Cable maintenance robot system in intelligent building

Abstract

The invention discloses a cable maintenance robot system in an intelligent building, and belongs to the technical field of cable apparent defect detection. The cable cracking degree can be represented more accurately through the total length of all the contour lines reserved in the RGB images of the cables on the two sides, and the cable bulging degree is represented more accurately by screening the abrupt change pixel points and utilizing the total number of all the abrupt change pixel points in the depth images of the cables on the two sides. The mean value of y-axis coordinate differences yc1 and yc2 between the y-axis coordinate minimum value point and the y-axis coordinate maximum value point on the center lines of the cable areas on the two sides serves as the cable distortion parameter, the cable distortion degree of the current section can be represented more accurately, and then the accurate cracking degree score, the accurate bulging degree score and the accurate distortion degree score are obtained; and finally, accurately evaluating the state of the cable in the current section through the three scores.

Description

Technical Field

[0001] The present invention relates to the technical field of cable surface defect detection, and in particular to a cable maintenance robot system in a smart building. Background Art

[0002] Smart buildings are architectural forms that leverage modern information technology, automation, and intelligent devices to efficiently manage and control various building functions, creating a safe, comfortable, convenient, and energy-efficient working and living environment. The key functional characteristics of current smart buildings are as follows: high energy efficiency, safety and comfort, convenient management, and flexible adaptability. High energy efficiency refers to the rational utilization of energy and reduced energy consumption through intelligent management and control of equipment. For example, lighting brightness is automatically adjusted based on light intensity, and lights and equipment are automatically turned off in unoccupied areas. Safety and comfort refer to the safety of people and property through comprehensive security systems, while a comfortable indoor environment (such as appropriate temperature, humidity, and air quality) improves people's quality of work and life. Convenient management means that managers can centrally monitor and manage all building equipment and functions through a central management system, gaining real-time visibility into building operations, identifying and resolving problems promptly, and improving management efficiency. Flexible adaptability means that smart building systems and equipment can be flexibly adjusted and configured based on user needs and usage scenarios to meet the personalized requirements of different users. For example, conference rooms can automatically adjust the settings of audio and projection equipment based on the type and scale of the meeting. In short, smart buildings are an inevitable trend in the development of modern buildings. They perfectly combine information technology, automation technology and construction technology, bringing people a more efficient, safe, comfortable and convenient living and working environment.

[0003] In smart buildings, cables are crucial physical carriers, used for power transmission and device connectivity, ensuring the proper functioning of various devices. Therefore, regular cable maintenance is crucial. Cable surface defects typically include insulation cracks, bulging, and cable twisting.

[0004] How to use a maintenance robot to more accurately identify apparent defects in cables and thus accurately assess the cable status is an urgent problem to be solved. To this end, the present invention proposes a cable maintenance robot system for smart buildings. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: how to more accurately identify the apparent defects of cables through a maintenance robot, and provides a cable maintenance robot system in a smart building.

[0006] The present invention solves the above technical problems through the following technical solutions, which include an image acquisition module, a cracking parameter detection module, a bulging parameter detection module, a distortion parameter detection module and a comprehensive evaluation module; The image acquisition module is used to photograph the cable in the current section from both sides of the cable, obtain the cable RGB image and cable depth image of the current section, and preprocess the cable RGB image and cable depth image to obtain the preprocessed cable RGB image and cable depth image; The cracking parameter detection module is used to detect the preprocessed cable RGB image and obtain the cable cracking parameters of the current section; The bulge parameter detection module is used to obtain the cable bulge parameters of the current section based on the preprocessed cable depth image; The distortion parameter detection module is used to obtain the cable distortion parameter of the current section based on the preprocessed cable RGB image; The comprehensive evaluation module is used to search the corresponding database according to the cable cracking parameters, cable bulging parameters and cable twisting parameters of the current section to obtain the corresponding cracking degree score, bulging degree score and twisting degree score, and then perform weighted sum calculation on the cracking degree score, bulging degree score and twisting degree score to obtain a comprehensive score as the cable status score of the current section, thereby achieving accurate evaluation of the cable status of the current section.

