Manhole cover disease detection method, device, system, electronic equipment and storage medium
The image of the manhole cover is obtained through the Internet of Vehicles platform for disease detection, which solves the problem that the inspection vehicle cannot detect the manhole cover being blocked and improves the inspection efficiency.
Patent Information
- Application Number
- CN202310487893.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-04-28
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2043-04-28
Smart Images

Figure CN116665164B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a method, device, electronic equipment and storage medium for detecting manhole cover defects. Background Art
[0002] With the implementation of artificial intelligence in the field of transportation, visual technology can be used to replace manual inspections of road defects. A road defect inspection vehicle is a vehicle equipped with an image sensor and an image processor. It uses image sensors to collect images or point cloud data of the road for image recognition and processing, thereby identifying road defects on the road. Manhole cover defect detection is an important item among road defects. During the inspection, the inspection vehicle will detect defects such as missing manhole covers, deep sinking or protruding manhole covers. Generally speaking, due to the tight traffic resources and people's driving needs, inspection vehicles usually do not prohibit other vehicles on the road when they are on the road. Instead, they consider going on the road for inspection when there are fewer vehicles on the road. Even if the inspection is carried out when there are fewer vehicles on the road, it is impossible to avoid the obstruction of the manhole cover by vehicles. The main reason is that the moving vehicles and the parked vehicles will obstruct the manhole cover on the road, making it impossible for the inspection vehicle to obtain the image of the manhole cover at the corresponding position for defect identification. It is obviously not in line with the actual traffic and cost to ask the inspection vehicle to stop and wait or re-inspect. Therefore, the existing inspection vehicle cannot detect defects on the obstructed manhole cover. Summary of the invention
[0003] The embodiment of the present invention provides a method for detecting defects in a manhole cover, aiming to solve the problem that existing inspection vehicles are unable to detect defects in obstructed manhole covers. When the inspection vehicle finds that a predetermined manhole cover position is obstructed during inspection, the vehicle networking platform is used to obtain an image of the manhole cover position as an image to be identified, and the defect detection result of the manhole cover position can be obtained by performing defect detection on the image to be identified. In the case where the manhole cover is obstructed, the inspection vehicle does not need to stop and wait or re-inspect, thereby solving the problem that existing inspection vehicles are unable to detect defects in obstructed manhole covers and improving the inspection efficiency of the inspection vehicle.
[0004] In a first aspect, an embodiment of the present invention provides a method for detecting manhole cover defects, the method comprising:
[0005] When the inspection vehicle inspects a predetermined manhole cover position, determining whether the manhole cover position is blocked by the vehicle;
[0006] If it is determined that the manhole cover position is blocked by a vehicle, an image to be identified of the manhole cover position is obtained based on the Internet of Vehicles platform;
[0007] Based on the image to be identified, the manhole cover position is subjected to disease detection to obtain a disease detection result of the manhole cover position.
[0008] Optionally, if it is determined that the manhole cover position is blocked by a vehicle, obtaining an image to be identified at the manhole cover position based on a vehicle networking platform includes:
[0009] If it is determined that the manhole cover is blocked by a vehicle, image information of the blocked vehicle is obtained;
[0010] Acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle;
[0011] Requesting a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information;
[0012] Based on the requested vehicle-mounted video, an image to be identified at the manhole cover location is determined.
[0013] Optionally, acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle includes:
[0014] Determining a travel speed of the obstructing vehicle according to the image information of the obstructing vehicle;
[0015] If the travel speed of the obstructing vehicle is less than or equal to a preset first speed threshold, determining the license plate information of the obstructing vehicle according to the image information of the obstructing vehicle;
[0016] The target vehicle information of the obstructing vehicle is obtained in the Internet of Vehicles platform according to the license plate information of the obstructing vehicle.
[0017] Optionally, after determining the travel speed of the obstructing vehicle according to the image information of the obstructing vehicle, the method includes:
[0018] If the speed of the blocking vehicle is greater than a preset first speed threshold, determining a candidate vehicle passing the manhole cover position based on the Internet of Vehicles platform;
[0019] Determine the travel speed of the candidate vehicle passing through the area where the manhole cover is located;
[0020] A target vehicle is determined from candidate vehicles whose driving speed is less than or equal to a preset second speed threshold, and target vehicle information of the target vehicle is determined in the connected vehicle network.
