Cable reel rotation angle calibration method and system based on digital image processing

Abstract

The present application provides a method and system for calibrating the rotation angle of a cable reel based on digital image processing. During the operation of the overhead crane, the present application captures real-time images of the cable reel below the overhead crane through an image acquisition device installed on the overhead crane, and uses a server to detect the cable reel target in the real-time images of the cable reel, identify the frame boundary line corresponding to the cable reel target, thereby calculating the rotation angle of the frame boundary line relative to the overhead crane. According to the rotation angle of the frame boundary line relative to the overhead crane, the overhead crane clamp is triggered to clamp the cable reel accordingly and adjust its rotation angle and / or position. Thus, the present application can detect the offset position and rotation angle of the cable reel through simple digital image processing steps and adjust it in a timely manner, avoiding abnormal tilting of the cable reel, thereby greatly reducing the potential safety hazards during the process of the overhead crane loading and unloading the cable reel, avoiding accidents, and reducing economic losses.

Description

Technical Field

[0001] This application relates to the technical field of cable reel loading and unloading for unmanned cranes, and specifically relates to a method and system for calibrating the rotation angle of a cable reel based on digital image processing. Background Art

[0002] Large cable reels are common materials in the warehousing of the power industry. Currently, in the industry, it is generally defaulted that the cable reel is stationary during the loading and unloading operation of the cable reel on the unmanned crane. If, under certain external force factors, the cable reel undergoes simple translation, under the existing technology, the target detection algorithm can directly find the moved cable reel and correct its position. However, if the external force factor causes the cable reel to rotate, the existing target detection algorithm will not be able to determine the rotation angle of the cable reel, resulting in the inability to identify the abnormal tilt state of the cable reel, thus greatly increasing the safety hazards during the process of the crane loading and unloading the cable reel. Once an accident occurs, it will cause great economic losses.

[0003] To reduce the above risks, in existing warehouses, the on-site condition of the cable reel is often monitored by manual observation. However, due to the large area where the cable reel is placed, the manual observation method often cannot obtain accurate results, and this method also greatly increases the safety hazards of these workers.

[0004] In response to the above problems, some warehouses install fixed iron frames at the bottom of the cable reel. This method has the following defects: First, it increases the warehousing cost. In particular, since the cable reels vary in size, a single-size fixed iron frame cannot meet the requirements, and it is necessary for the warehouse to always keep fixed iron frames of different sizes for fixing, which greatly increases the complexity of the warehousing project; second, the process of fixing the iron frame often needs to be completed manually, which will further increase the labor cost and reduce the intelligence level of the warehouse. Summary of the Invention

[0005] In view of the deficiencies of the prior art, this application provides a method and system for calibrating the rotation angle of a cable reel based on digital image processing. In response to the problem in the actual warehouse scenario that the cable reel is heavy and inconvenient to handle and calibrate, but there is always a deflection probability that affects the operation of warehouse equipment, an identification algorithm for the rotation angle of the cable reel is added on the basis of the target detection algorithm, which increases the ability to handle abnormal situations during the operation of the crane and improves the safety during the process of the unmanned crane loading and unloading the cable reel. The specific technical solutions adopted in this application are as follows.

[0006] First, to achieve the above object, a method for calibrating the rotation angle of a cable reel based on digital image processing is proposed, and its steps include: First step, during the operation of the traveling crane, a real-time image of the cable reel under the traveling crane is captured; Second step, detecting the cable reel target in the real-time image of the cable reel; Third step, identifying the frame boundary line corresponding to the cable reel target and calculating the rotation angle of the frame boundary line relative to the traveling crane; Fourth step, according to the rotation angle of the frame boundary line relative to the traveling crane, triggering the traveling crane's clamping device to clamp the cable reel accordingly and adjust its rotation angle and / or position.

[0007] Optionally, for the method for calibrating the rotation angle of a cable reel based on digital image processing as described in any of the above, wherein, in the second step, the cable reel target in the real-time image of the cable reel is detected specifically according to the following steps: Step 201, input the real-time image of the cable reel into a deep learning model trained by a cable reel picture data set; Step 202, mark the cable reel recognition frame according to the detection result of the deep learning model; Step 203, calculate the straight-line distance between the center point M1(m1, m2) of the cable reel recognition frame and the center point M2(m3, m4) of the real-time image of the cable reel. When the straight-line distance L between the two exceeds the alarm threshold, trigger an alarm prompt of the traveling crane to perform manual processing and adjustment on the cable reel. When the straight-line distance L between the two does not exceed the alarm threshold, trigger the traveling crane to clamp the cable reel according to the position of the cable reel recognition frame and adjust its offset position until it is restored to its assigned position.