[0007] Furthermore, the image acquisition module includes an image capturing unit and an image preprocessing unit; the image capturing unit uses an image capturing mechanism to capture the cables in the current section to obtain an RGB image and a depth image of the cables in the current section; the image preprocessing unit is used to perform noise reduction and enhancement processing on the RGB image and the depth image of the cables; The image capture mechanism includes two RGBD cameras and a positioning connection frame. The RGBD cameras are symmetrically mounted on both sides of the positioning connection frame and on both sides of the cable. The positioning connection frame is mounted at the lower end of the robot body. In the image capture unit, a total of two pairs of cable RGB images and cable depth images are obtained, each pair of cable RGB images and cable depth images includes one side surface of the current section cable and the local surface of the surrounding devices; in the same pair of cable RGB images and cable depth images, the pixel positions of the cable RGB image and the cable depth image correspond one to one.

[0008] Furthermore, the cracking parameter detection module includes a cable identification unit, a contour detection unit and a cracking parameter acquisition unit; the cable identification unit is used to identify the cable area in the cable RGB image through a trained cable detection model, and obtain the cable area detection frame and its position information in the cable RGB image; the contour detection unit is used to detect the contour lines in the cable area detection frame, and obtain the contour lines in the cable area detection frame and their position information in the cable RGB image; the cracking parameter acquisition unit is used to obtain the cable cracking parameters of the current section based on the contour lines in the cable area detection frame and their position information in the cable RGB image.

[0009] Furthermore, in the contour detection unit, the specific processing process is as follows: S11: For the cable RGB image on one side, obtain a cable area detection frame and its position information in the cable RGB image, where the position information of the cable area detection frame in the cable RGB image is the coordinates of the upper left corner and the lower right corner of the cable area detection frame; S12: Detect the contour lines in the cable area detection frame using the contour detection function in OpenCV; S13: Obtain each contour line in the cable area detection frame and its position information in the cable RGB image. The position information of the contour line in the cable RGB image is the coordinates of each pixel point on the contour line in the cable RGB image: S14: Process the cable RGB image on the other side using the processing methods of S11 to S13, thereby obtaining the contour lines in the cable area detection frame in the cable RGB images on both sides and their position information in the cable RGB images; In the cracking parameter acquisition unit, the specific processing process is as follows: S21: Obtaining the contour lines in the cable area detection frame in the cable RGB images on both sides and their position information in the cable RGB images; S22: For the cable RGB image on one side, use the contour length threshold L0 to filter the contours, retain the contours whose length is less than the contour length threshold L0, and obtain the number of contours retained in the cable RGB image on one side, which is recorded as Cb1; S23: For the cable RGB image on the other side, similarly use the contour length threshold L0 to filter the contours, retaining contours whose lengths are less than the contour length threshold L0, and obtain the number of contours retained in the cable RGB image on the other side, which is recorded as Cb2; S24: Calculate the total length of all the remaining contour lines in the cable RGB images on both sides, recorded as Cb total , as the cable cracking parameter of the current section.

[0010] Furthermore, the bulge parameter detection module includes a bulge area detection unit and a bulge parameter acquisition unit; the bulge area detection unit is used to segment the cable area detection frame depth image from the corresponding cable depth image based on the cable area detection frame obtained by the cable identification unit and its position information in the cable RGB image, and then determine the mutation pixel point set based on the cable area detection frame depth image; the bulge parameter acquisition unit is used to obtain the cable bulge parameters of the current section based on the determined mutation pixel point set.

[0011] Furthermore, in the bulge area detection unit, the specific processing process is as follows: S31: For the cable depth image on one side, segment the corresponding cable depth image to obtain a cable area detection frame depth image based on the cable area detection frame obtained by the cable identification unit and its position information in the cable RGB image; S32: In the cable area detection frame depth image, calculate the change in the pixel values of two adjacent pixels. When the pixel value change exceeds the pixel value change threshold Pb0, include the two pixels involved in the calculation into a sudden change pixel point set, thereby obtaining a sudden change pixel point set in the cable area detection frame depth image on one side. The change in the pixel values of two adjacent pixels is the difference between the pixel values of the two adjacent pixels. S33: Processing the cable depth image on the other side using the processing methods in S31 to S32, thereby obtaining a set of sudden pixel points in the cable depth images on both sides; In the bulge parameter acquisition unit, the total number of all pixels in the sudden pixel set in the cable depth image on both sides is obtained by counting, which is recorded as P total , as the cable bulge parameter of the current section.