[0021] Optionally, determining the target vehicle from candidate vehicles having a driving speed less than or equal to a preset second speed threshold comprises:
[0022] Screening out to-be-determined vehicles whose travel speed is less than or equal to a preset second speed threshold from the candidate vehicles, wherein the number of the to-be-determined vehicles is greater than or equal to 2;
[0023] Acquire vehicle parameters and shooting equipment parameters of the vehicle to be determined, and compare the vehicle parameters and shooting equipment parameters of the vehicle to be determined with the vehicle parameters and shooting equipment parameters of the inspection vehicle to obtain a comparison result between the vehicle to be determined and the inspection vehicle;
[0024] A target vehicle is determined from the vehicles to be determined based on the comparison result.
[0025] Optionally, requesting a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information includes:
[0026] Determine the video request length according to the manhole cover position, and determine the video request object according to the target vehicle information;
[0027] Generate video request information based on the video request length and the video request object;
[0028] Based on the video request information, a corresponding vehicle video is requested in the Internet of Vehicles platform.
[0029] In a second aspect, an embodiment of the present invention further provides a manhole cover defect detection device, the manhole cover defect detection device comprising:
[0030] A first determination module is used to determine whether the manhole cover position is blocked by the vehicle when the inspection vehicle inspects the predetermined manhole cover position;
[0031] A first processing module is used to obtain an image to be identified at the manhole cover position based on the Internet of Vehicles platform if it is determined that the manhole cover position is blocked by a vehicle;
[0032] The second processing module is used to perform disease detection on the manhole cover position based on the image to be identified to obtain a disease detection result of the manhole cover position.
[0033] In a third aspect, an embodiment of the present invention further provides a manhole cover defect detection system, the manhole cover defect detection system comprising: a patrol vehicle and a vehicle networking platform, the patrol vehicle communicating with the vehicle networking platform via a vehicle networking protocol;
[0034] The inspection vehicle is used to determine whether the manhole cover position is blocked by the vehicle when inspecting a predetermined manhole cover position;
[0035] If it is determined that the manhole cover position is blocked by a vehicle, an image to be identified of the manhole cover position is obtained based on the Internet of Vehicles platform;
[0036] Based on the image to be identified, the manhole cover position is subjected to disease detection to obtain a disease detection result of the manhole cover position.
[0037] In a fourth aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the manhole cover disease detection method provided in the embodiment of the present invention when executing the computer program.
[0038] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the manhole cover disease detection method provided in the embodiment of the invention are implemented.
[0039] In an embodiment of the present invention, when the inspection vehicle inspects a predetermined manhole cover position, it determines whether the manhole cover position is blocked by a vehicle; if it is determined that the manhole cover position is blocked by a vehicle, the image to be identified of the manhole cover position is obtained based on the vehicle networking platform; the manhole cover position is subjected to defect detection based on the image to be identified to obtain a defect detection result of the manhole cover position. When the inspection vehicle inspects that the predetermined manhole cover position is blocked, the vehicle networking platform is used to obtain an image of the manhole cover position as the image to be identified, and the defect detection result of the manhole cover position can be obtained by performing defect detection on the image to be identified. In the case where the manhole cover is blocked, the inspection vehicle does not need to stop and wait or re-inspect, which solves the problem that the existing inspection vehicle cannot perform defect detection on the blocked manhole cover, and improves the inspection efficiency of the inspection vehicle. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0041] Figure 1 is a flow chart of a method for detecting manhole cover defects provided by an embodiment of the present invention;
[0042] Figure 2 is a structural schematic diagram of a manhole cover disease detection device provided in an embodiment of the present invention;
[0043] Figure 3 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0045] The embodiment of the present invention provides a method for detecting defects in a manhole cover, which is mounted on a manhole cover defect detection system. The manhole cover defect detection system at least includes a patrol vehicle and a vehicle networking platform. The patrol vehicle and the vehicle networking platform communicate via a vehicle networking protocol. The vehicle networking platform stores vehicle information and vehicle-mounted images or videos uploaded by the corresponding vehicle. The vehicle information may include vehicle identification, vehicle driving trajectory, vehicle parameters, equipment parameters and other information. The vehicle-mounted images or videos include shooting time and shooting location. The patrol vehicle can request the corresponding vehicle-mounted images or videos on the vehicle networking platform based on the vehicle information. After receiving the request from the patrol vehicle, the vehicle networking platform can search for the corresponding vehicle-mounted images or videos and return them to the patrol vehicle.
