Tower crane jacking operation key connection point monitoring method

Abstract

The invention provides a tower crane jacking operation key connection point monitoring method. The tower crane jacking operation key connection point monitoring method comprises the following steps that 1, key connection points are arranged on a tower crane; step 2, adopting an image recognition method to recognize the key connection points during jacking operation of the tower crane to obtain a recognition result; and step 3, judging according to the identification result, giving an alarm, and completing the monitoring of the key connection points of the jacking operation of the tower crane. According to the method, a camera is installed on a tower crane jacking mechanism, whether key connection points are in a normal state or not during jacking operation is detected through an image recognition technology, signals are transmitted to an industrial personal computer according to various states of the connection points, and early warning is conducted on the abnormal state.

Description

Technical Field

[0001] The present invention relates to a tower crane monitoring method, in particular to a monitoring method for key connection points during the tower crane jacking operation. Background Art

[0002] The information provided in this section is only background information related to the present disclosure, and it does not necessarily represent prior art.

[0003] Tower cranes are widely used in construction. Among them, the accident rate of the installation and disassembly operations is relatively high. The disassembly and installation of tower cranes are inseparable from the jacking operation, which is mainly achieved by using a hydraulic cylinder to push the jacking crossbeam to act on the standard section steps. Unsafe behaviors of people are the main causes of installation and disassembly operation accidents. For example, mistakes or failures to perform operations on the movable climbing claws and safety pins of the jacking crossbeam often occur. The current monitoring schemes can only monitor the load transportation link of the tower crane and lack the monitoring of the jacking operation link.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] Object of the Invention: The technical problem to be solved by the present invention is to provide a monitoring method for key connection points during the tower crane jacking operation in view of the deficiencies of the prior art.

[0006] To solve the above technical problem, the present invention discloses a monitoring method for key connection points during the tower crane jacking operation, including the following steps:

[0007] Step 1, set key connection points on the tower crane;

[0008] Step 2, adopt an image recognition method to recognize the key connection points during the tower crane jacking operation to obtain a recognition result;

[0009] Step 3, make a judgment according to the recognition result and give an alarm to complete the monitoring of the key connection points during the tower crane jacking operation.

[0010] Furthermore, the tower crane in Step 1 includes:

[0011] The main limb of the standard section, on the side of which there are steps fixedly provided. The groove on the step is movably connected to the shaft end of the jacking crossbeam. A safety pin is movably arranged on the jacking crossbeam through a safety pin handle. The safety pin passes through the pin hole on the step to lock the jacking crossbeam; the movable climbing claws are installed on the main limb of the jacking sleeve of the jacking sleeve and act on the top end of the step.

[0012] Further, the setting of the key connection points in step 1 includes:

[0013] Step 1-1, setting color labels on the safety pin, including:

[0014] Setting a first color label, a second color label, and a third color label on the action axis end of the safety pin; wherein, when both the first color label and the second color label are exposed, it is the state where the safety pin is not in place, when only the first color label is exposed, it is the state where the safety pin is not safely in place, and when only the third color label is exposed, it is the state where the safety pin is safely in place;

[0015] Step 1-2, setting shape labels on the movable climbing claw, including:

[0016] Setting shape labels with linear features on the fixed end and the movable end of the movable climbing claw respectively, for calculating the angles of the fixed end and the movable end.

[0017] Further, the setting of the color labels on the safety pin in step 1-1 specifically includes:

[0018] The first color label is set at the outer end of the safety pin and occupies 1 / 4 of the area of the safety pin;

[0019] After an interval of 1 / 4 of the area of the safety pin, set the second color label and it occupies 1 / 4 of the area of the safety pin;

[0020] The third color label is set at the inner end of the safety pin, and its area is 1 / 2 of the area of the handle of the safety pin.

[0021] Further, the shape label in step 1-2 is a rectangular label.