[0012] Furthermore, the distortion parameter detection module includes an edge contour line determination unit, a cable area center line determination unit and a distortion parameter acquisition unit; the edge contour line determination unit is used to determine the cable area edge contour line and its position information in the cable RGB image based on the contour lines in the cable area detection frame obtained by the contour detection unit and their position information in the cable RGB image; the cable area center line determination unit is used to determine the cable area center line and its position information in the cable RGB image based on the cable area edge contour line and its position information in the cable RGB image; the distortion parameter acquisition unit is used to obtain the cable distortion parameter of the current section based on the cable area center line and its position information in the cable RGB image.

[0013] Furthermore, in the edge contour line determination unit, the contour lines are screened by the contour line length threshold L0, and the contour lines with a length greater than or equal to the contour line length threshold L0 are used as the cable area edge contour lines, thereby obtaining the cable area edge contour lines in the cable RGB images on both sides and their position information in the cable RGB images; wherein, for the cable RGB images on one side, the number of cable area edge contour lines is 4, including two edge contour lines along the length direction of the cable area detection frame and two edge contour lines along the width direction of the cable area detection frame, and the length of any edge contour line along the length direction of the cable area detection frame is greater than the length of any edge contour line along the width direction of the cable area detection frame.

[0014] Furthermore, in the cable area centerline determination unit, the specific processing process is as follows: S41: Acquire the edge contour line of the cable area and its position information in the cable RGB image, where the position information of the edge contour line of the cable area in the cable RGB image is the coordinates of each pixel point on the contour line in the cable RGB image; S42: For the RGB image of the cable on one side, at the same x-axis coordinate value, calculate the midpoint coordinates of the line connecting two corresponding pixel points on two edge contour lines along the length direction of the cable area detection frame, then calculate the midpoint coordinates at all x-axis coordinate values, and connect all midpoints in sequence to form the center line of the cable area; S43: Processing the cable RGB image on the other side using the processing method in S42, thereby obtaining the cable area center line in the cable RGB images on both sides and its position information in the cable RGB image, where the position information of the cable area center line in the cable RGB image is the coordinates of each pixel point on the cable area center line in the cable RGB image; In the distortion parameter acquisition unit, the specific processing process is as follows: S51: For the RGB image of the cable on one side, obtain the minimum y-axis coordinate point and the maximum y-axis coordinate point of each pixel point on the center line of the cable area; S52: Calculate the y-axis coordinate difference between the y-axis coordinate minimum point and the y-axis coordinate maximum point; S53: Process the cable RGB image on the other side using the processing method in S51 to S52, and then obtain the y-axis coordinate differences between the minimum y-axis coordinate point and the maximum y-axis coordinate point on the center line of the cable area in the cable RGB images on both sides, which are recorded as yc1 and yc2 respectively; S54: Calculate the mean of yc1 and yc2, recorded as yc avge , as the cable twist parameter for the current segment.

[0015] Furthermore, in the comprehensive evaluation module, the comprehensive score T total The calculation formula is as follows: T total =w1*T1+w2*T2+w3*T3; Among them, T1, T2, and T3 are the cracking degree score, bulging degree score, and distortion degree score, respectively; w1, w2, and w3 are the cracking degree score, bulging degree score, and distortion degree score, respectively. total The weight ratio of .

[0016] Compared with the existing technology, the present invention has the following advantages: the cable inspection robot system in the smart building can more accurately characterize the degree of cable cracking through the total length of all retained contour lines in the RGB images of the cables on both sides, and by screening the mutation pixel points, the total number of all mutation pixel points in the depth images of the cables on both sides is used to more accurately characterize the degree of cable bulging. In addition, the average of the y-axis coordinate differences yc1 and yc2 between the minimum y-axis coordinate point and the maximum y-axis coordinate point on the center line of the cable area on both sides is used as the cable distortion parameter. The cable distortion degree of the current section can be more accurately characterized, and then an accurate cracking degree score, bulging degree score and distortion degree score can be obtained. Finally, the above three scores are used to accurately evaluate the cable status of the current section. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic block diagram of the structure of a cable maintenance robot system in a smart building according to an embodiment of the present invention; Figure 2 3 is a schematic diagram of the partial structure of the image capturing mechanism in an embodiment of the present invention (along the cable axis). DETAILED DESCRIPTION

[0018] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process. However, the protection scope of the present invention is not limited to the following embodiment.