[0046] The inspection vehicle is provided with an image capturing device, which can capture the inspection road through the image capturing device to obtain corresponding captured images. It should be noted that when the inspection vehicle is driving, the image capturing device will capture images at a certain frame rate to obtain continuous captured images.
[0047] Specifically, the above-mentioned manhole cover position is predetermined, and the manhole cover in the road can be measured by surveying and mapping technology to obtain the position of each manhole cover in the road and record it, so that the position of each manhole cover is known in advance. When the inspection vehicle is inspecting, when it inspects the predetermined manhole cover position, it can first detect whether there is a vehicle above the manhole cover position. If there is a vehicle, it means that the manhole cover is blocked. The blocked vehicle can be a moving vehicle or a parked vehicle. The moving vehicle refers to a moving vehicle, and the parked vehicle refers to a stopped vehicle. It can be understood that if the manhole cover position is in the lane where the inspection vehicle is located, the inspection vehicle can capture the manhole cover image. If the manhole cover position is in the lane where the inspection vehicle is not located, there may be a vehicle running in parallel with the inspection vehicle to block the manhole cover position, or if the manhole cover position is in the rightmost lane, and the rightmost lane is a lane where vehicles can be temporarily parked, there may be a parked vehicle blocking the manhole cover position, or if the manhole cover position is in the lane where the inspection vehicle is not located at the traffic intersection, there may be a parked vehicle blocking the manhole cover.
[0048] Furthermore, the inspection vehicle uses an image capture device to take pictures during the inspection process. When it inspects a predetermined manhole cover position, it captures the manhole cover position to obtain a captured image of the manhole cover position. The captured image of the manhole cover position is subjected to vehicle detection or manhole cover detection, and it is determined whether the manhole cover position is blocked by a vehicle based on the detection result.
[0049] Specifically, the vehicle detection model can be used to detect vehicles in the captured images of the manhole cover position. If a vehicle is detected in the captured images of the manhole cover position, it can be determined that the manhole cover position is blocked by the vehicle. If no vehicle is detected in the captured images of the manhole cover position, it can be determined that the manhole cover position is not blocked by a vehicle. The above vehicle detection model can be a detection model based on machine learning, such as a vehicle detection model based on Haar feature + Adaboost algorithm, HOG feature + SVM algorithm, FasterR-CNN network, YOLO network, SSD network, etc.
[0050] Alternatively, the manhole cover detection model can be used to perform vehicle detection on the captured image of the manhole cover position. If a manhole cover is detected in the captured image of the manhole cover position, it can be determined that there is no vehicle blocking the manhole cover position. If no manhole cover is detected in the captured image of the manhole cover position, it can be determined that the manhole cover position is blocked by a vehicle. The above-mentioned manhole cover detection model can be a detection model based on machine learning, such as a manhole cover detection model based on Haar feature + Adaboost algorithm, HOG feature + SVM algorithm, FasterR-CNN network, YOLO network, SSD network, etc.
[0051] Furthermore, if it is determined that there is no vehicle blocking the manhole cover position, the captured image of the manhole cover position can be determined as the image to be identified, and the image to be identified can be subjected to defect detection to obtain the defect detection result of the manhole cover position. If it is determined that the manhole cover position is blocked by a vehicle, the vehicle-mounted image corresponding to the manhole cover position can be obtained based on the Internet of Vehicles platform, and the vehicle-mounted image can be pre-processed so that the vehicle-mounted image and the image captured by the inspection vehicle have the same resolution size to obtain the image to be identified, and the image to be identified can be subjected to defect detection to obtain the defect detection result of the manhole cover position. Specifically, if it is determined that the manhole cover position is blocked by a vehicle, a search can be performed on the Internet of Vehicles platform based on the manhole cover position to find a vehicle-mounted image whose shooting position is close to the manhole cover position; if there are multiple vehicle-mounted images whose shooting position is close to the manhole cover position, vehicle-mounted detection or manhole cover detection is performed on the vehicle-mounted image to determine a vehicle-mounted image with no obstruction to the manhole cover, and the vehicle-mounted image with the highest image quality is selected from the vehicle-mounted images blocked by the manhole cover as the vehicle-mounted image returned to the inspection vehicle. After obtaining the vehicle-mounted image corresponding to the manhole cover position, the inspection vehicle pre-processes the vehicle-mounted image so that the vehicle-mounted image and the image shot by the inspection vehicle have the same resolution size, thereby obtaining an image to be identified, and performing disease detection on the image to be identified to obtain a disease detection result at the manhole cover position.
