Automatic cable reel stacking system based on image processing technology

By using Gaussian filtering and bilateral filtering in the automatic stacking system of the online cable tray, combined with data analysis and decision-making modules, the problem of low image noise processing and accuracy in traditional systems is solved, and the grab point and stacking position are realized and improve the accuracy and system efficiency.

CN118701646BActive Publication Date: 2024-12-03SHANGHAI SHIJING HUATU INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410818565.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-24
Publication Date
2024-12-03
Estimated Expiration
2044-06-24

AI Technical Summary

Technical Problem

The traditional cable tray automatic stacking system has shortcomings in data processing and accuracy, poor image noise processing effect, low edge detection accuracy, resulting in large errors in determining the grab point and stacking position, and lack of comprehensive analysis and decision-making modules, which is difficult to Cope with complex operating environments.

Method used

Using a system based on image processing technology, including data collection and processing module, data analysis module, comprehensive analysis module and decision-making module, the image is denoised through Gaussian filtering and bilateral filtering, the contour and shape of the cable tray are extracted, and the best calculation is calculated. Grab the points and stacking positions, and dynamically adjust the image processing algorithm according to the detection results to ensure the accuracy of the grab points and stacking positions.

Benefits of technology

It improves the accuracy and reliability of image data, ensures the accuracy of the working path planning of the robot and the efficient operation, can operate effectively in complex environments, and improves the efficiency and accuracy of automatic stacking of cable trays.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118701646B_ABST
    Figure CN118701646B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of image processing technology, and specifically to an automatic cable reel stacking system based on image processing technology, which includes a data collection and processing module, a data analysis module, a comprehensive analysis module and a decision module, wherein: the data collection and processing module collects cable entanglement images and stacking area images; the data analysis module calculates the outline and shape of the cable entanglement and the size and position of the cable entanglement according to the cable entanglement image; calculates the free position of the stacking area according to the stacking area image; the comprehensive analysis module calculates the best gripping point of the cable reel according to the outline and shape of the cable entanglement; calculates the best stacking position of the stacking area according to the size and position of the cable entanglement and the free position of the stacking area; the decision module detects whether the best gripping point of the cable reel is within the size and position of the cable entanglement, and performs optimization or calculation of control instructions according to the detection result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to an automatic cable reel stacking system based on image processing technology. Background Art

[0002] Traditional cable reel automatic stacking systems have multiple disadvantages during operation, mainly in terms of data processing, accuracy and system optimization capabilities.

[0003] First, in terms of data processing, traditional systems usually use simple image processing technology, which has poor processing effect on image noise. Due to the lack of efficient denoising methods, such as Gaussian filtering and bilateral filtering, the noise in the image cannot be effectively removed, affecting the accuracy of subsequent data analysis.

[0004] Secondly, in terms of accuracy, traditional systems often rely on basic edge detection and contour extraction algorithms when extracting the contour and shape of cable reels, and the calculated size and position accuracy is low. This low-precision data leads to large errors in the determination of the grasping point and stacking position, and the operation path planning of the robot is therefore not accurate enough, which can easily lead to grasping failure or unstable stacking.

[0005] In addition, the traditional system lacks comprehensive analysis and decision-making modules, and the selection of grabbing points and stacking positions is too simple, and it cannot be dynamically adjusted and optimized according to actual conditions. The system cannot adjust the image processing algorithm according to the detection results, nor can it ensure the accuracy of the grabbing points and stacking positions; the various modules of the traditional system lack effective coordination, the overall operation efficiency is low, and it is difficult to cope with complex operating environments. Summary of the invention

[0006] The object of the present invention is to provide an automatic cable reel stacking system based on image processing technology to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: an automatic stacking system for cable reels based on image processing technology, characterized in that it includes a data collection and processing module, a data analysis module, a comprehensive analysis module and a decision-making module, wherein:

[0008] The data collection and processing module collects the cable entanglement image and the stacking area image, and uses Gaussian filtering to perform denoising on the cable entanglement image and the stacking area image;