[0019] like Figure 1 As shown, this embodiment provides a technical solution: a cable maintenance robot system in a smart building, comprising the following modules: an image acquisition module, a cracking parameter detection module, a bulging parameter detection module, a twisting parameter detection module, and a comprehensive evaluation module; In this embodiment, the image acquisition module is used to photograph the cable in the current section from both sides of the cable, obtain the cable RGB image and cable depth image of the current section, and preprocess the cable RGB image and cable depth image to obtain the preprocessed cable RGB image and cable depth image.

[0020] More specifically, when performing the detection and acquisition of the above-mentioned various parameters, a complete cable is divided into a plurality of equal-length and continuous sections, and only one section is photographed each time.

[0021] More specifically, the image acquisition module includes an image capturing unit and an image preprocessing unit; the image capturing unit utilizes an image capturing mechanism to capture the cables in the current section to obtain the cable RGB image and cable depth image of the current section; the image preprocessing unit is used to perform noise reduction and enhancement processing on the cable RGB image and cable depth image.

[0022] As more specific, Figure 2 As shown, the image capture mechanism includes two RGBD cameras 1 and a positioning connecting frame 2. The RGBD cameras 1 are symmetrically installed on both sides of the positioning connecting frame 2 and are located on both sides of the cable. The positioning connecting frame 2 is installed at the lower end of the robot body 3. The upper end of the robot body 3 is provided with multiple walking gears 31. The walking gears 31 are engaged and transmitted with the racks 42 provided on the cable outer frame 4. The driving motor provided inside the robot body 3 drives the walking gears 31 to move on the racks 42.

[0023] More specifically, in the image capturing unit, two pairs of cable RGB images and cable depth images are acquired. Each pair of cable RGB images and cable depth images includes one side surface of the cable in the current section and the local surfaces of its surrounding devices (positioning connection frame 2, cable outer frame 4). In the same pair of cable RGB images and cable depth images, the pixel positions of the cable RGB image and the cable depth image correspond one to one.

[0024] In this embodiment, the cracking parameter detection module is used to detect the preprocessed cable RGB image to obtain the cable cracking parameter of the current section.

[0025] More specifically, the cracking parameter detection module includes a cable identification unit, a contour detection unit and a cracking parameter acquisition unit; the cable identification unit is used to identify the cable area in the cable RGB image through a trained cable detection model, and obtain the cable area detection frame and its position information in the cable RGB image; the contour detection unit is used to detect the contour lines in the cable area detection frame, and obtain the contour lines in the cable area detection frame and their position information in the cable RGB image; the cracking parameter acquisition unit is used to obtain the cable cracking parameters of the current section based on the contour lines in the cable area detection frame and their position information in the cable RGB image.

[0026] More specifically, in the cable identification unit, the cable detection model is obtained by training the SSD target detection network.

[0027] More specifically, in the contour detection unit, the specific processing process is as follows: Step 1: For the cable RGB image on one side, obtain the cable area detection frame and its position information in the cable RGB image. The position information of the cable area detection frame in the cable RGB image is the coordinates of the upper left corner and lower right corner of the cable area detection frame; Step 2: Use the contour detection function in OpenCV to detect the contour lines in the cable area detection frame; Step 3: Obtain the contour lines in the cable area detection frame and their position information in the cable RGB image. The position information of the contour lines in the cable RGB image is the coordinates of each pixel point on the contour line in the cable RGB image: Step 4: Process the cable RGB image on the other side using the processing methods of steps 1 to 3 above, and then obtain the contour lines in the cable area detection frame in the cable RGB images on both sides and their position information in the cable RGB images.

[0028] More specifically, in the cracking parameter acquisition unit, the specific processing process is as follows: Step 1: Obtain the contour lines in the cable area detection frame in the cable RGB images on both sides and their position information in the cable RGB images; Step 2: For the cable RGB image on one side, use the contour length threshold L0 to filter the contours, and retain the contours whose length is less than the contour length threshold L0. The number of contours retained in the cable RGB image on one side is obtained and recorded as Cb1. Step 3: For the cable RGB image on the other side, similarly use the contour length threshold L0 to filter the contours, retaining contours whose length is less than the contour length threshold L0. The number of contours retained in the cable RGB image on the other side is obtained, which is recorded as Cb2. Step 4: Calculate the total length of all the remaining contour lines in the cable RGB images on both sides, denoted as Cb total , as the cable cracking parameter of the current section.

