Snowfield cableway autonomous inspection system and method based on unmanned aerial vehicle

Abstract

The invention is suitable for the technical field of snow yard cableway inspection, and provides a snow yard cableway autonomous inspection system and method based on an unmanned aerial vehicle, and the system comprises an inspection management system, a data collection module, a data storage module and a fault analysis module. Image data of a cableway support and a cable are accurately and effectively acquired through a data acquisition module and an inspection management system; the fault analysis module is used for carrying out preprocessing, classification and fault detection on the image data, fault diagnosis of all parts of the cableway is achieved, and results are stored in the historical data storage unit. The inspection management module can control the route of the unmanned aerial vehicle and make an inspection plan, thereby achieving the autonomous inspection of the cableway. The high flexibility of the unmanned aerial vehicle and the high precision of the machine vision technology are utilized to automatically execute the cableway inspection task, so that the labor cost is reduced, the inspection efficiency and accuracy are improved, the safety risk of operation and maintenance personnel is reduced, and the ice and snow industry is promoted to develop towards the intelligent direction.

Description

Technical Field

[0001] The present invention belongs to the technical field of inspection of ski resort ropeways, and in particular relates to an autonomous inspection system and method of ski resort ropeways based on unmanned aerial vehicles. Background Art

[0002] In recent years, China's ice and snow industry has ushered in a golden period of development, driving the rapid development of related industries such as ice and snow tourism and ice and snow events. However, the surrounding industries of ice and snow are still in their infancy, and safety accidents in ice and snow sports occur frequently. The safety inspection of facilities and equipment in ski resorts still needs to be solved. As more and more tourists participate in ice and snow sports, ski resort cableways face new challenges. Long hours of work not only accelerate the wear and aging of equipment, but also the severe low temperature and strong winds increase the difficulty of cableway inspection.

[0003] At present, the inspection of ski resort cableways mainly relies on manual inspections, that is, the operation and maintenance personnel take the cable car to perform fault detection and diagnosis on the various components of the cableway. However, manual inspection consumes a lot of manpower, the accuracy and efficiency of the inspection depend on the experience of the operation and maintenance personnel, and manual inspection has certain safety hazards. In addition, manual inspection has a certain lag and cannot detect faults that occur during cableway operation in a timely and effective manner.

[0004] With the rapid progress of drone technology and machine vision technology in the past two decades, combining the two for inspection of ski resort ropeways has become a potential solution. Drones collect image data of the target area through cameras, and machine vision technology detects and identifies the collected image data. The combination of the two can effectively solve similar low-altitude inspection problems. Drone inspection technology has shown great potential and application prospects in the fields of power inspection and wind power inspection. However, in the field of ropeway inspection, related technologies and applications are still in a blank stage. In order to meet these challenges, it is particularly important to develop an autonomous inspection system for ski resort ropeways based on drones. Summary of the invention

[0005] The purpose of the present invention is to provide a ski resort cableway autonomous inspection system and method based on drones, aiming to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for autonomous inspection of a ski resort cableway based on a drone comprises the following steps:

[0008] Step 1: The meteorological monitoring unit monitors the meteorological conditions of the current environment of the ski resort cableway and determines whether the current meteorological conditions are suitable for performing the inspection task; if it is suitable for performing the inspection task, the drone nest unit opens the nest door and sends the inspection task; if it is not suitable for performing the inspection task, the inspection task is suspended and waits for suitable weather;

[0009] Step 2: After the drone starts the inspection, it uses the high-resolution camera and infrared camera on board to collect cableway image data, and uses the infrared ranging sensor on board to measure the distance between the drone and the cableway support;

[0010] Step 3: The route control unit receives the real-time location information of the UAV and determines whether the distance between the UAV and the cableway support is greater than two meters; if the distance is greater than or equal to two meters, the overall inspection route is executed; if the distance is less than two meters, the local inspection route is executed;

[0011] When executing the local inspection route, determine whether the cableway bracket is a vulnerable bracket; if the cableway bracket is a low-vulnerability bracket, the drone executes the low-vulnerability route; if the cableway bracket is a high-vulnerability bracket, the drone executes the high-vulnerability route;

[0012] Step 4: Determine the image data location information according to the cableway bracket number in the route;

[0013] Step 5: After completing the inspection task, the drone returns to the nest;

[0014] Step 6: The fault analysis module pre-processes the collected image data to obtain a cableway image data set, and performs different fault detection methods for different parts;

[0015] For the bracket wheel set, wheel set wear fault detection is adopted;

[0016] Bolt connection fault detection is used for the connection of the bracket bolt group;

[0017] For ropeway cables, cable foreign body detection is used;

[0018] Step 7: The fault analysis module sends the detection results to the fault summary unit, which displays the faults of this inspection, summarizes the fault report, and sends it to the operation and maintenance personnel;

[0019] Step 8: Save the inspection image of this time into the historical data storage unit, and save the fault image into the fault data storage unit;