[0008] Optionally, for the method for calibrating the rotation angle of a cable reel based on digital image processing as described in any of the above, wherein, in the third step, the frame boundary line corresponding to the cable reel target is identified specifically according to the following steps and the rotation angle of the frame boundary line relative to the traveling crane is calculated: Step 301, perform digital image processing and Hough transform processing on the real-time image of the cable reel in the cable reel recognition frame, and filter out the frame boundary line corresponding to the cable reel target; Step 302, extract the pixel coordinates (X1, Y1), (X2, Y2) of two points in the frame boundary line corresponding to the cable reel target, and calculate the slope k of the cable reel frame boundary line as k = (Y2 - Y1) / (X2 - X1); Step 303, calculate the rotation angle θ of the cable reel frame boundary line as θ = 90° - |arctan(k)|, and when arctan(k) > 0, judge that the rotation direction of the cable reel is clockwise, and when arctan(k) < 0, judge that the rotation direction of the cable reel is counterclockwise, and trigger the traveling crane's clamping device to adjust the rotation angle of the cable reel in the reverse direction until it is restored to its assigned position.

[0009] Optionally, for the method for calibrating the rotation angle of a cable reel based on digital image processing as described in any of the above, wherein, in step 301, the longest straight line obtained by the Hough transform processing is filtered out as the frame boundary line corresponding to the cable reel target.

[0010] Optionally, for any of the above-described cable reel rotation angle calibration methods based on digital image processing, in step 301, the digital image processing performed on the real-time image of the cable reel in the cable reel recognition frame sequentially includes: grayscale processing, filtering, binarization, edge detection, dilation, and erosion.

[0011] Meanwhile, to achieve the above object, the present application also provides a cable reel rotation angle calibration system based on digital image processing, which includes: an image acquisition device installed on the traveling crane for capturing a real-time image of the cable reel below the traveling crane; a server communicatively connected to the image acquisition device, receiving the real-time image of the cable reel and detecting the cable reel target therein, identifying the frame boundary line corresponding to the cable reel target, and calculating the rotation angle of the frame boundary line relative to the traveling crane; a communication module connected to the traveling crane control end and the server, for triggering the traveling crane clamp to clamp the cable reel accordingly and adjust its rotation angle and / or position according to the rotation angle of the frame boundary line relative to the traveling crane.

[0012] Optionally, for any of the above-described cable reel rotation angle calibration systems based on digital image processing, the server includes: a target detection module for inputting the real-time image of the cable reel into a deep learning model trained by a cable reel picture data set, then marking the cable reel recognition frame according to the detection result of the deep learning model, and finally calculating the straight-line distance between the center point M1(m1, m2) of the cable reel recognition frame and the center point M2(m3, m4) of the real-time image of the cable reel When the straight-line distance L between the two exceeds the alarm threshold, trigger the traveling crane to give an alarm prompt for manual processing and adjustment of the cable reel. When the straight-line distance L between the two does not exceed the alarm threshold, trigger the traveling crane to clamp the cable reel according to the position of the cable reel recognition frame and adjust its offset position until it is restored to its assigned position; a digital image processing module for performing digital image processing and Hough transform processing on the real-time image of the cable reel in the cable reel recognition frame, screening out the frame boundary line corresponding to the cable reel target, then extracting the pixel coordinates (X1, Y1), (X2, Y2) of two points in the frame boundary line corresponding to the cable reel target, calculating the slope k of the cable reel frame boundary line as k = (Y2 - Y1) / (X2 - X1), and finally calculating the rotation angle θ of the cable reel frame boundary line as θ = 90° - |arctan(k)|. When arctan(k) > 0, it is determined that the rotation direction of the cable reel is clockwise, and when arctan(k) < 0, it is determined that the rotation direction of the cable reel is counterclockwise, and trigger the traveling crane clamp to adjust the rotation angle of the cable reel in the reverse direction until it is restored to its assigned position.