[0029] It should be noted that in step 2 and step 3, the contour line length threshold L0 is used to remove the edge contour lines of the cable area in the cable area detection frame from the total number of contour lines to ensure the accuracy of subsequent cable cracking parameters.

[0030] In the present invention, the total length of all retained contour lines in the RGB images of the cables on both sides is used to more accurately characterize the degree of cable cracking, thereby facilitating subsequent comprehensive evaluation of the cable status in the current section.

[0031] In this embodiment, the bulge parameter detection module is used to obtain the cable bulge parameters of the current section according to the preprocessed cable depth image.

[0032] To be more specific, the bulge parameter detection module includes a bulge area detection unit and a bulge parameter acquisition unit; the bulge area detection unit is used to segment the cable area detection frame depth image from the corresponding cable depth image based on the cable area detection frame obtained by the cable identification unit and its position information in the cable RGB image, and then determine the mutation pixel point set based on the cable area detection frame depth image; the bulge parameter acquisition unit is used to obtain the cable bulge parameters of the current section based on the determined mutation pixel point set.

[0033] More specifically, in the bulge area detection unit, the specific processing process is as follows: Step 1: For the cable depth image on one side, based on the cable area detection frame obtained by the cable recognition unit and its position information in the cable RGB image, the cable area detection frame depth image is segmented from the corresponding cable depth image; Step 2: In the cable area detection frame depth image, calculate the change in the pixel values of two adjacent pixels. When the pixel value change exceeds the pixel value change threshold Pb0, the two pixels involved in the calculation are included in the mutation pixel set, thereby obtaining the mutation pixel set in the cable area detection frame depth image on one side. The change in the pixel values of two adjacent pixels is the difference between the pixel values of the two adjacent pixels. Step 3: Process the cable depth image on the other side using the processing methods in steps 1 and 2 above, and then obtain a set of sudden pixel points in the cable depth images on both sides.

[0034] More specifically, in the bulge parameter acquisition unit, the total number of all pixels in the sudden pixel set in the cable depth image on both sides is obtained by counting, which is recorded as P total , as the cable bulge parameter of the current section.

[0035] In the present invention, by screening the mutation pixels, the total number of all mutation pixels in the cable depth images on both sides is used to more accurately characterize the cable bulging degree, which facilitates the subsequent comprehensive evaluation of the cable status of the current section.

[0036] In this embodiment, the distortion parameter detection module is used to obtain the cable distortion parameter of the current section according to the pre-processed cable RGB image.

[0037] More specifically, the distortion parameter detection module includes an edge contour line determination unit, a cable area center line determination unit and a distortion parameter acquisition unit; the edge contour line determination unit is used to determine the cable area edge contour line and its position information in the cable RGB image based on the contour lines in the cable area detection frame obtained by the contour detection unit and their position information in the cable RGB image; the cable area center line determination unit is used to determine the cable area center line and its position information in the cable RGB image based on the cable area edge contour line and its position information in the cable RGB image; the distortion parameter acquisition unit is used to obtain the cable distortion parameter of the current section based on the cable area center line and its position information in the cable RGB image.

[0038] More specifically, in the edge contour line determination unit, the contour lines are screened by the contour line length threshold L0, and the contour lines with a length greater than or equal to the contour line length threshold L0 are used as the cable area edge contour lines, thereby obtaining the cable area edge contour lines in the cable RGB images on both sides and their position information in the cable RGB images; wherein, for the cable RGB images on one side, the number of cable area edge contour lines is 4, including two edge contour lines along the length direction of the cable area detection frame and two edge contour lines along the width direction of the cable area detection frame, and the length of any edge contour line along the length direction of the cable area detection frame is greater than the length of any edge contour line along the width direction of the cable area detection frame.