[0009] The data analysis module calculates the target detection result according to the cable entanglement image, wherein the target detection result includes the outline and shape of the cable entanglement and the size and position of the cable entanglement; and calculates the free position of the stacking area according to the stacking area image;

[0010] The comprehensive analysis module calculates the best gripping point of the cable reel according to the contour and shape of the cable reel; calculates the best stacking position of the stacking area according to the size and position of the cable reel and the free position of the stacking area;

[0011] The decision module detects whether the optimal gripping point of the cable reel is within the size and position of the cable reel, wherein:

[0012] If the optimal gripping point of the cable reel is not within the size and position of the cable reel, the Gaussian filter in the data collection and processing module is modified to a bilateral filter to perform optimization processing;

[0013] If the optimal gripping point of the cable reel is within the size and position of the cable reel, the control instructions of the manipulator are calculated based on the optimal gripping point of the cable reel and the optimal stacking position of the stacking area.

[0014] As a further improvement of the present technical solution, the data collection and processing module uses Gaussian filtering to reduce the noise of the image by weighted averaging of neighborhood pixels, uses a Gaussian function as a weight distribution, traverses each pixel in the image, and replaces the value of the current pixel with the weighted sum of the neighborhood pixels.

[0015] As a further improvement of the technical solution, the data analysis module includes a cable entanglement image analysis unit, and the cable entanglement image analysis unit calculates the outline and shape of the cable entanglement and the size and position of the cable entanglement according to the cable entanglement image. The specific calculation process includes:

[0016] Converting the cable reel image into a grayscale cable reel image by calculating the grayscale value of each pixel of the image;

[0017] The edge of the grayscale cable reel image is extracted by edge detection to generate a cable reel edge image;

[0018] Use contour detection algorithm to extract all contours in the cable reel edge image as the cable reel contour;

[0019] According to the outline of the cable, calculate the circumscribed rectangle, which is represented by the coordinates of the upper left corner and the width, height and depth; use the calibration parameters of the camera to convert the pixel unit into the actual unit, and convert the width, height and depth in the circumscribed rectangle into the actual width, actual height and actual depth, and use the actual width, actual height and actual depth as the size of the cable;

[0020] Calculate the shape characteristics of the cable coil according to the size of the cable coil;

[0021] Calculate the center position of the cable reel as the position of the cable reel. The calculation formula for the center position is: center position X = upper left corner X coordinate + half of the width; center position Y = upper left corner Y coordinate + half of the height; center position Z = upper left corner Z coordinate + half of the depth.

[0022] As a further improvement of the technical solution, the data analysis module includes a stacking area image analysis unit, and the stacking area image analysis unit calculates the free position of the stacking area according to the stacking area image, specifically including:

[0023] The stacking area image is converted into a grayscale image, edges in the image are extracted using edge detection, and all contours are extracted from the image after edge detection as the stacking area contour;

[0024] Marking the stacking area outline on the stacking area image to generate an occupied area image;

[0025] The occupied area image is subjected to morphological dilation operation to expand the boundary of the occupied area, and the idle area image is obtained by image subtraction. Contour detection is performed again on the idle area image to extract the contour of the idle area; and the center position of each idle area is calculated as the idle position of the stacking area.

[0026] As a further improvement of the technical solution, the comprehensive analysis module includes a grab point analysis unit, and the grab point analysis unit calculates the optimal grab point of the cable reel, specifically including:

[0027] Extract key points from the cable contour by using a scale-invariant feature transformation algorithm;

[0028] According to the shape characteristics of the cable reel, determine its main geometric characteristics;

[0029] Calculate the center of gravity of the cable reel by obtaining the coordinates of all points on the outline of the cable reel, where the center of gravity X is the sum of the x coordinates of all points divided by the number of x coordinates, the center of gravity Y is the sum of the y coordinates of all points divided by the number of y coordinates, and the center of gravity Z = the sum of the z coordinates of all points / the number of z coordinates;