[0020] Step 9: The data storage module uses the bracket fragility level determination formula to determine the fragility of the brackets inspected this time;

[0021] Step 10: The drone nest unit charges and performs maintenance operations on the drone;

[0022] The determination formula of the support fragility level is:

[0023] S = 0.5 × P + 0.3 × M + 0.2 × H;

[0024] Among them, S represents the vulnerability level score, with a full score of 10 points; 0-6 points represent low vulnerability levels, and 7-10 points represent high vulnerability levels; P represents the number of failures of the bracket within a year, with a full score of 10 points. The more failures there are, the higher the score; M represents the usage time of the bracket when the failure occurs, with a full score of 10 points. The longer the usage time, the higher the score; H represents the environmental conditions of the bracket, including temperature and humidity, with a full score of 10 points. The worse the environment, the higher the score.

[0025] Furthermore, the specific process of the step of determining the image data location information according to the cableway bracket number in the route is as follows:

[0026] The entire route is divided into multiple sections using the cableway bracket as a node;

[0027] The cables between two adjacent brackets are numbered in sections according to the existing numbers of the two adjacent brackets;

[0028] The high-resolution camera on the drone is used to obtain the bracket number, and the bracket position is matched to the number;

[0029] After the image data of the current bracket is collected, the drone flies along the cable to the next bracket and obtains its number, and determines the number of the cable section according to the numbers of the two adjacent brackets; and the image data taken by the drone are digitally marked in sequence;

[0030] The position mark is combined with the serial number to form the position information corresponding to each image data.

[0031] Furthermore, the process of preprocessing the collected image data by the fault analysis module is as follows:

[0032] Use image hashing to remove similar images;

[0033] Gaussian filter is used to denoise the image data;

[0034] Histogram equalization is used to enhance the image data to improve the contrast and brightness of the image;

[0035] Correct image distortion caused by the UAV’s flight posture;

[0036] The preprocessed image data is constructed into a cableway image dataset;

[0037] Threshold segmentation technology is used on the cableway image dataset to separate the bracket from the background, and the ratio of the number of bracket pixels to the total pixels of the image is calculated; if the ratio is greater than or equal to 70%, the image is classified as the bracket subset; if the ratio is less than 70%, the image is classified as the cable subset; the bracket subset is used to detect wheel wear faults and bolt connection faults, and the cable subset is used to detect cable foreign body faults.

[0038] Furthermore, the steps of wheel set wear fault detection are as follows:

[0039] The pixel value of the wheel set infrared image is converted into an actual temperature value by a preset temperature conversion formula to obtain the actual temperature of the wheel set;

[0040] Calculate the average temperature value of all pixels in the wheel group area to obtain the average temperature of all wheels;

[0041] According to the temperature range of normal operation of the wheel set, a temperature threshold is pre-set to judge the wear condition of the wheel set;

[0042] Calculate the standard deviation of all wheel group temperatures to determine the temperature distribution;

[0043] Combining the average wheel temperature, standard deviation, and set temperature threshold, the temperature of each wheel is compared:

[0044] If the temperature of a wheel set exceeds the set temperature threshold, or is higher than the average temperature of the wheel set plus one standard deviation, it is determined that the wheel set has abnormal wear;

[0045] If the temperature of a wheel set shows abnormal wear for five or more days during a week of routine inspection, the wheel set should be repaired or replaced;

[0046] The detection results and image data are stored in the data storage module, including the location of wheel wear and the detection time.

[0047] Furthermore, the steps of bolt connection fault detection are as follows:

[0048] The YOLOv8 model is trained using a dataset of labeled images of bolt groups, nuts, and bolts.

[0049] The image data in the bracket subset is input into the trained YOLOv8 model to perform object detection, identify and locate the nuts and bolts in the image;

[0050] Mark the detected nuts and bolts and output the test results, including the number, location and confidence of nuts and bolts;

[0051] According to the detection results of the target detection algorithm, determine whether there is a missing nut or bolt;

[0052] Comparing the number of nuts and bolts detected with a predetermined standard number, if the number of bolts or nuts in the detection result is less than the predetermined number, it is determined that the bolt group has a connection fault;

[0053] Verify the detected nut and bolt positions. If the nut and bolt positions in the detection result do not match the predetermined positions, it is determined to be a connection failure.

[0054] For bolt groups with missing nuts or bolts, the specific location and fault type are recorded and a fault alarm is issued;

[0055] The detection results and image data are stored in the data storage module, including the location of the faulty bolt group, the fault type and the detection time.