[0039] More specifically, in the cable area centerline determination unit, the specific processing process is as follows: Step 1: Obtain the edge contour line of the cable area and its position information in the cable RGB image. The position information of the edge contour line of the cable area in the cable RGB image is the coordinates of each pixel point on the contour line in the cable RGB image. Step 2: For the RGB image of one side of the cable, at the same x-axis coordinate value, calculate the midpoint coordinates of the line connecting the two corresponding pixels on the two edge contour lines along the length direction of the cable area detection frame. Then calculate the midpoint coordinates at all x-axis coordinate values and connect all the midpoints in sequence to form the center line of the cable area. Step 3: Process the cable RGB image on the other side using the processing method in step 2 above, and then obtain the cable area center line in the cable RGB images on both sides and its position information in the cable RGB image. The position information of the cable area center line in the cable RGB image is the coordinates of each pixel point on the cable area center line in the cable RGB image.

[0040] It should be noted that, in the cable RGB image, the x-axis is set along the length direction of the cable area detection frame, and the y-axis is set along the width direction of the cable area detection frame.

[0041] More specifically, in the distortion parameter acquisition unit, the specific processing process is as follows: Step 1: For the RGB image of the cable on one side, obtain the minimum and maximum y-axis coordinate points of each pixel on the center line of the cable area; Step 2: Calculate the y-axis coordinate difference between the minimum y-axis coordinate point and the maximum y-axis coordinate point; Step 3: Process the cable RGB image on the other side using the same processing methods as in steps 1 and 2 above, and obtain the y-axis coordinate differences between the minimum y-axis coordinate point and the maximum y-axis coordinate point on the center line of the cable area in the RGB images on both sides, which are recorded as yc1 and yc2 respectively. Step 4: Calculate the mean of yc1 and yc2, denoted as yc avge , as the cable twist parameter for the current segment.

[0042] In the present invention, by taking the average of the y-axis coordinate differences yc1 and yc2 between the minimum y-axis coordinate point and the maximum y-axis coordinate point on the center lines of the cable areas on both sides as the cable twist parameter, the cable twist degree of the current section can be more accurately characterized, facilitating the subsequent comprehensive evaluation of the cable status of the current section.

[0043] In this embodiment, the comprehensive assessment module is used to search the corresponding database based on the cable cracking parameters, cable bulging parameters, and cable twisting parameters of the current section to obtain the corresponding cracking degree score, bulging degree score, and twisting degree score. The module then performs a weighted sum calculation on the cracking degree score, bulging degree score, and twisting degree score to obtain a comprehensive score as the cable status score for the current section. This allows for an accurate assessment of the cable status of the current section.

[0044] More specifically, in the comprehensive evaluation module, the comprehensive score T total The calculation formula is as follows: T total =w1*T1+w2*T2+w3*T3; Among them, T1, T2, and T3 are the cracking degree score, bulging degree score, and distortion degree score, respectively; w1, w2, and w3 are the cracking degree score, bulging degree score, and distortion degree score, respectively. total The weight ratio of .

[0045] It should be noted that the cracking degree score is used to calculate the cable cracking parameter Cb of the current section. totalThe cracking degree score database is searched and obtained. The cracking degree score database stores the correspondence between the cable cracking parameter value and the cracking degree score; the bulging degree score is used to calculate the cable bulging parameter P of the current section. total The cable bulge degree score is searched and obtained in the bulge degree score database, which stores the corresponding relationship between the cable bulge parameter value and the bulge degree score; the twist degree score is used to calculate the cable twist parameter yc of the current section. avge The twisting degree score database is searched and obtained, and the twisting degree score database stores the corresponding relationship between the cable twisting parameter value and the twisting degree score; the cracking degree score database, the bulging degree score database, and the twisting degree score database are all pre-established by experts.

[0046] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A cable maintenance robot system in a smart building, characterized in that: include: Image acquisition module, cracking parameter detection module, bulging parameter detection module, distortion parameter detection module and comprehensive evaluation module; The image acquisition module is used to photograph the cable in the current section from both sides of the cable, obtain the cable RGB image and cable depth image of the current section, and preprocess the cable RGB image and cable depth image to obtain the preprocessed cable RGB image and cable depth image; The cracking parameter detection module is used to detect the preprocessed cable RGB image and obtain the cable cracking parameters of the current section; The bulge parameter detection module is used to obtain the cable bulge parameters of the current section based on the preprocessed cable depth image; The distortion parameter detection module is used to obtain the cable distortion parameter of the current section based on the preprocessed cable RGB image; The comprehensive evaluation module is used to search the corresponding database according to the cable cracking parameters, cable bulging parameters and cable twisting parameters of the current section to obtain the corresponding cracking degree score, bulging degree score and twisting degree score, and then perform weighted sum calculation on the cracking degree score, bulging degree score and twisting degree score to obtain a comprehensive score as the cable status score of the current section, thereby achieving accurate evaluation of the cable status of the current section.