[0030] Divide the outline of the cable into several small areas, calculate the area and center of gravity of each area, analyze the area distribution and center of gravity position of each small area, select the area with uniform mass distribution, and ensure the uniform mass distribution in the area where the grab point is located by comparing the area and center of gravity of each area;

[0031] Select points close to the edge as candidate grasping points, and points on the contour are edge points;

[0032] The moment analysis is used to select the candidate grasping point with the minimum moment. The moment formula is moment = force × lever arm length, where the lever arm length is the distance from the candidate grasping point to the center of gravity, and force is the force applied during grasping;

[0033] The point with the smallest torque is selected, and the distance from each candidate grasping point to the center of gravity is calculated, and the candidate grasping point with the smallest torque is selected as the optimal grasping point of the cable reel.

[0034] As a further improvement of the technical solution, the comprehensive analysis module includes a stacking position analysis unit, which calculates the optimal stacking position of the stacking area according to the size and position of the cables and the free position of the stacking area, specifically including:

[0035] According to the free positions in the stacking area, the coordinates and sizes of all free positions in the stacking area are obtained;

[0036] Filter out suitable free positions according to the size of the cable reel, and filter out suitable free positions by filtering out cable reels whose size is smaller than the size of the free positions in the stacking area;

[0037] Find the area with the best matching cable reel size in the free positions in the stacking area, calculate the difference between the free positions in the stacking area and the cable reel size, and select the free position in the stacking area with the smallest size difference as the best stacking position in the stacking area.

[0038] As a further improvement of the technical solution, the decision module includes a detection unit, which detects whether the optimal gripping point of the cable reel is within the size and position of the cable reel. The detection process specifically includes:

[0039] Calculate the minimum and maximum coordinates of the cable reel according to the size and position of the cable reel, where: the minimum x coordinate is x_min=x; the maximum x coordinate is x_max=x+l; the minimum y coordinate is y_min=y; the maximum y coordinate is y_max=y+w; the minimum z coordinate is z_min=z; the maximum z coordinate is z_max=z+h; where (x, y, z) is the position coordinate of the cable reel, and (l, w, h) is the size coordinate of the cable reel;

[0040] The coordinates of the best gripping point of the cable reel are (x_g, y_g, z_g). By detecting the x coordinate, we can determine whether x_g is between x_min and x_max; by detecting the y coordinate, we can determine whether y_g is between y_min and y_max; by detecting the z coordinate, we can determine whether z_g is between z_min and z_max; where:

[0041] If the x, y, and z coordinates are all within the corresponding bounds, then the optimal gripping point for the cable reel is within the size and position of the cable reel;

[0042] If any coordinate is not within the corresponding boundary range, then the optimal gripping point of the cable reel is not within the size and position of the cable reel.

[0043] As a further improvement of the present technical solution, the decision module includes an optimization control unit, and the optimization control unit makes corresponding optimization and control according to the detection result of the detection unit, specifically including:

[0044] If the best gripping point of the cable reel is not within the size and position of the cable reel, the Gaussian filter in the data collection and processing module is modified to a bilateral filter, the spatial weight is based on the weight distribution of the spatial distance, and the value weight is based on the weight distribution of the pixel value difference. The closer the pixel value is, the greater the weight is, and vice versa. The product of the spatial weight and the value weight is used for normalized weighted averaging to obtain the filtered pixel value;

[0045] If the optimal gripping point of the cable reel is within the size and position of the cable reel, the control instructions of the manipulator are calculated based on the optimal gripping point of the cable reel and the optimal stacking position of the stacking area.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The automatic stacking system of cable reels based on image processing technology obtains the size and position of the cable reel according to the calculation of the data analysis module, and the comprehensive analysis module calculates the best gripping point of the cable reel and the best stacking position of the stacking area; and detects the best gripping point of the cable reel according to the size and position of the cable reel. Specifically, the process first uses image processing technology to obtain the image data of the cable reel, and performs denoising through Gaussian filtering to ensure the accuracy and reliability of the data; then, the data analysis module extracts the contour and shape of the cable reel according to the processed image data, and calculates its size and position. This information is crucial for determining the best gripping point and stacking position; the comprehensive analysis module further analyzes this data, calculates the best gripping point through a detailed analysis of the contour and shape of the cable reel, and determines the best stacking position in combination with the free position in the stacking area; in this way, the system can effectively plan the working path of the robot to ensure the efficiency and accuracy of the gripping and stacking process.