[0056] Furthermore, the steps of detecting foreign matter in the cable are as follows:

[0057] Use the EDlines algorithm to perform edge detection on the images of the cable subset, extract the edge features of the cables, and obtain the first feature map;

[0058] Use the LSD algorithm to perform line segment detection on the first feature map, extract the line segment features of the cable, and obtain the second feature map;

[0059] Fitting the line segment detection result of the second feature map to construct a cable model;

[0060] Based on the cable model, the currently acquired cable image is subjected to difference calculation with the cable model to obtain a difference image, and the difference image is subjected to threshold segmentation to extract the area with the largest difference as the foreground area;

[0061] Extract the color, shape and texture features of the foreground area and construct a feature vector;

[0062] Support vector machine is used to classify the extracted features and determine whether the foreground area is a foreign object;

[0063] If the classification result is a foreign object, the foreground area is determined to be a foreign object on the cable;

[0064] Record the specific location and detection time of foreign matter, and issue a fault alarm;

[0065] The detection results and image data are stored in the data storage module, including the location, type and detection time of the foreign matter in the cable.

[0066] A ski resort cableway autonomous inspection system based on drones, used to implement the above-mentioned ski resort cableway autonomous inspection method based on drones, comprising:

[0067] Inspection management system, used to monitor the ski resort environment and weather conditions; receive and display inspection images sent back by drones, receive and display the real-time location information of drones and their flight trajectory; determine the relationship between the current location of the drone and the position of the bracket, and perform different inspection methods according to the vulnerability level of the bracket; display the inspection results, including fault images and fault location information; control the opening and closing of the machine nest door when the drone departs and returns;

[0068] The data acquisition module is used to control the UAV to obtain the relevant image data of each cableway component during the inspection process and the location information when obtaining it according to the specified route; wherein the location information of the image data is determined according to the existing number on each cableway bracket in the route;

[0069] A data storage module is used to store all inspection image data within a month and all fault image data within a year; wherein the fault image data is used to determine the fragility level of the bracket;

[0070] The fault analysis module is used to pre-process the image data, generate a cableway image data set, and divide the cableway image data set into a bracket subset and a cable subset according to the proportion of the cableway components in the image screen; wherein the bracket subset is used to detect wheel wear faults and bolt connection faults, and the cable subset is used to detect cable foreign body faults; the fault analysis module performs wheel wear fault detection, bolt connection fault detection and cable foreign body detection.

[0071] Furthermore, the data acquisition module includes:

[0072] The drone is used to fly along the designated route and send the collected image data, infrared image data and location information to the inspection management system; the designated route is divided into an overall inspection route and a local inspection route; among them, the overall inspection route is suitable for flying along the cableway cable, and the local inspection route is suitable for inspecting the cableway bracket;

[0073] A high-resolution camera, fixed to the front of the drone frame, is used to capture high-definition image data of the cableway;

[0074] An infrared camera, fixed under the UAV frame, is used to capture infrared image data of the cableway;

[0075] The infrared ranging sensor is fixed below the infrared camera and is used to measure the distance between the drone and the cableway support.

[0076] Furthermore, the data storage module includes:

[0077] A historical data storage unit is used to store all inspection image data within one month;

[0078] A fault data storage unit is used to store all fault image data, fault types, fault locations and fault times within one year;

[0079] The data retrieval unit is used to retrieve the stored image data according to the fault type, fault location or fault time.

[0080] Furthermore, the inspection management system includes:

[0081] Meteorological monitoring unit, including temperature sensor, humidity sensor and TENG wind speed sensor, is used to collect meteorological data of the environment where the cableway is located in real time;

[0082] Inspection image display unit, used to receive and display the inspection images sent back by the drone;

[0083] The route control unit is used to receive and display the real-time location information of the UAV and its flight route, determine the relationship between the current position of the UAV and the position of the bracket, and perform different inspection methods according to the vulnerability level of the bracket;

[0084] The fault summary unit is used to display the fault results of this inspection, including the fault image and fault location information, and send them to the data storage module;

[0085] The drone nest unit is used to control the opening and closing of the nest door when the drone takes off and returns.

[0086] Compared with the prior art, the present invention has the following beneficial effects:

[0087] 1. The present invention uses drones to automatically and accurately collect image data of cableway supports and cables, and combines machine vision technology to perform real-time fault detection and diagnosis. This innovative method significantly improves the efficiency and accuracy of inspections, reduces errors and omissions caused by manual inspections, and also reduces labor costs and safety risks.

[0088] 2. The present invention can pre-process and classify the acquired image data through the fault analysis module, perform detailed fault detection and analysis on various components of the cableway, and save the results in the data storage module, providing detailed fault information for operation and maintenance personnel, and helping to formulate more effective maintenance plans and strategies through historical data analysis.

[0089] 3. The present invention can intelligently formulate and adjust the inspection plan according to the real-time environment and meteorological data through the inspection management system, and accurately control the route of the drone. The system performs different inspection methods according to the vulnerability level of the cableway support to ensure that key parts are fully inspected, thereby improving the pertinence and effectiveness of the inspection work.

[0090] 4. The present invention can not only detect wheel wear and bolt connection failure of the cableway support, but also identify foreign objects on the cable. Through a variety of detection methods, the fault detection of various cableway components is fully covered, thereby improving the reliability and safety of the cableway operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 It is a system block diagram of the present invention.