2. A cable maintenance robot system in a smart building according to claim 1, characterized in that: The image acquisition module includes an image capturing unit and an image preprocessing unit; the image capturing unit uses an image capturing mechanism to capture the cables in the current section to obtain an RGB image and a depth image of the cables in the current section; The image preprocessing unit is used to perform noise reduction and enhancement processing on the cable RGB image and the cable depth image; The image capture mechanism includes two RGBD cameras and a positioning connection frame. The RGBD cameras are symmetrically mounted on both sides of the positioning connection frame and on both sides of the cable. The positioning connection frame is mounted at the lower end of the robot body. In the image capture unit, a total of two pairs of cable RGB images and cable depth images are obtained, each pair of cable RGB images and cable depth images includes one side surface of the current section cable and the local surface of the surrounding devices; in the same pair of cable RGB images and cable depth images, the pixel positions of the cable RGB image and the cable depth image correspond one to one.

3. The cable maintenance robot system in a smart building according to claim 1, characterized in that: The cracking parameter detection module includes a cable identification unit, a contour detection unit, and a cracking parameter acquisition unit; the cable identification unit is used to identify the cable area in the cable RGB image using a trained cable detection model, and obtain the cable area detection frame and its position information in the cable RGB image; The contour detection unit is used to detect the contour lines in the cable area detection frame and obtain the position information of each contour line in the cable area detection frame and its position information in the cable RGB image; The cracking parameter acquisition unit is used to acquire the cable cracking parameter of the current section according to the position information of each contour line in the cable area detection frame and its position information in the cable RGB image.

4. A cable maintenance robot system in a smart building according to claim 3, characterized in that: In the contour detection unit, the specific processing process is as follows: S11: For the cable RGB image on one side, obtain a cable area detection frame and its position information in the cable RGB image, where the position information of the cable area detection frame in the cable RGB image is the coordinates of the upper left corner and the lower right corner of the cable area detection frame; S12: Detect the contour lines in the cable area detection frame using the contour detection function in OpenCV; S13: Obtain each contour line in the cable area detection frame and its position information in the cable RGB image. The position information of the contour line in the cable RGB image is the coordinates of each pixel point on the contour line in the cable RGB image: S14: Process the cable RGB image on the other side using the processing methods of S11 to S13, thereby obtaining the contour lines in the cable area detection frame in the cable RGB images on both sides and their position information in the cable RGB images; In the cracking parameter acquisition unit, the specific processing process is as follows: S21: Obtaining the contour lines in the cable area detection frame in the cable RGB images on both sides and their position information in the cable RGB images; S22: For the cable RGB image on one side, use the contour length threshold L0 to filter the contours, retain the contours whose length is less than the contour length threshold L0, and obtain the number of contours retained in the cable RGB image on one side, which is recorded as Cb1; S23: For the cable RGB image on the other side, similarly use the contour length threshold L0 to filter the contours, retaining contours whose lengths are less than the contour length threshold L0, and obtain the number of contours retained in the cable RGB image on the other side, which is recorded as Cb2; S24: Calculate the total length of all the remaining contour lines in the cable RGB images on both sides, recorded as Cb total , as the cable cracking parameter of the current section.

5. A cable maintenance robot system in a smart building according to claim 4, characterized in that: The bulge parameter detection module includes a bulge area detection unit and a bulge parameter acquisition unit; the bulge area detection unit is used to segment the cable area detection frame obtained by the cable identification unit and its position information in the cable RGB image from the corresponding cable depth image to obtain a cable area detection frame depth image, and then determine a sudden pixel point set based on the cable area detection frame depth image; The bulge parameter acquisition unit is used to acquire the cable bulge parameter of the current section according to the determined mutation pixel point set.