[0022] Further, the identification of the key connection points in step 2 includes:

[0023] Step 2-1, collecting images of the key connection points set in step 1;

[0024] Step 2-2, using the YOLO v8 method to identify the images of the key connection points, including:

[0025] Step 2-2-1, obtaining the color of the label that is exposed among the color labels set on the safety pin;

[0026] Step 2-2-2, obtaining the linear features of the shape labels set on the movable climbing claw and calculating the included angle between the fixed end and the movable end of the movable climbing claw.

[0027] Further, the judgment in step 3 includes:

[0028] Step 3-1: Based on the color of the exposed label obtained in Step 2-2-1, determine the position of the safety pin and issue an alarm.

[0029] Step 3-2: Based on the included angle obtained in Step 2-2-2, determine the angle of the movable climbing claw and issue an alarm.

[0030] Further, the determination of the position of the safety pin in Step 3-1 includes:

[0031] When the color of the exposed label includes the colors of the first color label and the second color label, it is determined that the safety pin is not in place, and an alarm is issued.

[0032] When the color of the exposed label only includes the color of the first color label, it is determined that the safety pin is not safely in place, and an alarm is issued.

[0033] When the color of the exposed label only includes the color of the third color label, it is determined that the safety pin is safely in place, and a prompt is given.

[0034] Further, the obtaining of the linear feature of the shape label set on the movable climbing claw and the calculation of the included angle between the fixed end and the movable end of the movable climbing claw in Step 2-2-2 include:

[0035] Step 2-2-2-1: Use grayscale conversion to convert the key connection point image corresponding to the movable climbing claw into a grayscale image.

[0036] Step 2-2-2-2: Use Gaussian filtering to remove the noise in the grayscale image, which is expressed as follows:

[0037]

[0038] Among them, G(x, y) is the processed Gaussian weight matrix, (x, y) = (5, 5), indicating the size of the Gaussian kernel, that is, a 5×5 neighborhood is taken for calculation; the standard deviation of the Gaussian kernel in the X-axis direction is σ.

[0039] Step 2-2-2-3: Use the Canny edge detection method to extract the edge information in the image to obtain a binary image after Canny edge detection, and the binary image is scattered points.

[0040] Step 2-2-2-4: Use the Hough transform to detect the straight lines in the image, that is, transform the scattered points into straight lines.

[0041] Step 2-2-2-5: Straight line fitting, fit the straight lines obtained in Step 2-2-2-4 using the RANSAC method, which specifically includes:

[0042] Randomly sample points and calculate the straight-line equation:

[0043] y = mx + b

[0044] where m is the slope and b is the intercept;

[0045] Calculate the distance d from all sampled points to the straight line:

[0046]

[0047] Select the straight line with the most inliers as the final fitting result;

[0048] Step 2-2-2-5, calculate the angle, specifically including:

[0049] Assume that in the image corresponding to the fixed end and the movable end of the movable climbing claw, the slopes of the two straight lines obtained by fitting are m1 and m2 respectively; then calculate the included angle θ between the two straight lines:

[0050]

[0051] Step 2-2-2-5, perform smoothing processing on the calculated included angle value θ using Kalman filtering.

[0052] Furthermore, the determination of the angle of the movable climbing claw described in step 3-2 includes:

[0053] When the included angle θ is within the preset range, it is determined to be in a safe in-place state; otherwise, it is in an unsafe in-place state, and corresponding alarms or prompts are given.

[0054] Beneficial effects:

[0055] The key connection point monitoring method for tower crane jacking operations provided by the present invention, compared with traditional manual detection methods and current monitoring solutions, adds safety detection of key connection points in tower crane jacking operations, facilitates intuitive monitoring of jacking operations by management personnel, standardizes the operation process, effectively solves the safety supervision problem during tower crane jacking operations, and improves the safety management and safety of jacking operations. Brief description of the drawings

[0056] The following further specifically describes the present invention in conjunction with the drawings and specific embodiments, and the above and / or other advantages of the present invention will become clearer.