[0092] Figure 2 This is a local inspection route map of the UAV in the present invention.

[0093] Figure 3 It is a flow chart of bolt connection fault detection in the present invention.

[0094] Figure 4 This is a flow chart of wheel set wear fault detection in the present invention.

[0095] Figure 5 The figure is a flow chart of cable foreign body detection in the present invention.

[0096] Figure 6 The figure is a flow chart of the method of the present invention.

[0097] In the figure: inspection management system 100, meteorological monitoring unit 101, inspection screen display unit 102, route control unit 103, fault summary unit 104, drone nest unit 105, data acquisition module 200, drone 201, high-resolution camera 202, infrared camera 203, infrared ranging sensor 204, data storage module 300, historical data storage unit 301, fault data storage unit 302, data retrieval unit 303, fault analysis module 400, wheel wear fault detection 401, bolt connection fault detection 402, cable foreign body detection 403. DETAILED DESCRIPTION

[0098] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0099] like Figure 1 As shown, an autonomous inspection system for a ski resort cableway based on a drone is provided as an embodiment of the present invention, comprising:

[0100] The inspection management system 100 is used to monitor the snow field environment and weather conditions; receive and display the inspection images sent back by the drone 201, receive and display the real-time location information and flight trajectory of the drone 201; determine the position relationship between the current position of the drone 201 and the bracket, and perform different inspection methods according to the vulnerability level of the bracket; display the inspection results, including fault images and fault location information; control the opening and closing of the machine nest door when the drone 201 departs and returns;

[0101] The data acquisition module 200 is used to control the drone 201 to obtain the relevant image data of each cableway component during the inspection process and the location information when obtaining it according to the specified route; wherein the location information of the image data is determined according to the existing number on each cableway bracket in the route, specifically:

[0102] Taking the cableway bracket as a node, the entire route is divided into multiple sections; according to the existing numbers of two adjacent brackets, the cables between the two adjacent brackets are segmented and numbered; the bracket number is obtained through the high-resolution camera 202 carried by the drone 201, and the position of the bracket is matched with the number; after the image data of the current bracket is collected, the drone 201 flies along the cable to the next bracket and obtains its number, and determines the number of the cable section according to the numbers of the two adjacent brackets; and the image data taken by the drone 201 are digitally marked in sequence in turn to obtain the corresponding serial number of each image data; the position mark is combined with the serial number to form the position information corresponding to each image data.

[0103] If the pre-set multiple serial numbers are T001 and T002, the serial number of the first acquired high-definition image and / or infrared image is recorded as T001; if a section of cable corresponds to two adjacent brackets with bracket numbers Z003 and Z004 respectively, the number of the section of cable is Z003-Z004, and the position of the bracket is marked as Z003Z004. At the same time, if the serial number of the high-definition image and / or infrared image is T0045, the position information of the image is Z003Z004T0045.

[0104] The data storage module 300 is used to store the inspection image data of a month and the fault image data of a year; wherein the fault image data is used to determine the fragility level of the bracket;

[0105] The fault analysis module 400 is used to pre-process the image data, generate a cableway image data set, and divide the cableway image data set into a bracket subset and a cable subset according to the proportion of the cableway components in the image screen; wherein the bracket subset is used to detect wheel wear faults and bolt connection faults, and the cable subset is used to detect cable foreign body faults; the fault analysis module 400 performs wheel wear fault detection 401, bolt connection fault detection 402 and cable foreign body detection 403.

[0106] The inspection management system 100 includes:

[0107] The meteorological monitoring unit 101 includes a temperature sensor, a humidity sensor, and a TENG wind speed sensor, which is used to collect meteorological data of the environment where the cableway is located in real time;

[0108] The inspection image display unit 102 is used to receive and display the inspection image sent back by the drone 201;

[0109] The route control unit 103 is used to receive and display the real-time position information of the drone 201 and its flight route, determine the position relationship between the current position of the drone 201 and the support, and perform different inspection methods according to the vulnerability level of the support;

[0110] The fault summary unit 104 is used to display the fault results of this inspection, including the fault image and fault location information, and send them to the data storage module 300;

[0111] The drone nest unit 105 is used to control the opening and closing of the nest door when the drone 201 departs and returns.

[0112] The data acquisition module 200 includes:

[0113] The drone 201 is used to fly along a designated route and send the collected image data, infrared image data, and location information to the inspection management system 100; the designated route is divided into an overall inspection route and a local inspection route; the overall inspection route is suitable for flying along the cableway cable, based on the existing cableway line and flying at a preset distance above it; the local inspection route is suitable for detecting the cableway bracket, and the specific route is as follows Figure 2 As shown, L001 is the inspection route for low-fragility brackets, which flies to the rear, left, right and front of the bracket; L002 is the inspection route for high-fragility brackets, which takes the bracket column as the center, flies half a circle around the bracket, and then passes through the middle of the route to complete the other half of the flight.