[0013] Optionally, for any of the cable reel rotation angle calibration systems based on digital image processing described above, a grayscale processing unit, a filtering processing unit, a binarization processing unit, an edge detection processing unit, a dilation processing unit, and an erosion processing unit are provided in the digital image processing module, which are connected in sequence to perform digital image processing on the real-time image of the cable reel in the cable reel recognition frame in sequence, and perform a Hough transform on the processed image to detect straight lines therein.

[0014] Beneficial effects

[0015] During the operation of the crane, the real-time image of the cable reel under the crane is captured by the image acquisition device installed on the crane. The server is used to detect the cable reel target in the real-time image of the cable reel, identify the frame boundary line corresponding to the cable reel target, thereby calculating the rotation angle of the frame boundary line relative to the crane. According to the rotation angle of the frame boundary line relative to the crane, the crane clamp is triggered to clamp the cable reel accordingly and adjust its rotation angle and / or position. Thus, the present application can detect the offset position and rotation angle of the cable reel through simple digital image processing steps and adjust it in time, avoiding abnormal inclination of the cable reel, thereby greatly reducing the potential safety hazards in the process of the crane loading and unloading the cable reel, avoiding accidents, and reducing economic losses.

[0016] Other features and advantages of the present application will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present application. Brief description of the drawings

[0017] The drawings are used to provide a further understanding of the present application, and constitute a part of the specification, and together with the embodiments of the present application, are used to explain the present application, and do not constitute a limitation to the present application. In the drawings:

[0018] Figure 1 is a schematic diagram of the real-time image of the square cable reel captured by the image acquisition device in the system of the present application;

[0019] Figure 2 is for Figure 1 the schematic diagram of filtering processing the image in;

[0020] Figure 3 is for Figure 2 the schematic diagram of binarization processing the image in;

[0021] Figure 4 is for Figure 3 the schematic diagram of edge detection processing the image in;

[0022] Figure 5 is for Figure 4 the schematic diagram of dilation processing the image in;

[0023] Figure 6 is a schematic diagram of the erosion processing of the image in Figure 5 ;

[0024] Figure 7 is a schematic diagram of the Hough transform processing of the image in Figure 6 ;

[0025] Figure 8 is a schematic diagram of the clamping and turning of the cable reel. Specific embodiments

[0026] To make the objectives and technical solutions of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present application. Apparently, the described embodiments are only a part rather than all of the embodiments of the present application. Based on the described embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0027] Those skilled in the art of the present technology can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the field to which the present application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless defined as herein.

[0028] The meaning of "and / or" as described in the present application refers to the situation where each exists alone or both exist simultaneously.

[0029] The meaning of "connection" as described in the present application can be a direct connection between components or an indirect connection between components through other components.

[0030] The meaning of "up and down" as described in the present application refers to the direction from the crossbeam of the overhead crane to the goods being carried and lifted during the operation of the overhead crane, which is the down direction, and vice versa, rather than a specific limitation on the device mechanism of the present application.

[0031] The cable reel rotation angle calibration system based on digital image processing provided by the present application is implemented by the following device units:

[0032] An image acquisition device, which is installed on the overhead crane and is used to capture and obtain the real-time image of the cable reel under the overhead crane;

[0033] A server, which is communicatively connected to an image acquisition device. It can be directly implemented through an independent processor or directly integrate the corresponding functions on a vehicle control system. A target detection module and a digital image processing module are preset in the server. The two cooperate with each other to detect the cable reel target in the real-time image of the cable reel received by the server, identify the frame boundary line corresponding to the cable reel target, calculate the rotation angle of the frame boundary line relative to the vehicle, and mark the rotation angle of the cable reel itself with the rotation angle of the cable reel frame boundary line to calculate the rotation state of the cable reel;

[0034] A communication module, which is connected to the vehicle control terminal and the server, and is used to send the rotation angle and offset condition of the cable reel to the unmanned vehicle control terminal by using a communication protocol, and trigger the vehicle to clamp the cable reel according to the rotation angle of the frame boundary line relative to the vehicle, so as to adjust the cable reel to a suitable rotation angle and / or position.

[0035] Thus, through digital image processing technology, the present application can directly calculate the rotation angle and direction of the cable reel through the cable reel boundary line, thereby accurately calibrating the angle and position of the cable reel, greatly improving the safety during the operation of the unmanned vehicle, reducing the warehousing cost, and improving the unmanned and intelligent level of the warehouse work.