[0052] In another possible embodiment, the manhole cover disease detection system includes a patrol vehicle, a vehicle networking platform, and a networked vehicle. The patrol vehicle and the vehicle networking platform communicate via a vehicle networking protocol. The networked vehicle and the vehicle networking platform communicate via a vehicle networking protocol. The vehicle information is stored on the vehicle networking platform. The vehicle information may include vehicle identification, vehicle driving trajectory, vehicle parameters, equipment parameters, and other information. The networked vehicle may store vehicle images or videos locally. Specifically, the patrol vehicle may request corresponding vehicle images or videos on the vehicle networking platform based on the vehicle information. After receiving the request from the patrol vehicle, the vehicle networking platform may request vehicle images or videos from the corresponding networked vehicle based on the vehicle information, and return the requested vehicle images or videos to the patrol vehicle.
[0053] Furthermore, if it is determined that the manhole cover position is blocked by a vehicle, the networked vehicle passing the manhole cover position can be obtained based on the Internet of Vehicles platform, and the networked vehicle can be requested for a vehicle-mounted image close to the manhole cover position. The networked vehicle returns the vehicle-mounted image close to the manhole cover position to the Internet of Vehicles platform, which then returns the vehicle-mounted image close to the manhole cover position to the inspection vehicle. The inspection vehicle pre-processes the vehicle-mounted image close to the manhole cover position so that the vehicle-mounted image and the image taken by the inspection vehicle have the same resolution size, and an image to be identified is obtained. The image to be identified is subjected to defect detection to obtain a defect detection result at the manhole cover position. Specifically, if it is determined that the manhole cover position is blocked by a vehicle, the vehicle information whose vehicle trajectory coincides with the manhole cover position can be determined in the Internet of Vehicles platform, and the corresponding networked vehicle can be requested for a vehicle-mounted image close to the manhole cover position based on the vehicle information. The networked vehicle returns the vehicle-mounted image close to the manhole cover position to the Internet of Vehicles platform. If the Internet of Vehicles platform receives vehicle-mounted images returned by multiple networked vehicles, the Internet of Vehicles platform can perform on-board detection or manhole cover detection on the vehicle-mounted image to determine the vehicle-mounted image without obstruction of the manhole cover, and select the vehicle-mounted image with the highest image quality from the vehicle-mounted images blocked by the manhole cover as the vehicle-mounted image returned to the inspection vehicle. After obtaining the vehicle-mounted image corresponding to the manhole cover position, the inspection vehicle pre-processes the vehicle-mounted image so that the vehicle-mounted image and the image taken by the inspection vehicle have the same resolution size, thereby obtaining the image to be identified, and performing defect detection on the image to be identified to obtain the defect detection result of the manhole cover position.
[0054] In an embodiment of the present invention, when the inspection vehicle finds that a predetermined manhole cover position is blocked during inspection, the Internet of Vehicles platform is used to obtain an image of the manhole cover position as an image to be identified, and the defect detection result of the manhole cover position can be obtained by performing defect detection on the image to be identified. In the case where the manhole cover is blocked, the inspection vehicle does not need to stop and wait or re-inspect, which solves the problem that the existing inspection vehicle cannot perform defect detection on blocked manhole covers, and improves the inspection efficiency of the inspection vehicle.
[0055] like Figure 1 As shown, Figure 1 1 is a flow chart of a method for detecting manhole cover defects provided by an embodiment of the present invention. The method for detecting manhole cover defects comprises the following steps:
[0056] 101. When the inspection vehicle inspects a predetermined manhole cover position, it determines whether the manhole cover position is blocked by the vehicle.