[0114] A high-resolution camera 202 is fixed in front of the UAV frame and is used to capture high-definition image data of the cableway;

[0115] An infrared camera 203 is fixed below the UAV frame and is used to capture infrared image data of the cableway;

[0116] The infrared distance measuring sensor 204 is fixed below the infrared camera 203 and is used to measure the distance between the drone 201 and the cableway support, and to determine whether to execute the overall inspection route or the partial inspection route according to the distance; if the distance is less than two meters, the partial inspection route is executed; if the distance is equal to or greater than two meters, the overall inspection route is executed;

[0117] The data storage module 300 includes:

[0118] The historical data storage unit 301 is used to store the inspection image data of each month;

[0119] The fault data storage unit 302 is used to store the image data, fault type, fault location and fault time of all faults within one year;

[0120] The data retrieval unit 303 is used to retrieve the stored image data according to the fault type, fault location or fault time.

[0121] In the embodiment of the present invention, the fault analysis module 400 is used to pre-process the image data, delete redundant images, and improve the quality of the data body. The YOLOv8 algorithm is also used to detect and locate the bolts to determine their fault conditions, which greatly improves the efficiency and accuracy of cableway inspections.

[0122] By setting up the data storage module 300, it can help the operation and maintenance personnel to store inspection images and historical fault images, and provide a retrieval function, which can help to formulate more effective maintenance plans and strategies through historical data analysis.

[0123] Through the organic combination of the above modules, the present invention realizes all-round and autonomous inspection of the cableways in ski resorts, ensures the safe operation of the cableways, improves inspection efficiency and accuracy, reduces labor costs and risks, and promotes the development of the ice and snow industry towards intelligence.

[0124] like Figure 6 As shown, a method for autonomous inspection of a ski resort cableway based on a drone according to an embodiment of the present invention is applied to the above-mentioned autonomous inspection system for a ski resort cableway based on a drone, and includes the following steps:

[0125] Step 1: The weather monitoring unit 101 monitors the weather conditions of the current environment of the ski resort cableway and determines whether the current weather conditions are suitable for performing the inspection task; if suitable, the drone nest unit 105 opens the nest door and sends the inspection task; if not suitable, the inspection task is suspended and waits for suitable weather;

[0126] Step 2: After the drone 201 starts the inspection, it uses the high-resolution camera 202 and infrared camera 203 on board to collect cableway image data;

[0127] Step 3: The route control unit 103 receives the real-time position information of the drone 201 and determines whether the distance between the drone 201 and the cableway support is greater than two meters; if the distance is greater than or equal to two meters, the overall inspection route is executed; if the distance is less than two meters, the partial inspection route is executed;

[0128] When executing the local inspection route, it is determined whether the bracket is a vulnerable bracket; if the bracket is a low-vulnerability bracket, the UAV 201 executes the low-vulnerability route L001; if the bracket is a high-vulnerability bracket, the UAV 201 executes the high-vulnerability route L002;

[0129] Step 4: Determine the image data location information according to the cableway bracket number in the route;

[0130] Step 5: After completing the inspection task, the drone 201 returns to the nest;

[0131] Step 6: The fault analysis module 400 pre-processes the collected image data to obtain a cableway image data set, and performs different fault detection methods for different parts;

[0132] For the bracket wheel set, wheel set wear fault detection 401 is used;

[0133] Bolt connection fault detection 402 is used for the connection condition of the bracket bolt group;

[0134] For ropeway cables, cable foreign body detection 403 is used;

[0135] Step 7: The fault analysis module 400 sends the detection result to the fault summary unit 104, which displays the fault of this inspection, summarizes the fault report, and sends it to the operation and maintenance personnel;

[0136] Step 8: Save the inspection image of this time to the historical data storage unit 301, and save the fault image to the fault data storage unit 302;

[0137] Step nine, the data storage module 300 uses the bracket fragility level determination formula to determine the fragility of the brackets inspected this time;

[0138] Step 10: The drone nest unit 105 performs charging and maintenance operations on the drone.

[0139] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0140] Embodiment 1. In this embodiment, the image data is high-definition image data obtained by photographing the cableway with a high-resolution camera 202 carried by a drone 201;

[0141] Through high-definition image data, the connection status of the bracket bolt group can be detected, such as Figure 3 As shown, the steps of bolt connection fault detection 402 are as follows:

[0142] S402-01. Train the YOLOv8 model using a dataset of labeled images containing bolt groups, nuts, and bolts.

[0143] S402-02, input the image data in the bracket subset into the trained YOLOv8 model to perform target detection, identify and locate the nuts and bolts in the image;

[0144] S402-03, marking the detected nuts and bolts, and outputting the detection results, including the number, location and confidence of the nuts and bolts;

[0145] S402-04. Determine whether there is a missing nut or a missing bolt based on the detection result of the target detection algorithm;

[0146] S402-05, comparing the number of nuts and bolts detected with a predetermined standard number, and if the number of bolts or nuts in the detection result is less than the predetermined number, determining that the bolt group has a connection fault;

[0147] S402-06, verifying the detected positions of nuts and bolts. If the positions of nuts and bolts in the detection results do not match the predetermined positions, it can also be determined as a connection failure;

[0148] S402-07. For bolt groups with missing nuts or bolts, record their specific locations and fault types, and send fault alarms to operation and maintenance personnel so that they can be repaired or replaced in a timely manner;

[0149] S402-08, storing the detection results and image data in the data storage module 300, including the location of the faulty bolt group, the fault type and the detection time, for subsequent analysis and maintenance.