[0048] 2. The automatic stacking system for cable reels based on image processing technology, according to the detection results of the decision-making module, if the optimal gripping point of the cable reel is not within the size and position of the cable reel, the Gaussian filter in the data collection and processing module is modified to a bilateral filter for optimization; and if the optimal gripping point of the cable reel is within the size and position of the cable reel, the control instructions of the manipulator are calculated according to the optimal gripping point of the cable reel and the optimal stacking position of the stacking area; through this dual detection and optimization mechanism, the system can dynamically adjust the image processing algorithm to improve the accuracy of gripping point detection. When it is detected that the gripping point is not in the appropriate position, the system automatically switches to bilateral filtering, which not only considers the spatial distance, but also the pixel value difference, thereby effectively removing noise while retaining image details. If the gripping point is in the appropriate position, the system directly generates the control instructions of the manipulator based on the existing data to ensure the accuracy and real-time performance of the operation.

[0049] 3. The automatic stacking system of cable reels based on image processing technology This system realizes the automatic stacking of cable reels through multiple modules and complex algorithm processes; the data collection and processing module performs image denoising through Gaussian filtering and bilateral filtering to ensure data quality; the data analysis module uses advanced image processing algorithms, such as edge detection and contour extraction, to accurately calculate the size and position of cable reels. The comprehensive analysis module determines the best grasping point by extracting key points and calculating the center of gravity, and calculates the best stacking position based on the free position in the stacking area; the decision-making module ensures the accuracy of the grasping point and stacking position through detection and optimization control. The whole process ensures the efficient operation of the system in complex environments and greatly improves the efficiency and accuracy of automatic stacking of cable reels. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the overall module of the present invention;

[0051] Figure 2 It is a schematic diagram of a data analysis module unit of the present invention;

[0052] Figure 3 It is a schematic diagram of a comprehensive analysis module unit of the present invention;

[0053] Figure 4 It is a schematic diagram of the decision module unit of the present invention.

[0054] In the figure: 100, data collection and processing module; 200, data analysis module; 201, cable entanglement image analysis unit; 202, stacking area image analysis unit; 300, comprehensive analysis module; 301, grabbing point analysis unit; 302, stacking position analysis unit; 400, decision module; 401, detection unit; 402, optimization control unit. DETAILED DESCRIPTION

[0055] The following will be combined with the accompanying drawings in the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] For examples, see Figure 1-4 The present invention provides a technical solution: an automatic cable reel stacking system based on image processing technology, including a data collection and processing module 100, a data analysis module 200, a comprehensive analysis module 300 and a decision module 400.

[0057] The data collection and processing module 100 collects cable entanglement images and stacking area images, uses a camera to shoot the cable reel to obtain the cable reel image, shoots the stacking area to obtain the stacking area image, and uses Gaussian filtering to denoise the cable entanglement image and the stacking area image, wherein Gaussian filtering is a linear smoothing filter that reduces image noise by weighted averaging of neighboring pixels, and uses Gaussian function as weight distribution, with the center pixel having the largest weight and the farther from the center, the smaller the weight, and traverses each pixel in the image, replacing the value of the current pixel with the weighted sum of the neighboring pixels.