[0057] Figure 1 It is a schematic diagram of the structure of the jacking crossbeam.

[0058] Figure 2 It is a system block diagram of the key connection point monitoring method in an embodiment.

[0059] Figure 3aSchematic diagram of the installation of red and white labels on the safety pin in an embodiment.

[0060] Figure 3b Schematic diagram of the installation of the red label on the safety pin in an embodiment.

[0061] Figure 3c Schematic diagram of the installation of the green label on the safety pin in an embodiment.

[0062] Figure 4 Schematic diagram of the installation of the color label on the climbing claw in an embodiment.

[0063] Figure 5 Schematic diagram of the recognition result of the color label on the climbing claw in an embodiment.

[0064] Figure 6 Flow chart of the climbing claw image recognition.

[0065] Figure 7 Schematic diagram of the overall structure of the jacking mechanism.

[0066] In the figure, 1 is the main limb of the standard section, 2 is the step, 3 is the jacking cross beam, 4 is the safety pin, 5 is the safety pin handle, 6 is the movable climbing claw, 7 is the main limb of the jacking sleeve frame, 8 is the first blue label, 9 is the second blue label, 10 is the red label, 11 is the white label, and 12 is the green label. Detailed implementation manner

[0067] The overall idea of the present invention is as follows: A method for monitoring key connection points during tower crane jacking operation, which uses image recognition technology to detect the working states of the safety pin of the jacking cross beam and the movable climbing claw. When it is detected that the safety pin of the jacking cross beam is not in the normal state required for the jacking operation, it is determined that the tower crane cannot perform the jacking operation. The industrial control computer will give a voice alarm, and the ground management personnel will simultaneously monitor and require the operators to make rectifications according to the operation requirements. The present invention can increase the safety detection of key connection points during tower crane jacking operation, facilitate the intuitive monitoring of the jacking operation by the management personnel, standardize the operation process, effectively solve the safety supervision problem during tower crane jacking operation, and improve the safety management and safety of the jacking operation.

[0068] The specific technical solution adopted by the present invention is as follows:

[0069] The present invention proposes a method for monitoring key connection points during tower crane jacking operation, wherein the tower crane structure includes: the main limb 1 of the standard section, the step 2, the jacking cross beam 3, the safety pin 4, the safety pin handle 5, the movable climbing claw 6, and the main limb 7 of the jacking sleeve frame. The overall structure schematic diagram is as Figure 7 shown.

[0070] Set around the target tower crane: a high-definition camera, an industrial control computer, a wireless transmission unit and a ground tower crane monitoring terminal for monitoring key connection points during the jacking operation. The system block diagram is asFigure 2 as shown

[0071] Among them, the images of the high-definition cameras can cover the jacking crossbeam 3, the safety pin 4, the safety pin handle 5, and the movable climbing claw 6. The high-definition cameras can identify whether the safety pin 4 and the movable climbing claw 5 are in what working state, transmit signals to the industrial control computer according to various states of the key connection points and conduct voice announcements, and display them on the ground tower crane monitoring terminal interface, facilitating the supervision of the jacking operation or the tower crane operation.

[0072] The state of the key connection points is the working state of the jacking crossbeam safety pin 4 and the movable climbing claw 6.

[0073] The number of high-definition cameras is at least 4.

[0074] The installation positions of the four high-definition cameras are on the jacking sleeve. Two of the cameras are directed at the movable climbing claw 6, and the other two high-definition cameras are directed at the safety pins 4 at both ends of the jacking crossbeam.

[0075] The high-definition cameras are installed using strong magnets.

[0076] Only when all the movable climbing claws 6 and the jacking crossbeam safety pins 4 are in the normal working state required by this operation process, is the next process allowed. During the jacking operation, voice announcements are made in real time, and for abnormal working states, the operators are required to make rectifications.

[0077] The working states include the safe in-place state, the unsafe in-place state, and the not in-place state.