[0150] Embodiment 2: Embodiment 2 of the present invention is different from Embodiment 1 in that the source of the image data and the fault detection object are different;

[0151] In this embodiment, the image data is infrared image data obtained by photographing the cableway support through the infrared camera 203 carried by the drone 201;

[0152] Through infrared image data, wheel wear faults can be detected, such as Figure 4 As shown, the steps of the wheel set wear fault detection 401 are as follows:

[0153] S401-01, converting the pixel value of the wheel set infrared image into the actual temperature value by using a preset temperature conversion formula to obtain the actual temperature of the wheel set;

[0154] S401-02, calculating the average temperature value of all pixels in the wheel group area to obtain the average temperature of all wheel groups;

[0155] S401-03, presetting a temperature threshold according to the temperature range of normal operation of the wheel set, for judging the wear condition of the wheel set;

[0156] S401-04, calculate the standard deviation of all wheel group temperatures to determine the temperature distribution;

[0157] S401-05. Compare the temperature of each wheel group by combining the average wheel group temperature, standard deviation and set temperature threshold:

[0158] S401-06, if the temperature of a wheel set exceeds the set temperature threshold, or is higher than the average temperature of the wheel set plus one standard deviation, it is determined that the wheel set has abnormal wear;

[0159] S401-07: If the temperature of a wheel set shows abnormal wear for five or more days during a week of routine inspection, the operation and maintenance personnel must be notified to repair or replace the wheel set;

[0160] S401-08, storing the detection results and image data in the data storage module 300, including the location of wheel wear and the detection time, for subsequent analysis and maintenance.

[0161] In this embodiment, the temperature distribution of the wheelset can be accurately obtained through the infrared image data of the wheelset. Through the preset temperature conversion formula and temperature threshold, abnormal temperature can be quickly identified, thereby discovering potential wear problems, so as to perform preventive maintenance, reduce downtime, and improve the safety and reliability of the entire cableway system.

[0162] Embodiment 3. Embodiment 3 of the present invention is different from Embodiment 1 in that the fault detection object is different;

[0163] In this embodiment, the object of fault detection is the cableway rope, and the purpose of detection is to determine whether there is a foreign object on the rope, such as Figure 5 As shown, the steps of cable foreign body detection 403 are as follows:

[0164] S403-01, using the EDlines algorithm to perform edge detection on the image of the cable subset, extracting edge features of the cables, and obtaining a first feature map;

[0165] S403-02, using the LSD algorithm to perform line segment detection on the first feature map, extracting line segment features of the cable, and obtaining a second feature map;

[0166] S403-03, fitting the line segment detection result of the second feature graph to construct a cable model;

[0167] S403-04, based on the cable model, performing a difference calculation between the currently acquired cable image and the cable model to obtain a difference image, performing threshold segmentation on the difference image, and extracting an area with a large difference as a foreground area;

[0168] S403-05, extracting the color, shape, texture and other features of the foreground area and constructing a feature vector;

[0169] S403-06, using a support vector machine to classify the extracted features and determine whether the foreground area is a foreign object;

[0170] S403-07, if the classification result is a foreign object, determining that the foreground area is a foreign object on the cable;

[0171] S403-08, record the specific location and detection time of the foreign object, and send a fault alarm to the operation and maintenance personnel so that the foreign object can be removed in time;

[0172] S403-09, storing the detection results and image data in the data storage module 300, including the location, type and detection time of the cable foreign body, for subsequent analysis and maintenance.

[0173] In this embodiment, the image data is processed by machine vision technology, a cable model is constructed, and a support vector machine is used to determine whether there is a foreign object, so as to quickly and accurately detect foreign objects on the cable and ensure the safe operation of the cableway. The data storage and alarm system is used to promptly notify the operation and maintenance personnel to handle the problem, thereby improving the inspection efficiency and fault response speed.