[0058] The cable entanglement image analysis unit 201 in the data analysis module 200 calculates the target detection result according to the cable entanglement image, wherein the target detection result includes the outline and shape of the cable entanglement and the size and position of the cable entanglement. The specific calculation process includes:

[0059] Converting the cable reel image into a grayscale cable reel image by calculating the grayscale value of each pixel of the image;

[0060] The edge of the grayscale cable reel image is extracted by edge detection to generate a cable reel edge image;

[0061] Use contour detection algorithm to extract all contours in the cable reel edge image as the cable reel contour;

[0062] According to the outline of the cable, calculate the circumscribed rectangle, which is represented by the coordinates of the upper left corner and the width, height and depth; use the calibration parameters of the camera to convert the pixel unit into the actual unit, and convert the width, height and depth in the circumscribed rectangle into the actual width, actual height and actual depth, and use the actual width, actual height and actual depth as the size of the cable;

[0063] According to the size of the cable coil, calculate the shape characteristics of the cable coil, such as aspect ratio, circularity, perimeter, and area;

[0064] Calculate the center position of the cable reel as the position of the cable reel. The calculation formula for the center position is: center position X = upper left corner X coordinate + half of the width; center position Y = upper left corner Y coordinate + half of the height; center position Z = upper left corner Z coordinate + half of the depth.

[0065] The stacking area image analysis unit 202 in the data analysis module 200 calculates the free position of the stacking area according to the stacking area image, specifically including:

[0066] Similarly, the stacking area image is converted into a grayscale image, the edges in the image are extracted using edge detection, and all contours are extracted from the image after edge detection as the stacking area contour;

[0067] Marking the stacking area outline on the stacking area image to generate an occupied area image;

[0068] The occupied area image is subjected to morphological dilation operation to expand the boundary of the occupied area, and the idle area image is obtained by image subtraction. Contour detection is performed again on the idle area image to extract the contour of the idle area; and the center position of each idle area is calculated as the idle position of the stacking area.

[0069] The grabbing point analysis unit 301 in the comprehensive analysis module 300 calculates the optimal grabbing point of the cable reel according to the contour and shape of the cable reel, specifically including:

[0070] Extract key points from the cable's coiled contours by using a scale-invariant feature transformation algorithm;

[0071] According to the shape characteristics of the cable reel, determine its main geometric characteristics, such as aspect ratio, circularity, circumference, area, etc., for subsequent calculation of the best grabbing point;

[0072] Calculate the center of gravity of the cable reel by obtaining the coordinates of all points on the outline of the cable reel, where the center of gravity X is the sum of the x coordinates of all points divided by the number of x coordinates, the center of gravity Y is the sum of the y coordinates of all points divided by the number of y coordinates, and the center of gravity Z = the sum of the z coordinates of all points / the number of z coordinates; the center of gravity represents the center of balance of the object in space, and grasping the center of gravity can ensure that the object remains balanced during the grasping process;

[0073] Balance analysis ensures that the mass distribution of the area selected by the grasping point is uniform, avoiding tilting or imbalance of the object during grasping. The outline of the cable is divided into several small areas, the area and center of gravity of each area are calculated, the area distribution and center of gravity position of each small area are analyzed, and the area with uniform mass distribution is selected. By comparing the area and center of gravity of each area, the mass distribution of the area where the grasping point is located is ensured to be uniform.

[0074] In order to facilitate the grasping of the robot, points close to the edge are selected as candidate grasping points, and points on the contour are edge points;

[0075] The moment analysis is used to select the candidate grasping point with the minimum moment to ensure the grasping stability. The moment formula is moment = force × lever arm length, where the lever arm length is the distance from the candidate grasping point to the center of gravity, and force is the force applied during grasping;

[0076] Select the point with the smallest torque by calculating the distance from each candidate grasping point to the center of gravity (arm length). Because the smaller the torque, the more stable the grasping, the candidate grasping point with the smallest torque is selected as the best grasping point for the cable reel.