[0078] Draw different color labels, such as red, green, and white, or labels of different shapes, such as square or other shapes, on the working axis end of the safety pin 4.

[0079] Taking the color labels as an example, the sizes of the red and white color labels occupy 1 / 4 of the area of the tail end of the safety pin 4 when it is not in use, and the interval size of the labels occupies 1 / 4 of the area of the tail end of the safety pin 4 when it is not in use. The size of the green color label occupies 1 / 2 of the area of the safety pin handle 5. Use the YOLO v8 algorithm (reference: https: / / blog.csdn.net / Natsuago / article / details / 135785356?spm=1001.2014.3001.5506) to identify the color labels to determine what working state the safety pin is in. The specific judgment method is as follows:

[0080] When both the red and white color labels are recognized simultaneously, it is in the not in-place state. When and only when the red color label is recognized, it is in the unsafe in-place state. When and only when the green color label is recognized, it is in the safe in-place state.

[0081] Draw color labels on the flipping action end and the fixed end of the movable climbing claw 6, such as blue strip-shaped labels. It is best to be strip-shaped (convenient for extracting linear features and calculating the angle), and the color can be other colors but avoid repeating the color of the safety pin label.

[0082] Taking the color label as an example, use the YOLO v8 algorithm to identify the color label and calculate the angle formed by the color label to obtain the inclination angle of the climbing claw 6 with respect to the reference plane. Compare the inclination angle with the in-place state angle of the climbing claw 6 to determine what working state the climbing claw is in. The specific judgment method is as follows:

[0083] When no blue label is recognized or only one blue color label is recognized, it is in the not-in-place state. When the angle is within the range of 0° to 85°, it is in the unsafe in-place state. When the angle is within the range of 85° to 95°, it is in the safe in-place state. The above angle range settings are for reference only and can be adjusted according to the actual situation.

[0084] Taking the use of color labels as an example, collect the images containing color labels, such as images of red, green, white, and blue color labels on a yellow background that conforms to the tower crane color, and use the labelimg method (reference: https: / / blog.csdn.net / didiaopao / article / details / 119808973?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522bc53c740dc05388eaf8fb6146f0efb4f%2522%252C%2522scm%2522%253A%252220140713.130102334..%2522%257D&request_id=bc53c740dc05388eaf8fb6146f0efb4f&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~top_positive~default-2-119808973-null-null.142^v101^pc_search_result_base9&utm_term=labelimg&spm=1018.2226.3001.4187) for annotation and train the YOLOv8n model.

[0085] As Figure 6As shown, when determining what working state the active climbing claw is in, a pre-trained YOLOv8n model is used to detect the targets in the image, identify the target categories (such as the blue label image category shows blue). When no blue label is recognized or only one blue color label is recognized, a voice alarm prompts that the jacking operation is not allowed. When two blue long-strip color labels are recognized simultaneously, image preprocessing is performed on the two blue label images respectively. The original image is converted into a grayscale image, which can reduce the computational complexity and improve the robustness of edge detection at the same time. Gaussian filtering is performed on the grayscale image to reduce noise interference. Then, the Canny edge detection (reference: https: / / blog.csdn.net / Natsuago / article / details / 143665040?ops_request_misc=%257B%2522r equest%255Fid%2522%253A%2522bc53c740dc05388eaf8fb6146f0efb4f%2522%252C%2522scm%2522%253A%252220140713.130102334..%2522%257D&request_id=bc53c740dc05388eaf8fb6146f0efb4f&biz_id=0&utm_medium=distribute.pc_search_result.none-task-bl og-2~all~sobaiduend~default-6-143665040-null-null.142^v101^pc_search_result_base9&ut m_term=labelimg&spm=1018.2226.3001.4187) method is used to extract the edge information in the image to provide feature data for subsequent line detection. Then, the Hough transform is used to extract the line features from the edge image.