[0174] The above are only preferred embodiments of the present invention. It should be pointed out that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention. These should also be regarded as the protection scope of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A method for autonomous inspection of a ski resort cableway based on a drone, characterized in that: The following steps are involved: Step 1: The meteorological monitoring unit monitors the meteorological conditions of the current environment of the ski resort cableway and determines whether the current meteorological conditions are suitable for performing the inspection task; If it is suitable to perform the inspection task, the drone nest unit opens the nest door and sends the inspection task; If it is not suitable to perform the inspection task, the inspection task will be suspended and wait for suitable weather; Step 2: After the drone starts the inspection, it uses the high-resolution camera and infrared camera on board to collect cableway image data, and uses the infrared ranging sensor on board to measure the distance between the drone and the cableway support; Step 3: The route control unit receives the real-time location information of the UAV and determines whether the distance between the UAV and the cableway support is greater than two meters; if the distance is greater than or equal to two meters, the overall inspection route is executed; if the distance is less than two meters, the local inspection route is executed; When executing the local inspection route, determine whether the cableway bracket is a vulnerable bracket; if the cableway bracket is a low-vulnerability bracket, the drone executes the low-vulnerability route; if the cableway bracket is a high-vulnerability bracket, the drone executes the high-vulnerability route; Step 4: Determine the image data location information according to the cableway bracket number in the route; Step 5: After completing the inspection task, the drone returns to the nest; Step 6: The fault analysis module pre-processes the collected image data to obtain a cableway image data set, and performs different fault detection methods for different parts; For the bracket wheel set, wheel set wear fault detection is adopted; Bolt connection fault detection is used for the connection of the bracket bolt group; For ropeway cables, cable foreign body detection is used; Step 7: The fault analysis module sends the detection results to the fault summary unit, which displays the faults of this inspection, summarizes the fault report, and sends it to the operation and maintenance personnel; Step 8: Save the inspection image of this time into the historical data storage unit, and save the fault image into the fault data storage unit; Step 9: The data storage module uses the bracket fragility level determination formula to determine the fragility of the brackets inspected this time; Step 10: The drone nest unit charges and performs maintenance operations on the drone; The determination formula of the support fragility level is: S = 0.5 × P + 0.3 × M + 0.2 × H; Among them, S represents the vulnerability level score, with a full score of 10 points; 0-6 points represent low vulnerability levels, and 7-10 points represent high vulnerability levels; P represents the number of failures of the bracket within a year, with a full score of 10 points. The more failures there are, the higher the score; M represents the usage time of the bracket when the failure occurs, with a full score of 10 points. The longer the usage time, the higher the score; H represents the environmental conditions of the bracket, including temperature and humidity, with a full score of 10 points. The worse the environment, the higher the score.

2. The autonomous inspection method for ski resort cableways based on drones according to claim 1 is characterized in that: The specific process of the step of determining the image data location information according to the cableway bracket number in the route is as follows: The entire route is divided into multiple sections using the cableway bracket as a node; The cables between two adjacent brackets are numbered in sections according to the existing numbers of the two adjacent brackets; The high-resolution camera on the drone is used to obtain the bracket number, and the bracket position is matched to the number; After the image data of the current bracket is collected, the drone flies along the cable to the next bracket and obtains its number, and determines the number of the cable section according to the numbers of the two adjacent brackets; and the image data taken by the drone are digitally marked in sequence; The position mark is combined with the serial number to form the position information corresponding to each image data.

3. The autonomous inspection method for ski resort cableways based on drones according to claim 1 is characterized in that: The process of preprocessing the collected image data by the fault analysis module is as follows: Use image hashing to remove similar images; Gaussian filter is used to denoise the image data; Histogram equalization is used to enhance the image data to improve the contrast and brightness of the image; Correct image distortion caused by the UAV’s flight posture; The preprocessed image data is constructed into a cableway image dataset; The threshold segmentation technique is used on the cableway image dataset to separate the bracket from the background and calculate the ratio of the number of bracket pixels to the total number of image pixels; if the ratio is greater than or equal to 70%, the image is classified as the bracket subset; if the ratio is less than 70%, the image is classified as the cable subset; The bracket sub-assembly is used to detect wheel wear failure and bolt connection failure, and the cable sub-assembly is used to detect cable foreign body failure.

4. The autonomous inspection method for ski resort cableways based on drones according to claim 1 is characterized in that: The steps of wheel set wear fault detection are as follows: The pixel value of the wheel set infrared image is converted into an actual temperature value by a preset temperature conversion formula to obtain the actual temperature of the wheel set; Calculate the average temperature value of all pixels in the wheel group area to obtain the average temperature of all wheels; According to the temperature range of normal operation of the wheel set, a temperature threshold is pre-set to judge the wear condition of the wheel set; Calculate the standard deviation of all wheel group temperatures to determine the temperature distribution; Combining the average wheel temperature, standard deviation, and set temperature threshold, the temperature of each wheel is compared: If the temperature of a wheel set exceeds the set temperature threshold, or is higher than the average temperature of the wheel set plus one standard deviation, it is determined that the wheel set has abnormal wear; If the temperature of a wheel set shows abnormal wear for five or more days during a week of routine inspection, the wheel set should be repaired or replaced; The detection results and image data are stored in the data storage module, including the location of wheel wear and the detection time.