6. A cable maintenance robot system in a smart building according to claim 5, characterized in that: In the bulge area detection unit, the specific processing process is as follows: S31: For the cable depth image on one side, segment the corresponding cable depth image to obtain a cable area detection frame depth image based on the cable area detection frame obtained by the cable identification unit and its position information in the cable RGB image; S32: In the cable area detection frame depth image, calculate the change in the pixel values of two adjacent pixels. When the pixel value change exceeds the pixel value change threshold Pb0, include the two pixels involved in the calculation into a sudden change pixel point set, thereby obtaining a sudden change pixel point set in the cable area detection frame depth image on one side. The change in the pixel values of two adjacent pixels is the difference between the pixel values of the two adjacent pixels. S33: Processing the cable depth image on the other side using the processing methods in S31 to S32, thereby obtaining a set of sudden pixel points in the cable depth images on both sides; In the bulge parameter acquisition unit, the total number of all pixels in the sudden pixel set in the cable depth image on both sides is obtained by counting, which is recorded as P total , as the cable bulge parameter of the current section.

7. A cable maintenance robot system in a smart building according to claim 6, characterized in that: The distortion parameter detection module includes an edge contour line determination unit, a cable area center line determination unit, and a distortion parameter acquisition unit; the edge contour line determination unit is used to determine the edge contour line of the cable area and its position information in the cable RGB image based on the contour lines in the cable area detection frame obtained by the contour detection unit and their position information in the cable RGB image; The cable area centerline determination unit is used to determine the cable area centerline and its position information in the cable RGB image based on the cable area edge contour line and its position information in the cable RGB image; The distortion parameter acquisition unit is used to acquire the cable distortion parameter of the current section according to the cable area center line and its position information in the cable RGB image.

8. The cable maintenance robot system in a smart building according to claim 7, characterized in that: In the edge contour line determination unit, the contour lines are screened by the contour line length threshold L0, and the contour lines with a length greater than or equal to the contour line length threshold L0 are used as the cable area edge contour lines, thereby obtaining the cable area edge contour lines in the cable RGB images on both sides and their position information in the cable RGB images; wherein, for the cable RGB images on one side, the number of cable area edge contour lines is 4, including two edge contour lines along the length direction of the cable area detection frame and two edge contour lines along the width direction of the cable area detection frame, and the length of any edge contour line along the length direction of the cable area detection frame is greater than the length of any edge contour line along the width direction of the cable area detection frame.

9. A cable maintenance robot system in a smart building according to claim 8, characterized in that: In the cable area centerline determination unit, the specific processing process is as follows: S41: Acquire the edge contour line of the cable area and its position information in the cable RGB image, where the position information of the edge contour line of the cable area in the cable RGB image is the coordinates of each pixel point on the contour line in the cable RGB image; S42: For the RGB image of the cable on one side, at the same x-axis coordinate value, calculate the midpoint coordinates of the line connecting two corresponding pixel points on two edge contour lines along the length direction of the cable area detection frame, then calculate the midpoint coordinates at all x-axis coordinate values, and connect all midpoints in sequence to form the center line of the cable area; S43: Processing the cable RGB image on the other side using the processing method in S42, thereby obtaining the cable area center line in the cable RGB images on both sides and its position information in the cable RGB image, where the position information of the cable area center line in the cable RGB image is the coordinates of each pixel point on the cable area center line in the cable RGB image; In the distortion parameter acquisition unit, the specific processing process is as follows: S51: For the RGB image of the cable on one side, obtain the minimum y-axis coordinate point and the maximum y-axis coordinate point of each pixel point on the center line of the cable area; S52: Calculate the y-axis coordinate difference between the y-axis coordinate minimum point and the y-axis coordinate maximum point; S53: Process the cable RGB image on the other side using the processing method in S51 to S52, and then obtain the y-axis coordinate differences between the minimum y-axis coordinate point and the maximum y-axis coordinate point on the center line of the cable area in the cable RGB images on both sides, which are recorded as yc1 and yc2 respectively; S54: Calculate the mean of yc1 and yc2, recorded as yc avge , as the cable twist parameter for the current segment.

10. A cable maintenance robot system in a smart building according to claim 1 or 9, characterized in that: In the comprehensive evaluation module, the comprehensive score T total The calculation formula is as follows: <h2 style=";text-align:left;direction:ltr">T<h2 style=";text-align:left;direction:ltr"> total <h2 style=";text-align:left;direction:ltr"> =w1*T1+w2*T2+w3*T3; Among them, T1, T2, and T3 are the cracking degree score, bulging degree score, and distortion degree score, respectively; w1, w2, and w3 are the cracking degree score, bulging degree score, and distortion degree score, respectively. total The weight ratio of .

Images