[0057] In an embodiment of the present invention, the above-mentioned manhole cover disease detection method can be mounted on a server or an edge end. The above-mentioned edge end can be an electronic device for image processing and analysis on a patrol vehicle. The above-mentioned patrol vehicle is provided with an image shooting device, and the patrol road can be photographed by the image shooting device to obtain a corresponding photographed image. It should be noted that when the patrol vehicle performs the patrol task on the patrol road, the image shooting device will shoot at a certain frame rate to obtain continuous photographed images.
[0058] The above manhole cover positions are predetermined, and the manhole covers on the road can be measured by surveying and mapping technology to obtain the positions of each manhole cover on the road and record them. In this way, the position of each manhole cover is known in advance. When the inspection vehicle performs inspection tasks on the inspection road, the position of each manhole cover on the inspection road can be obtained in advance. When the inspection vehicle inspects the predetermined manhole cover position, it can first detect whether there is a vehicle above the manhole cover position. If there is a vehicle, it means that the manhole cover is blocked.
[0059] In one embodiment, the position of the manhole cover can be detected by an ultrasonic radar to determine whether there is a vehicle at the manhole cover position. Specifically, an ultrasonic radar can be set on the inspection vehicle, and when the inspection vehicle inspects a predetermined manhole cover position, an ultrasonic signal is emitted to the manhole cover position, and whether there is a vehicle at the manhole cover position is determined based on the received echo signal.
[0060] In another embodiment, the manhole cover position may also be visually inspected by visual inspection technology to determine whether there is a vehicle at the entry position. Specifically, when the inspection vehicle inspects a predetermined manhole cover position, the manhole cover position is photographed by an image capturing device to obtain a captured image of the manhole cover position. Vehicle detection or manhole cover detection is performed on the captured image of the manhole cover position, and whether the manhole cover position is blocked by a vehicle is determined based on the detection result.
[0061] Specifically, the vehicle detection model can be used to detect vehicles in the captured images of the manhole cover position. If a vehicle is detected in the captured images of the manhole cover position, it can be determined that the manhole cover position is blocked by the vehicle. If no vehicle is detected in the captured images of the manhole cover position, it can be determined that the manhole cover position is not blocked by a vehicle. The above vehicle detection model can be a detection model based on machine learning, such as a vehicle detection model based on Haar feature + Adaboost algorithm, HOG feature + SVM algorithm, FasterR-CNN network, YOLO network, SSD network, etc.
[0062] Alternatively, the manhole cover detection model can be used to perform vehicle detection on the captured image of the manhole cover position. If a manhole cover is detected in the captured image of the manhole cover position, it can be determined that there is no vehicle blocking the manhole cover position. If no manhole cover is detected in the captured image of the manhole cover position, it can be determined that the manhole cover position is blocked by a vehicle. The above-mentioned manhole cover detection model can be a detection model based on machine learning, such as a manhole cover detection model based on Haar feature + Adaboost algorithm, HOG feature + SVM algorithm, FasterR-CNN network, YOLO network, SSD network, etc.
[0063] If it is determined that there is no vehicle blocking the manhole cover position, the captured image of the manhole cover position can be determined as the image to be identified, and the image to be identified is subjected to defect detection to obtain the defect detection result of the manhole cover position.
[0064] 102. If it is determined that the manhole cover position is blocked by a vehicle, an image to be identified at the manhole cover position is obtained based on the Internet of Vehicles platform.
[0065] In an embodiment of the present invention, if it is determined that the manhole cover position is blocked by a vehicle, it can be determined that the inspection vehicle cannot capture an image of the manhole cover position during this inspection mission, that is, the inspection vehicle cannot obtain the captured image of the manhole cover position, and thus cannot perform disease detection on the manhole cover position.
[0066] When the inspection vehicle cannot capture images of the manhole cover location, the corresponding image to be identified can be obtained through the onboard images of the manhole cover location taken by other vehicles. Using a third-party data source to provide the corresponding image to be identified is equivalent to asynchronously obtaining the image to be identified corresponding to the manhole cover location in the time dimension, without the need for the inspection vehicle to stop and wait or re-inspect.