5. The autonomous inspection method for ski resort cableways based on drones according to claim 1 is characterized in that: The steps of bolt connection fault detection are as follows: The YOLOv8 model is trained using a dataset of labeled images of bolt groups, nuts, and bolts. The image data in the bracket subset is input into the trained YOLOv8 model to perform object detection, identify and locate the nuts and bolts in the image; Mark the detected nuts and bolts and output the test results, including the number, location and confidence of nuts and bolts; According to the detection results of the target detection algorithm, determine whether there is a missing nut or bolt; Comparing the number of nuts and bolts detected with a predetermined standard number, if the number of bolts or nuts in the detection result is less than the predetermined number, it is determined that the bolt group has a connection fault; Verify the detected nut and bolt positions. If the nut and bolt positions in the detection result do not match the predetermined positions, it is determined to be a connection failure. For bolt groups with missing nuts or bolts, the specific location and fault type are recorded and a fault alarm is issued; The detection results and image data are stored in the data storage module, including the location of the faulty bolt group, the fault type and the detection time.

6. The autonomous inspection method for ski resort cableways based on drones according to claim 1 is characterized in that: The steps of cable foreign body detection are as follows: Use the EDlines algorithm to perform edge detection on the images of the cable subset, extract the edge features of the cables, and obtain the first feature map; Use the LSD algorithm to perform line segment detection on the first feature map, extract the line segment features of the cable, and obtain the second feature map; Fitting the line segment detection result of the second feature map to construct a cable model; Based on the cable model, the currently acquired cable image is subjected to difference calculation with the cable model to obtain a difference image, and the difference image is subjected to threshold segmentation to extract the area with the largest difference as the foreground area; Extract the color, shape and texture features of the foreground area and construct a feature vector; Support vector machine is used to classify the extracted features and determine whether the foreground area is a foreign object; If the classification result is a foreign object, the foreground area is determined to be a foreign object on the cable; Record the specific location and detection time of foreign matter, and issue a fault alarm; The detection results and image data are stored in the data storage module, including the location, type and detection time of the foreign matter in the cable.

7. An autonomous inspection system for a ski resort cableway based on a drone, used to implement the autonomous inspection method for a ski resort cableway based on a drone as described in any one of claims 1 to 6, characterized in that: include: Inspection management system to monitor the ski resort environment and weather conditions; Receive and display the inspection images sent back by the drone, receive and display the real-time location information of the drone and its flight trajectory; determine the relationship between the current position of the drone and the position of the bracket, and perform different inspection methods according to the vulnerability level of the bracket; display the inspection results, including fault images and fault location information; control the opening and closing of the machine nest door when the drone departs and returns; The data acquisition module is used to control the UAV to obtain the relevant image data of each cableway component during the inspection process and the location information when obtaining it according to the specified route; wherein the location information of the image data is determined according to the existing number on each cableway bracket in the route; A data storage module is used to store all inspection image data within a month and all fault image data within a year; wherein the fault image data is used to determine the fragility level of the bracket; The fault analysis module is used to pre-process the image data, generate a cableway image data set, and divide the cableway image data set into a bracket subset and a cable subset according to the proportion of the cableway components in the image screen; wherein the bracket subset is used to detect wheel wear faults and bolt connection faults, and the cable subset is used to detect cable foreign body faults; the fault analysis module performs wheel wear fault detection, bolt connection fault detection and cable foreign body detection.

8. The autonomous inspection system for ski resort cableways based on drones according to claim 7 is characterized in that: The data acquisition module comprises: The drone is used to fly along the designated route and send the collected image data, infrared image data and location information to the inspection management system; the designated route is divided into an overall inspection route and a local inspection route; among them, the overall inspection route is suitable for flying along the cableway cable, and the local inspection route is suitable for inspecting the cableway bracket; A high-resolution camera, fixed to the front of the drone frame, is used to capture high-definition image data of the cableway; An infrared camera, fixed under the UAV frame, is used to capture infrared image data of the cableway; The infrared ranging sensor is fixed below the infrared camera and is used to measure the distance between the drone and the cableway support.

9. The autonomous inspection system for ski resort cableways based on drones according to claim 7 is characterized in that: The data storage module comprises: A historical data storage unit is used to store all inspection image data within one month; A fault data storage unit is used to store all fault image data, fault types, fault locations and fault times within one year; The data retrieval unit is used to retrieve the stored image data according to the fault type, fault location or fault time.

10. The autonomous inspection system for ski resort cableways based on drones according to claim 7, characterized in that: The inspection management system comprises: Meteorological monitoring unit, including temperature sensor, humidity sensor and TENG wind speed sensor, is used to collect meteorological data of the environment where the cableway is located in real time; Inspection image display unit, used to receive and display the inspection images sent back by the drone; The route control unit is used to receive and display the real-time location information of the UAV and its flight route, determine the relationship between the current position of the UAV and the position of the bracket, and perform different inspection methods according to the vulnerability level of the bracket; The fault summary unit is used to display the fault results of this inspection, including the fault image and fault location information, and send them to the data storage module; The drone nest unit is used to control the opening and closing of the nest door when the drone takes off and returns.

Images