[0036] In the specific implementation process, image acquisition devices such as camera hardware and image sensing devices in the system of the present application are installed on the vehicle clamp according to requirements. During specific installation, the position suitable for installing the camera hardware can be found according to actual scenarios, such as the height of the vehicle, the height of the vehicle clamp, and the size of the cable reel, etc., to obtain the real-time image of the cable reel under the vehicle in real time. The shooting position of the image acquisition device is generally selected directly above the hoisting position of the cable reel and parallel to the vehicle crossbeam, so as to obtain the complete frame boundary line of the cable reel and determine the angular relationship between it and the vehicle crossbeam.

[0037] At the same time, deploy the server, realize the data interaction between the image acquisition devices such as camera hardware and image sensing devices and the switch through network cables, and use the program running in the server to perform target detection and digital image processing on the images acquired by the image acquisition devices:

[0038] A target detection module can be integrated and set in the server. The computer runs a program to identify and detect the cable reel target in real-time images according to the following steps: First, use a camera to collect a large number of cable reel pictures. Then, manually label the positions and categories of cable reels in each picture, and divide the labeled pictures into a training set and a test set according to a certain ratio. Use a target detection algorithm to establish a deep learning model for target detection, use the training set to train the deep learning model for target detection, determine the parameters of the model in deep learning, and use the test set to test the trained deep learning model to evaluate the performance of the deep learning model, including the accuracy of detecting and identifying cable reels and the rate of detecting and identifying cable reels. Adopt methods such as parameter adjustment and data augmentation to improve the accuracy of the deep learning model in identifying cable reels in pictures; adopt methods such as model pruning, parameter quantization, and sharing to improve the rate of the deep learning model in identifying cable reels in pictures. Finally, deploy the optimized deep learning model to the server connected to the image acquisition device, input the real-time image of the cable reel into the deep learning model trained by the cable reel picture dataset, then mark the cable reel recognition frame according to the detection result of the deep learning model, and finally calculate the straight-line distance between the center point M1(m1, m2) of the cable reel recognition frame and the center point M2(m3, m4) of the real-time image of the cable reel. When the straight-line distance L between the two exceeds the alarm threshold, trigger a vehicle alarm prompt to manually process and adjust the cable reel. When the straight-line distance L between the two does not exceed the alarm threshold, trigger the vehicle to clamp the cable reel according to the position of the cable reel recognition frame and adjust its offset position until it is restored to its original position.

[0039] A digital image processing module is generally also integrated and set in the server: It uses a series of digital image processing operations, including: image grayscale conversion, filtering, binarization, edge detection, erosion, dilation, Hough transformation, etc. to complete the extraction of the boundary line of the cable reel frame. Use the target detection module to find the cable reel in the loading and unloading position. First, perform grayscale conversion on the real-time image of the cable reel in the cable reel recognition frame, and unify the RGB values of each pixel point in the image into the same grayscale value. As Figure 1 shown, the grayscale image will change from three channels to a single channel, and the single-channel data is much easier to process, which can simplify the subsequent image processing algorithm. Then, perform filtering processing on the grayscale image obtained by grayscale conversion to eliminate the noise in the image and obtain the Figure 2 shown image. Then perform binarization processing on the Figure 2 filtered image in Figure 3 to obtain the binarized image shown, so that the image becomes simple, reduces the data volume, and facilitates the digital image processing module to further process the binarized image to highlight the contour of the target of interest. Then perform edge detection processing on the binarized image to find the Figure 4The pixel points with drastic brightness changes in the image shown in the figure are a set, and the pixel points in this set show the overall outline of the cable drum frame in the image. If the edge in the image can be accurately measured and located, it means that the actual object can be located and measured, and the parameters including the area, diameter, and shape of the object can be obtained by identifying and calculating the image features. Then, the edge detected image is expanded to fill the gaps between the edge lines of the image and eliminate the boundary burrs, and the following is obtained: Figure 5 The dilated image is shown in Figure 1. Then, the dilated image is eroded to eliminate the boundary and shrink the boundary inward to eliminate the small and meaningless noise points in the boundary. Figure 6 Finally, the eroded image is subjected to Hough transform to find the possible straight lines in the image, and the straight line length is used to filter out the straight line that can represent the cable drum frame boundary to obtain Figure 7 The long straight line with a medium slant indicates the boundary line of the cable drum target frame.