[0077] The stacking position analysis unit 302 in the comprehensive analysis module 300 calculates the optimal stacking position of the stacking area according to the size and position of the cables and the free position of the stacking area, specifically including:

[0078] According to the free positions in the stacking area, the coordinates and sizes of all free positions in the stacking area are obtained;

[0079] Filter out suitable free spaces according to the size of the cable reel, and filter out suitable free spaces by filtering out cable reels whose size is smaller than the size of the free spaces in the stacking area, to ensure that the size of the free spaces in the stacking area can accommodate the cable reel;

[0080] Find the area with the best matching cable reel size in the free positions in the stacking area, calculate the difference between the free positions in the stacking area and the cable reel size, and preferentially select the free positions in the stacking area with the smallest size difference as the best stacking positions in the stacking area.

[0081] The detection unit 401 in the decision module 400 detects whether the optimal gripping point of the cable reel is within the size and position of the cable reel. The detection process specifically includes:

[0082] Calculate the minimum and maximum coordinates of the cable reel according to the size and position of the cable reel, where: the minimum x coordinate is x_min=x; the maximum x coordinate is x_max=x+l; the minimum y coordinate is y_min=y; the maximum y coordinate is y_max=y+w; the minimum z coordinate is z_min=z; the maximum z coordinate is z_max=z+h; where (x, y, z) is the position coordinate of the cable reel, and (l, w, h) is the size coordinate of the cable reel;

[0083] The coordinates of the best gripping point of the cable reel are (x_g, y_g, z_g). By detecting the x coordinate, we can determine whether x_g is between x_min and x_max; by detecting the y coordinate, we can determine whether y_g is between y_min and y_max; by detecting the z coordinate, we can determine whether z_g is between z_min and z_max; where:

[0084] If the x, y, and z coordinates are all within the corresponding bounds, then the optimal gripping point for the cable reel is within the size and position of the cable reel;

[0085] If any coordinate is not within the corresponding boundary range, then the optimal gripping point of the cable reel is not within the size and position of the cable reel.

[0086] The optimization control unit 402 in the decision module 400 performs corresponding optimization and control according to the detection result of the detection unit 401, specifically including:

[0087] If the best grabbing point of the cable reel is not within the size and position of the cable reel, the Gaussian filter in the data collection and processing module 100 is modified to a bilateral filter for optimization. The bilateral filter takes into account the spatial distance and the pixel value difference, and smoothes the image. The spatial weight is based on the weight distribution of the spatial distance, and the value weight is based on the weight distribution of the pixel value difference. The closer the pixel value is, the greater the weight, and vice versa. The product of the spatial weight and the value weight is used for normalized weighted averaging to obtain the filtered pixel value. Since Gaussian filtering denoising will blur the edges and affect the accuracy of subsequent edge detection and contour extraction, the bilateral filter retains the edge details, optimizes the results of edge detection and contour extraction, and thus improves the accuracy of data analysis.

[0088] If the optimal gripping point of the cable reel is within the size and position of the cable reel, the control instructions of the manipulator are calculated according to the optimal gripping point of the cable reel and the optimal stacking position of the stacking area, including:

[0089] Obtaining the grabbing point and the stacking point, wherein the grabbing point is the three-dimensional coordinate of the best grabbing point of the cable reel, and the stacking point is the three-dimensional coordinate of the best stacking position in the stacking area;

[0090] Perform straight line path interpolation calculation from the grabbing point to the stacking point to generate a series of paths passing through the intermediate points; evenly distribute the intermediate points according to the distance on the path. For example, assuming that the straight line distance is the total distance, interpolation generates several intermediate points;

[0091] Calculate the displacement in each direction based on the coordinate difference between the grasping point and the stacking point; ensure that the end effector of the manipulator is aligned with the target when grasping by calculating the required rotation angle based on the initial posture and end position of the manipulator; calculate the execution speed and acceleration of the manipulator considering the task requirements and the safety and speed requirements determined in the path planning;

[0092] Generate movement instructions for each intermediate point coordinate sequence, so that the robot arm can move step by step to the grasping point according to the calculated path and perform the grasping action; after reaching the grasping point, generate grasping action instructions to control the opening and closing of the end effector; after successful grasping, move along the path to the optimal stacking position and generate the same movement instruction sequence; after reaching the stacking position, generate placement action instructions to make the end effector withdraw from the target object.