[0086] Using the lines detected by the Hough transform in the previous step, the RANSAC algorithm is then used to fit the optimal line, obtaining the slope and intercept of the line after fitting, and drawing the line. Two lines are finally drawn for the two blue long-strip color labels. Calculate the included angle between the two lines. Then, in order to reduce the volatility of the detection data and improve the stability of the calculation results, the Kalman filter is used to smooth the angle values. Finally, the working state of the climbing claw is judged based on the smoothed angle data.

[0087] The Yolov8 pre-trained model is combined with an online data update strategy to improve the state perception ability of the camera for the safety pin 4 and the climbing claw 6, ensuring the stability and accuracy of the detection.

[0088] The industrial control computer uses a wireless transmission unit to transmit wireless signals to the ground tower crane monitoring terminal, facilitating supervisors to monitor the tower crane jacking operation in real time.

[0089] The high-definition camera adopts a wired connection scheme with the industrial control computer for wired signal transmission and power supply.

[0090] Embodiment:

[0091] The following uses a specific actual example to illustrate the key connection point monitoring method for the tower crane jacking operation proposed by the present invention. This method includes the standard section main limb 1, the step 2, the jacking crossbeam 3, the safety pin 4, the safety pin handle 5, the movable climbing claw 6, the jacking sleeve main limb 7, the high-definition camera, the industrial control computer, the wireless transmission unit, and the ground tower crane monitoring terminal. The image of the high-definition camera can cover the Figure 1 shown jacking crossbeam 3, safety pin 4, safety pin handle 5, and movable climbing claw 6.

[0092] Among them, the state of the key connection point is the working state of the jacking crossbeam safety pin 4 and the movable climbing claw 6. The structure of the jacking machine crossbeam is as Figure 7 shown.

[0093] Set around the target tower crane: a high-definition camera, an industrial control computer, a wireless transmission unit, and a ground tower crane monitoring terminal for monitoring the key connection points of the jacking operation. The system framework is as Figure 2 shown.

[0094] Among them, when the high-definition camera can identify that all safety pins 4 and movable climbing claws 6 are in the normal working state required by this operation process, the next process is allowed. During the jacking operation, voice announcements are made in real time, and operators are required to rectify abnormal working states.

[0095] Among them, the installation positions of the four high-definition cameras are on the jacking sleeve. Two of the cameras are directed at the movable climbing claws, and the other two high-definition cameras are directed at the safety pins at both ends of the jacking crossbeam.

[0096] Among them, the working states include the safe in-place state, the unsafe in-place state, and the not in-place state.

[0097] In a specific embodiment, for the safety pin 4 of the jacking crossbeam 3, red 10, green 12, and white square color labels 11 are drawn on the working axis end of the safety pin 4. The sizes of the red 10 and white color labels 11 occupy 1 / 4 of the area of the tail end of the safety pin 4 when it is not in use, and the label spacing size occupies 1 / 4 of the area of the tail end of the safety pin 4 when it is not in use. The size of the green color label 12 occupies 1 / 2 of the area of the safety pin handle 5. The specific installation positions are as Figure 3a , Figure 3b and Figure 3cAs shown in the figure, the YOLO v8 algorithm is used to identify color tags to determine the working state of the safety pin 4. When both the red color tag 10 and the white color tag 11 are identified simultaneously, it is in the not-in-place state. When and only when the red color tag 10 is identified, it is in the unsafe in-place state. When and only when the green color tag 12 is identified, it is in the safe in-place state.

[0098] In a specific embodiment, before the tower crane performs the jacking operation, when the camera monitoring is turned on, when no color tag is identified or the working state of the identified safety pin 4 is in an abnormal state, the industrial control computer will give a voice alarm and does not allow the jacking operation, requiring the operator to make rectifications.