[0040] At this point, the rotation angle of the cable drum relative to the driving direction, that is, the rotation angle of the boundary line of the cable drum target frame relative to the horizontal and vertical coordinates of the cable drum real-time image can be calculated: the rotation angle of the cable drum is equal to the rotation angle of the boundary line of its cable drum frame, and the rotation angle of the cable drum frame boundary line relative to the horizontal and vertical coordinates of the image obtained by the camera installation position on the driving beam is calculated to obtain the rotation angle of the cable drum relative to the driving vehicle. In the calculation process, the pixel point coordinates (X1, Y1) and (X2, Y2) of two points of the cable drum frame boundary line can be used to calculate the slope k = (Y2-Y1) / (X2-X1) of the cable drum frame boundary line, and the inclination angle θ = 90°-|arctan(K)| of the cable drum frame boundary line relative to the vertical coordinate of the image screen is calculated according to the slope. Since the image acquisition device is parallel to the driving structure itself, this inclination angle in the image screen is the inclination angle of the cable drum boundary line relative to the driving beam, that is, the rotation angle of the cable drum main structure in the cable drum frame relative to the driving vehicle. If arctan(K) is less than zero at this time, it can be determined that the cable drum is rotating clockwise or counterclockwise based on the position relationship between the horizontal and vertical coordinates of the screen and the vehicle. If arctan(K) is greater than zero, the corresponding cable drum is rotating clockwise.

[0041] Thus, the server can be connected to the vehicle control terminal through its communication module: judge whether the rotation angle of the cable reel exceeds a preset and determined threshold according to the angle obtained above. If the angle exceeds this threshold, it is considered that the rotation angle of the cable reel exceeds the acceptance level of the vehicle loading and unloading. At this time, this angle is sent to the unmanned vehicle control terminal using the TCP / IP or HTTP communication protocol, triggering the vehicle to clamp the cable reel according to this rotation angle and perform a corresponding rotation of the angle to correct the angle of the cable reel back to ensure the successful grasping of the cable reel by the vehicle. At the same time, during the grasping process, if the cable reel does not move, at this time the vehicle clamp should be just above the cable reel, then the position of the cable reel in the picture captured by the camera must be centered, and the target detection module will detect the cable reel and mark it with a rectangular cable reel recognition frame on the picture. Figure 1 As shown, at this time, the center point M1(m1, m2) of the rectangular cable reel recognition frame will be near the center point M2(m3, m4) of the real-time image of the cable reel. On the contrary, if the distance between these two center points M1 and M2 is far, then it is considered that the cable reel has undergone a large-scale displacement. This displacement degree can be reflected by the distance between the two points. Therefore, the system can set an alarm threshold. When the distance L between the two center points M1 and M2 exceeds the alarm threshold, it is considered that the cable reel has moved on a large scale, the cable reel has significantly deviated from the current position, and the vehicle cannot complete the cable reel clamping work. The communication module can accordingly send a signal to the vehicle control terminal through the communication protocol, and the vehicle will perform an alarm process to remind the operator to perform manual processing and adjustment to prevent more serious situations; on the contrary, if the distance between the two center points M1 and M2 does not exceed the alarm threshold, it is considered that the movement range of the cable reel is within the acceptable range of the vehicle operation, and at this time, the clamping work can be directly carried out.

[0042] The above is only the implementation mode of this application, and its description is relatively specific and detailed, but it cannot be understood as a limitation to the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several deformations and improvements can still be made, and these all belong to the protection scope of this application.