[0093] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. The automatic stacking system of cable reels based on image processing technology is characterized by: The system comprises a data collection and processing module (100), a data analysis module (200), a comprehensive analysis module (300) and a decision-making module (400), wherein: The data collection and processing module (100) collects cable entanglement images and stacking area images, and uses Gaussian filtering to perform denoising on the cable entanglement images and stacking area images; The data collection and processing module (100) uses Gaussian filtering to reduce image noise by weighted averaging of neighboring pixels, uses a Gaussian function as a weight distribution, traverses each pixel in the image, and replaces the value of the current pixel with the weighted sum of the neighboring pixels; The data analysis module (200) calculates target detection results based on the cable entanglement image, wherein the target detection results include the outline and shape of the cable entanglement and the size and position of the cable entanglement; and calculates free positions in the stacking area based on the stacking area image; The data analysis module (200) comprises a cable entanglement image analysis unit (201), wherein the cable entanglement image analysis unit (201) calculates the outline and shape of the cable entanglement as well as the size and position of the cable entanglement according to the cable entanglement image, and the specific calculation process comprises: Converting the cable reel image into a grayscale cable reel image by calculating the grayscale value of each pixel of the image; The edge of the grayscale cable reel image is extracted by edge detection to generate a cable reel edge image; Use contour detection algorithm to extract all contours in the cable reel edge image as the cable reel contour; According to the outline of the cable, calculate the circumscribed rectangle, which is represented by the coordinates of the upper left corner and the width, height and depth; use the calibration parameters of the camera to convert the pixel unit into the actual unit, and convert the width, height and depth in the circumscribed rectangle into the actual width, actual height and actual depth, and use the actual width, actual height and actual depth as the size of the cable; Calculate the shape characteristics of the cable coil according to the size of the cable coil; Calculate the center position of the cable reel as the position of the cable reel. The calculation formula for the center position is: center position X = upper left corner X coordinate + half of the width; center position Y = upper left corner Y coordinate + half of the height; center position Z = upper left corner Z coordinate + half of the depth; The data analysis module (200) comprises a stacking area image analysis unit (202), wherein the stacking area image analysis unit (202) calculates the free position of the stacking area according to the stacking area image, specifically comprising: The stacking area image is converted into a grayscale image, edges in the image are extracted using edge detection, and all contours are extracted from the image after edge detection as the stacking area contour; Marking the stacking area outline on the stacking area image to generate an occupied area image; Perform morphological dilation on the occupied area image to expand the boundary of the occupied area, obtain the idle area image by image subtraction, perform contour detection on the idle area image again to extract the outline of the idle area; and calculate the center position of each idle area as the idle position of the stacking area; The comprehensive analysis module (300) calculates the best gripping point of the cable reel according to the contour and shape of the cable reel; and calculates the best stacking position of the stacking area according to the size and position of the cable reel and the free position of the stacking area; The decision module (400) detects whether the optimal gripping point of the cable reel is within the size and position of the cable reel, wherein: If the optimal gripping point of the cable reel is not within the size and position of the cable reel, the Gaussian filter in the data collection and processing module (100) is modified to a bilateral filter to perform optimization processing; If the optimal gripping point of the cable reel is within the size and position of the cable reel, the control instructions of the manipulator are calculated based on the optimal gripping point of the cable reel and the optimal stacking position of the stacking area.