[0099] In a specific embodiment, for the movable climbing claw 6, blue long-strip color tags are respectively drawn on the flipping working end and the fixed end of the movable climbing claw 6. The specific installation positions are as Figure 4 shown; the YOLO v8 algorithm is used to identify the color tags, calculate the angle formed by the color tags, and thus obtain the inclination angle of the climbing claw 6 with respect to the reference plane. By comparing the inclination angle with the in-place state angle of the climbing claw 6, it is determined what working state the climbing claw 6 is in. When no blue tag is identified or only one blue color tag is identified, it is in the not-in-place state. When the angle satisfies the range of 0° to 85°, it is in the unsafe in-place state. When the angle satisfies the range of 85° to 95°, it is in the safe in-place state. This set angle range is for reference only and can be adjusted according to the actual situation.

[0100] In a specific embodiment, before the specific implementation, the images (images of red, green, white, and blue color tags based on the yellow background color of the tower crane) need to be collected and labeled with labelimg to train the YOLOv8n model.

[0101] The specific steps for judging the working state of the jacking crossbeam are as follows:

[0102] Video frames are collected through the camera, and the pre-trained YOLOv8 model is used to detect the targets in the image, identify the target categories (red, green, white). When both the red color tag 10 and the white color tag 11 are identified simultaneously, it is in the not-in-place state. When and only when the red color tag 10 is identified, it is in the unsafe in-place state. When and only when the green color tag 12 is identified, it is in the safe in-place state.

[0103] The specific steps for judging the working state of the climbing claw are as follows:

[0104] Step 1: Target recognition

[0105] Video frames are collected through the camera, and the pre-trained YOLOv8 model is used to detect the targets in the image, identify the target category (blue), and obtain its bounding box coordinates.

[0106] Step 2: Image Preprocessing

[0107] Input: The target area detected by YOLOv8.

[0108] Processing procedure:

[0109] 1. Adopt the grayscale conversion formula:

[0110] I gray = 0.2989R + 0.5870G + 0.1140B

[0111] Convert the color image to a grayscale image to reduce the computational complexity.

[0112] 2. Use Gaussian filtering to remove noise. Use a 5×5 Gaussian kernel to perform weighted averaging on each pixel point and its neighborhood, so that the noise can be smoothed.

[0113]

[0114] Among them, (x,y) = (5,5) represents the size of the Gaussian kernel, that is, a 5×5 neighborhood is taken for calculation; the standard deviation of the Gaussian kernel in the X direction (i.e., σ) is set to 1.0 here.

[0115] 3. Use the Canny algorithm to extract the edge information in the image to provide feature data for subsequent line detection.

[0116] Output: The edge image of the target

[0117] Step 3: Line Detection

[0118] Input: The binary image after Canny edge detection.

[0119] Processing procedure:

[0120] 1. Adopt the Hough transform to detect the lines in the image. The transformation formula is as follows:

[0121] ρ = xcosθ + ysinθ

[0122] Among them, ρ is the distance from the point to the origin, and θ is the polar angle.

[0123] 2. Determine the line that best matches the target features through a voting mechanism.

[0124] Output: The set of lines within the target area.

[0125] Step 4: Line Fitting

[0126] Input: The lines detected by the Hough transform.

[0127] Processing procedure:

[0128] 1. Fit the optimal straight line using the RANSAC algorithm, randomly sample points and calculate the straight line equation:

[0129] y = mx + b

[0130] where m is the slope and b is the intercept.

[0131] 2. Calculate the distances from all data points to the straight line:

[0132]

[0133] 3. Select the straight line with the most inliers as the final fitting result.

[0134] Output: the slope m and intercept b of the straight line.

[0135] Step 5: Angle calculation

[0136] Input: the slopes m1 and m2 of two straight lines obtained by straight line fitting.

[0137] Calculate the included angle between the two straight lines:

[0138]

[0139] Output: the included angle data between the targets.

[0140] Step 6: Kalman filter optimization

[0141] Input: the calculated included angle value.

[0142] Processing procedure: reduce the volatility of the data, and use the Kalman filter to smooth the angle data.

[0143] Output: the smoothed angle data.

