed on the outline image of the steel strip drill hole to determine the pixel coordinates of the center of the steel strip drill hole in the pixel coordinate system. The pixel coordinates of the center of the steel strip drilling in the pixel coordinate system are transformed to determine the image coordinates of the center of the steel strip drilling in the corresponding image coordinate system; the image coordinates of the center of the steel strip drilling in the corresponding image coordinate system constitute the data group of the center of the steel strip drilling.

3. The method for positioning steel strip drilling based on monocular vision according to claim 1, characterized in that, The light spot image features are preprocessed to obtain a light spot feature map, specifically including: The light spot image features are sequentially processed by pyramid downsampling, binarization segmentation, Canny edge detection, and pyramid upsampling to obtain a light spot feature map. The step of extracting the center of the light spot from the feature map using ellipse fitting to obtain a light spot data set specifically includes: The center of the light spot in the feature map is extracted by ellipse fitting based on Hough transform to determine multiple center points of the light spot; The center points of the multiple light spots are arranged in a first order; the first order is clockwise or counterclockwise. For two adjacent light spot center points after sorting, a midpoint is determined between the two light spot center points; the midpoint is located on the line connecting the two adjacent light spot center points; multiple light spot center points and multiple midpoints constitute a light spot data group.

4. The method for positioning steel strip drilling based on monocular vision according to claim 3, characterized in that, Based on the data set of the drilling center on the steel strip and the data set of the light spot, the pose of the drilling coordinate system on the steel strip in the camera coordinate system is calculated, specifically including: Calculate the first translation vector based on the center point of the light spot in the light spot data group; the first translation vector is the translation vector from the camera coordinate system to the roadway coal wall coordinate system; Calculate the first rotation matrix based on the midpoint of the light spot data set; the first rotation matrix is ​​the rotation matrix from the camera coordinate system to the roadway coal wall coordinate system; Calculate the second translation vector based on the data set of the center of the steel strip borehole; the second translation vector is the translation vector from the steel strip borehole coordinate system to the roadway coal wall coordinate system; The pose of the drilling coordinate system on the steel strip in the camera coordinate system is calculated based on the first translation vector, the first rotation matrix, and the second translation vector.

5. The method for positioning steel strip drilling based on monocular vision according to claim 4, characterized in that, Calculating the pose of the drilling coordinate system on the steel strip in the camera coordinate system based on the first translation vector, the first rotation matrix, and the second translation vector, specifically includes: According to the formula Calculate the pose of the drilling coordinate system on the steel strip in the camera coordinate system; in, This indicates the pose of the drilling coordinate system on the steel strip in the camera coordinate system. This indicates the pose of the coal face coordinate system in the roadway coordinate system within the camera coordinate system. This indicates the pose of the steel strip drilling coordinate system within the roadway coal wall coordinate system. , , Denotes the first rotation matrix. Denotes the first translation vector. Denotes the second translation vector. , This represents the second rotation matrix.

6. The method for positioning steel strip drilling based on monocular vision according to claim 4, characterized in that, Based on the center point of the light spot in the light spot data group, the first translation vector is calculated, specifically including: According to the formula Calculate the first translation vector; in, Denotes the first translation vector. , This represents half the distance between the center points of two adjacent light spots after sorting, where the distances between the center points of two adjacent light spots are equal. f This indicates the focal length of the camera used to capture the image to be processed. Point With point The distance between them Point With point The distance between points Let point be the origin of the light spot feature map in the camera coordinate system. This represents the center point of a spot in the aforementioned spot data set. This indicates that the spot data group contains points... The center point of another non-adjacent light spot.

7. The method for positioning steel strip drilling based on monocular vision according to claim 4, characterized in that, Calculate the first rotation matrix based on the midpoint of the light spot data group, specifically including: According to the formula Calculate the first rotation matrix; in, Denotes the first rotation matrix. Point In the coordinate system of the coal wall in the roadway X Coordinates, point This represents the mapping point of any intermediate point in the spot data set in the roadway coal wall coordinate system. Point In the coordinate system of the coal wall in the roadway X Coordinates, point This represents the mapping point of another intermediate point in the spot data set in the roadway coal wall coordinate system. Point In the coordinate system of the coal wall in the roadway Y Coordinates, point This represents the mapping point of another intermediate point in the spot data set in the roadway coal wall coordinate system. Point In camera coordinate system Z coordinate, Point In camera coordinate system Z coordinate, Point In camera coordinate system Z coordinate, This represents half the distance between the center points of two adjacent light spots after sorting, where the distance between the center points of the two adjacent light spots is equal.

8. The method for positioning steel strip drilling based on monocular vision according to claim 4, characterized in that, Based on the aforementioned steel strip borehole center data set, the second translation vector is calculated, specifically including: According to the formula Calculate the second translation vector; in, This indicates that the origin of the steel strip drilling coordinate system is in the roadway coal wall coordinate system. coordinate, This indicates that the origin of the steel strip drilling coordinate system is in the roadway coal wall coordinate system. coordinate, This indicates the thickness of the steel strip.

9. A steel strip drilling positioning system based on monocular vision, characterized in that, The monocular vision-based steel strip drilling positioning system includes: An image acquisition module is used to acquire an image to be processed; the image to be processed includes a roadway coal wall, a steel strip, a line laser beam, and multiple light spots; the steel strip is placed on the roadway coal wall, and multiple drill holes are provided on the steel strip; The steel strip feature extraction module is used to extract line laser image features from the image to be processed, and to process the line laser image features to obtain a steel strip feature map. The steel strip data calculation module is used to perform image enhancement processing on the image to be processed based on the steel strip feature map, and to determine the steel strip borehole center data group based on the image to be processed after image enhancement processing. A spot feature extraction module is used to extract spot image features from the image to be processed and to preprocess the spot image features to obtain a spot feature map. The spot data calculation module is used to extract the spot center of the spot feature map by ellipse fitting to obtain the spot data set; The drilling pose calculation module is used to calculate the pose of the drilling coordinate system on the steel strip in the camera coordinate system based on the drilling center data group of the steel strip and the spot data group; the pose is a transformation matrix. The drill arm motion control module is used to control the drill arm to move to the drilling position on the steel strip according to the pose of the drilling coordinate system on the steel strip in the camera coordinate system.

10. A steel strip drilling positioning device based on monocular vision, wherein the steel strip drilling positioning device is mounted on a drilling and anchoring robot, characterized in that, The steel strip drilling positioning device includes a circulating linear laser emitting component, a point laser emitting component, a monocular image acquisition component, and a data processing and motion control component. The circulating linear laser emitting assembly is positioned above the drilling and anchoring robot. The circulating linear laser emitting assembly is used to circulate and project linear laser beams onto the roadway coal wall and the steel strip. The steel strip is placed on the roadway coal wall and has multiple drill holes. Both the point laser emitting component and the monocular image acquisition component are mounted on the drill arm of the drilling and anchoring robot; the point laser emitting component is used to project multiple light spots onto the roadway coal wall and the steel strip; The monocular image acquisition component is used to acquire an image to be processed; the image to be processed includes the roadway coal wall, the steel strip, the line laser beam, and multiple light spots; The data processing and motion control component is connected to the monocular image acquisition component, and the data processing and motion control component is used to execute the monocular vision-based steel strip drilling positioning method according to any one of claims 1-8.

Images