2. The automatic cable reel stacking system based on image processing technology according to claim 1 is characterized in that: The comprehensive analysis module (300) comprises a gripping point analysis unit (301), wherein the gripping point analysis unit (301) calculates the optimal gripping point of the cable reel, specifically comprising: Extract key points from the cable contour by using a scale-invariant feature transformation algorithm; Determine the geometric characteristics of the cable reel according to its shape characteristics; Calculate the center of gravity of the cable reel by obtaining the coordinates of all points on the outline of the cable reel, where the center of gravity X is the sum of the x coordinates of all points divided by the number of x coordinates, the center of gravity Y is the sum of the y coordinates of all points divided by the number of y coordinates, and the center of gravity Z = the sum of the z coordinates of all points / the number of z coordinates; Divide the outline of the cable into several small areas, calculate the area and center of gravity of each area, analyze the area distribution and center of gravity position of each small area, select the area with uniform mass distribution, and ensure the uniform mass distribution in the area where the grab point is located by comparing the area and center of gravity of each area; Select points close to the edge as candidate grasping points, and points on the contour are edge points; The moment analysis is used to select the candidate grasping point with the minimum moment. The moment formula is moment = force × lever arm length, where the lever arm length is the distance from the candidate grasping point to the center of gravity, and force is the force applied during grasping; The point with the smallest torque is selected, and the distance from each candidate grasping point to the center of gravity is calculated, and the candidate grasping point with the smallest torque is selected as the optimal grasping point of the cable reel.

3. The automatic cable reel stacking system based on image processing technology according to claim 2 is characterized in that: The comprehensive analysis module (300) comprises a stacking position analysis unit (302), wherein the stacking position analysis unit (302) calculates the optimal stacking position of the stacking area according to the size and position of the cables and the free position of the stacking area, specifically comprising: According to the free positions in the stacking area, the coordinates and sizes of all free positions in the stacking area are obtained; Filter out suitable free positions according to the size of the cable reel, and filter out suitable free positions by filtering out cable reels whose size is smaller than the size of the free positions in the stacking area; Find the area with the best matching cable reel size in the free positions in the stacking area, calculate the difference between the free positions in the stacking area and the cable reel size, and select the free position in the stacking area with the smallest size difference as the best stacking position in the stacking area.

4. The automatic cable reel stacking system based on image processing technology according to claim 1 is characterized in that: The decision module (400) comprises a detection unit (401), wherein the detection unit (401) detects whether the optimal gripping point of the cable reel is within the size and position of the cable reel, and the detection process specifically comprises: Calculate the minimum and maximum coordinates of the cable reel according to the size and position of the cable reel, where: the minimum x coordinate is x_min=x; the maximum x coordinate is x_max=x+l; the minimum y coordinate is y_min=y; the maximum y coordinate is y_max=y+w; the minimum z coordinate is z_min=z; the maximum z coordinate is z_max=z+h; where (x, y, z) is the position coordinate of the cable reel, and (l, w, h) is the size coordinate of the cable reel; The coordinates of the best gripping point of the cable reel are (x_g, y_g, z_g). By detecting the x coordinate, we can determine whether x_g is between x_min and x_max; by detecting the y coordinate, we can determine whether y_g is between y_min and y_max; by detecting the z coordinate, we can determine whether z_g is between z_min and z_max. Among them: If the x, y, and z coordinates are all within the corresponding bounds, then the optimal gripping point for the cable reel is within the size and position of the cable reel; If any coordinate is not within the corresponding boundary range, then the optimal gripping point of the cable reel is not within the size and position of the cable reel.

5. The automatic cable reel stacking system based on image processing technology according to claim 4 is characterized in that: The decision module (400) comprises an optimization control unit (402), wherein the optimization control unit (402) performs corresponding optimization and control according to the detection result of the detection unit (401), specifically comprising: If the optimal gripping point of the cable reel is not within the size and position of the cable reel, the Gaussian filter in the data collection and processing module (100) is modified to a bilateral filter, the spatial weight is based on the weight distribution of the spatial distance, the value weight is based on the weight distribution of the pixel value difference, the closer the pixel value is, the greater the weight is, and vice versa, the smaller the weight is, and the product of the spatial weight and the value weight is used to perform normalized weighted averaging to obtain the filtered pixel value; If the optimal gripping point of the cable reel is within the size and position of the cable reel, the control instructions of the manipulator are calculated based on the optimal gripping point of the cable reel and the optimal stacking position of the stacking area.