Claims

1. A method for calibrating the rotation angle of a cable reel based on digital image processing, characterized in that the steps Including: The first step is to capture real-time images of the cable reel under the traveling crane during its operation. The second step is to detect the cable reel target in the real-time image of the cable reel. The third step is to identify the frame boundary line corresponding to the cable reel target and calculate the rotation angle of the frame boundary line relative to the traveling crane. The fourth step is to trigger the traveling crane's gripper to clamp the cable reel accordingly based on the rotation angle of the frame boundary line relative to the traveling crane and adjust its rotation angle and / or position. Wherein, In the second step, the cable reel target in the real-time image of the cable reel is detected according to the following specific steps: Step 201: Input the real-time image of the cable reel into the deep learning model trained by the cable reel picture dataset. Step 202: Mark the cable reel recognition frame according to the detection result of the deep learning model. Step 203, calculate the straight-line distance between the center point M1(m1, m2) of the cable reel recognition frame and the center point M2(m3, m4) of the real-time image of the cable reel When the straight-line distance L between the two exceeds the alarm threshold, trigger a vehicle alarm prompt to manually process and adjust the cable reel. When the straight-line distance L between the two does not exceed the alarm threshold, trigger the vehicle to clamp the cable reel according to the position of the cable reel recognition frame and adjust its offset position until it is restored to its assigned bin; Wherein, In the third step, the frame boundary line corresponding to the cable reel target is identified and the rotation angle of the frame boundary line relative to the traveling crane is calculated according to the following specific steps: Step 301: Perform digital image processing and Hough transform processing on the real-time image of the cable reel in the cable reel recognition frame to screen out the frame boundary line corresponding to the cable reel target. Step 302: Extract the pixel coordinates (X1, Y1), (X2, Y2) of two points in the frame boundary line corresponding to the cable reel target, and calculate the slope k of the cable reel frame boundary line as k = (Y2 - Y1) / (X2 - X1). Step 303: Calculate the rotation angle θ of the cable reel frame boundary line as θ = 90° - |arctan(k)|, and when arctan(k) > 0, determine that the rotation direction of the cable reel is clockwise, and when arctan(k) < 0, determine that the rotation direction of the cable reel is counterclockwise, and trigger the traveling crane's gripper to adjust the rotation angle of the cable reel in the reverse direction until it is restored to its original position.

2. The method for calibrating the rotation angle of a cable reel based on digital image processing according to claim 1, wherein In step 301, the longest straight line obtained by the Hough transform processing is selected as the frame boundary line corresponding to the cable reel target.

3. The method for calibrating the rotation angle of a cable reel based on digital image processing according to claim 2, characterized in that, In step 301, the digital image processing performed on the real-time image of the cable reel in the cable reel recognition frame successively includes: grayscale processing, filtering, binarization, edge detection, dilation, and erosion.

4. A cable reel rotation angle calibration system based on digital image processing, characterized in that, The system includes: An image acquisition device, which is installed on the traveling crane and is used to capture real-time images of the cable reel under the traveling crane. A server, which is communicatively connected to the image acquisition device, receives the real-time image of the cable reel, detects the cable reel target therein, identifies the frame boundary line corresponding to the cable reel target, and calculates the rotation angle of the frame boundary line relative to the traveling crane. A communication module, which is connected to the traveling crane control terminal and the server, and is used to trigger the traveling crane's gripper to clamp the cable reel accordingly based on the rotation angle of the frame boundary line relative to the traveling crane and adjust its rotation angle and / or position. Wherein, the server includes: The target detection module is used to input the real-time image of the cable reel into the deep learning model trained by the cable reel picture data set, then mark the cable reel recognition frame according to the detection result of the deep learning model, and finally calculate the straight-line distance between the center point M1(m1, m2) of the cable reel recognition frame and the center point M2(m3, m4) of the real-time image of the cable reel. When the straight-line distance L between the two exceeds the alarm threshold, a vehicle alarm prompt is triggered to manually process and adjust the cable reel. When the straight-line distance L between the two does not exceed the alarm threshold, the vehicle is triggered to clamp the cable reel according to the position of the cable reel recognition frame and adjust its offset position until it is restored to its assigned position. A digital image processing module is used to perform digital image processing and Hough transform processing on the real-time image of the cable reel in the cable reel recognition frame, screen out the frame boundary line corresponding to the cable reel target, and then extract the pixel coordinates (X1, Y1), (X2, Y2) of two points in the frame boundary line corresponding to the cable reel target, calculate the slope k of the cable reel frame boundary line as k = 2(Y - Y1) / (X2 - X1), finally calculate the rotation angle θ of the cable reel frame boundary line as θ = 90° - |arctan(k)|, and determine that the rotation direction of the cable reel is clockwise when arctan(k) > 0, and determine that the rotation direction of the cable reel is counterclockwise when arctan(k) < 0, trigger the vehicle clamping to reverse and adjust the rotation angle of the cable reel until it is restored to its belonging position.

5. The cable reel rotation angle calibration system based on digital image processing according to claim 4, wherein, A grayscale processing unit, a filtering processing unit, a binarization processing unit, an edge detection processing unit, a dilation processing unit and an erosion processing unit are set in the digital image processing module, which are connected in sequence to perform digital image processing on the real-time image of the cable reel in the cable reel recognition frame in turn, and perform Hough transform on the processed image to detect the straight lines therein.

Images