[0144] Finally, judge the working state of the climbing claws through the smoothed angle data, and the final displayed result image is as Figure 5 shown.

[0145] In a specific embodiment, a pre-trained model is adopted in combination with an online data update strategy to improve the state perception ability of the camera for the safety pin 4 and the climbing claw 6, and ensure the stability and accuracy of detection.

[0146] In a specific embodiment, the industrial control computer uses a wireless transmission unit to transmit wireless signals to the ground tower crane monitoring terminal, facilitating the supervisors to monitor the tower crane jacking operation in real time.

[0147] In a specific embodiment, the high-definition camera adopts a scheme of being wiredly connected to the industrial control computer, and transmits signals and supplies power through wires.

[0148] In a specific implementation, the present application provides a computer storage medium and a corresponding data processing unit. Among them, the computer storage medium can store a computer program, and when the computer program is executed by the data processing unit, it can run the invention content of a key connection point monitoring method for tower crane jacking operation and some or all of the steps in each embodiment. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), or the like.

[0149] Those skilled in the art can clearly understand that the technical solutions in the embodiments of the present invention can be implemented by means of a computer program and its corresponding general hardware platform. Based on such an understanding, the essence of the technical solutions in the embodiments of the present invention, or the part that contributes to the prior art, can be embodied in the form of a computer program, that is, a software product. The computer program software product can be stored in the storage medium and includes several instructions to enable a device including a data processing unit (which can be a personal computer, a server, a single-chip microcomputer, an MCU, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments of the present invention.

[0150] The present invention provides an idea and method for a key connection point monitoring method for tower crane jacking operation. There are many methods and ways to specifically implement this technical solution. The above is only the preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention. Each component not clearly defined in this embodiment can be implemented by the prior art.

Claims

1. A monitoring method for key connection points during tower crane jacking operation, characterized in that, It includes the following steps: Step 1, set key connection points on the tower crane; Step 2, use the image recognition method to recognize the key connection points during the tower crane jacking operation to obtain the recognition result; Step 3, make a judgment according to the recognition result and give an alarm to complete the monitoring of the key connection points during the tower crane jacking operation.

2. The key connection point monitoring method for tower crane jacking operation according to claim 1, characterized in that The tower crane in Step 1 includes: The main limb of the standard section (1), with a step (2) fixedly arranged on the side of the main limb of the standard section (1). The groove on the step (2) is movably connected to the shaft end of the jacking cross beam (3). A safety pin (4) is movably arranged on the jacking cross beam (3) through a safety pin handle (5). The safety pin (4) passes through the pin hole on the step (2) to lock the jacking cross beam (3); The movable climbing claw (6) is installed on the main limb (7) of the jacking sleeve of the jacking sleeve, and acts on the top end of the step (2).

3. A monitoring method for key connection points during tower crane jacking operation according to claim 2, characterized in that, The setting of the key connection points in Step 1 includes: Step 1-1, set color labels on the safety pin (4), including: Set a first color label (10), a second color label (11) and a third color label (12) on the acting shaft end of the safety pin (4); Among them, when both the first color label (10) and the second color label (11) are exposed, it is the state where the safety pin (4) is not in place. When only the first color label (10) is exposed, it is the state where the safety pin (4) is not safely in place. When only the third color label (12) is exposed, it is the state where the safety pin (4) is safely in place; Step 1-2, set shape labels on the movable climbing claw (6), including: Set shape labels with linear features at the fixed end and the movable end of the movable climbing claw (6) respectively to calculate the angle between the fixed end and the movable end.

4. A monitoring method for key connection points during tower crane jacking operation according to claim 3, characterized in that The setting of the color labels on the safety pin (4) in Step 1-1 specifically includes: The first color label (10) is set at the outer end of the safety pin (4) and accounts for 1 / 4 of the area of the safety pin (4); After an interval of 1 / 4 of the area of the safety pin (4), set the second color label (11) and it accounts for 1 / 4 of the area of the safety pin (4); The third color label (12) is set at the inner end of the safety pin (4), and its area is 1 / 2 of the area of the safety pin handle (5).

5. A monitoring method for key connection points during tower crane jacking operation according to claim 3, characterized in that, The shape label in Step 1-2 is a rectangular label.

6. A monitoring method for key connection points during tower crane jacking operation according to claim 3, characterized in that, The recognition of the key connection points in Step 2 includes: Step 2-1, collect the images of the key connection points set in Step 1; Step 2-2, use the YOLO v8 method to recognize the images of the key connection points, including: Step 2-2-1, obtain the color of the label exposed on the color labels set on the safety pin (4); Step 2-2-2, obtain the linear features of the shape labels set on the movable climbing claw (6) and calculate the included angle between the fixed end and the movable end of the movable climbing claw (6).

7. A monitoring method for key connection points during the tower crane jacking operation according to claim 6, characterized in that, The judgment in Step 3 includes: Step 3-1, judge the position of the safety pin (4) according to the color of the label exposed obtained in Step 2-2-1 and give an alarm; Step 3-2: Based on the included angle obtained in Step 2-2-2, determine the angle of the movable climbing claw (6) and issue an alarm.

8. A monitoring method for key connection points during tower crane jacking operation according to claim 7, characterized in that, The determination of the position of the safety pin (4) in Step 3-1 includes: When the colors of the exposed labels include the colors of the first color label (10) and the second color label (11), it is determined that the safety pin (4) is in an out-of-position state, and an alarm is issued; When the colors of the exposed labels only include the color of the first color label (10), it is determined that the safety pin (4) is in an insecurely in-position state, and an alarm is issued; When the colors of the exposed labels only include the color of the third color label (12), it is determined that the safety pin (4) is in a securely in-position state, and a prompt is given.

9. A monitoring method for key connection points during tower crane jacking operation according to claim 8, characterized in that, The obtaining of the linear feature of the shape label provided on the movable climbing claw (6) and the calculation of the included angle between the fixed end and the movable end of the movable climbing claw (6) in Step 2-2-2 include: Step 2-2-2-1: Use grayscale conversion to convert the key connection point image corresponding to the movable climbing claw (6) into a grayscale image; Step 2-2-2-2: Use Gaussian filtering to remove the noise in the grayscale image, which is expressed as follows: where G(x,y) is the processed Gaussian weight matrix, (x,y) = (5,5), representing the size of the Gaussian kernel, that is, a 5×5 neighborhood is taken for calculation; the standard deviation of the Gaussian kernel in the X-axis direction is σ; Step 2-2-2-3: Use the Canny edge detection method to extract the edge information in the image to obtain a binary image after Canny edge detection, and the binary image is scattered points; Step 2-2-2-4: Use the Hough transform to detect the straight lines in the image, that is, transform the scattered points into straight lines; Step 2-2-2-5: Straight line fitting, fit the straight lines obtained in Step 2-2-2-4 using the RANSAC method, which specifically includes: Randomly sample points and calculate the straight line equation: y = mx + b where m is the slope and b is the intercept; Calculate the distance d from all sampled points to the straight line: Select the straight line with the most inliers as the final fitting result; Step 2-2-2-5: Calculate the angle, which specifically includes: Assume that the slopes of the two straight lines obtained by fitting in the images corresponding to the fixed end and the movable end of the movable climbing claw (6) are m1 and m2 respectively; then calculate the included angle θ between the two straight lines: Step 2-2-2-5: Perform smoothing processing on the calculated included angle value θ using the Kalman filter.

10. A monitoring method for key connection points during the jacking operation of a tower crane according to claim 9, characterized in that, The determination of the angle of the movable climbing claw (6) in Step 3-2 includes: When the included angle θ satisfies the preset range, it is determined that the state is securely in-position, otherwise it is in an insecurely in-position state, and corresponding alarms or prompts are given.

Images