[0067] In one embodiment, the vehicle information of each connected vehicle and the corresponding vehicle-mounted image or vehicle-mounted video are stored in the vehicle networking platform. The vehicle information may include vehicle identification, vehicle driving trajectory, vehicle parameters, equipment parameters and other information. The vehicle-mounted image or vehicle-mounted video includes shooting time and shooting location. The vehicle identification may be a license plate number or driver identity information. The vehicle driving trajectory may be a vehicle driving trajectory based on GPS or Beidou satellite navigation. The vehicle parameters may include vehicle size, vehicle speed, etc. The equipment parameters include the height, resolution, frame rate, etc. of the vehicle-mounted image device. The inspection vehicle may request the corresponding vehicle-mounted image or vehicle-mounted video on the vehicle networking platform according to the vehicle information. After receiving the request of the inspection vehicle, the vehicle networking platform may search for the corresponding vehicle-mounted image or vehicle-mounted video and return it to the inspection vehicle. After receiving the vehicle-mounted image or vehicle-mounted video returned by the vehicle networking platform, the server or edge end of the inspection vehicle may pre-process the vehicle-mounted image or vehicle-mounted video to obtain the image to be identified corresponding to the manhole cover position.
[0068] In another embodiment, the above-mentioned Internet of Vehicles platform stores vehicle information corresponding to each networked vehicle, and the above-mentioned vehicle information may include vehicle identification, vehicle driving trajectory, vehicle parameters, equipment parameters and other information. The above-mentioned networked vehicle may store vehicle-mounted images or vehicle-mounted videos locally, and the above-mentioned vehicle-mounted images or vehicle-mounted videos include shooting time and shooting location. The above-mentioned vehicle identification may be a license plate number or driver identity information, the above-mentioned vehicle driving trajectory may be a vehicle driving trajectory based on GPS or Beidou satellite navigation, the above-mentioned vehicle parameters may include vehicle size, vehicle speed, etc., and the above-mentioned equipment parameters include the height, resolution, frame rate, etc. of the vehicle-mounted image device. Specifically, the inspection vehicle may request the corresponding vehicle-mounted image or vehicle-mounted video on the Internet of Vehicles platform according to the vehicle information. After receiving the request of the inspection vehicle, the Internet of Vehicles platform may request the corresponding vehicle-mounted image or vehicle-mounted video from the networked vehicle according to the vehicle information, and return the requested vehicle-mounted image or vehicle-mounted video to the inspection vehicle. After receiving the vehicle-mounted image or vehicle-mounted video returned by the Internet of Vehicles platform, the server or edge end of the inspection vehicle may pre-process the vehicle-mounted image or vehicle-mounted video to obtain the image to be identified corresponding to the manhole cover position.
[0069] 103. Perform disease detection on the manhole cover position based on the image to be identified to obtain a disease detection result of the manhole cover position.
[0070] In an embodiment of the present invention, a trained disease detection model can be used to perform disease detection on the image to be identified, thereby obtaining a disease detection result of the manhole cover position. The above-mentioned disease detection model is used to detect the disease type, disease location and disease degree of the manhole cover in the image to be identified. The above-mentioned disease detection model can be a deep convolutional neural network. Specifically, the above-mentioned disease detection model can be a model constructed based on a deep convolutional neural network such as ResNet, YOLOV, Faster R-CNN, SSD, RetinaNet, etc.
[0071] The above-mentioned disease detection model can be trained with a pre-prepared data set, and a trained disease detection model is obtained after the training is completed. The above-mentioned trained disease detection model is deployed on the server or edge end of the inspection vehicle. After the image to be identified is obtained on the server or edge end, the image to be identified is input into the disease detection model for processing. The disease detection model outputs the disease information such as the disease type, disease location and disease degree existing in the image to be identified as the disease identification result.
[0072] The above-mentioned data set includes sample images and manhole cover disease labels corresponding to the sample images. The manhole cover disease labels corresponding to the sample images can be annotated by relevant professionals. One sample image can correspond to a group of manhole cover disease labels, and a group of manhole cover disease labels includes annotation information such as disease type, disease location, and disease degree.
[0073] During the training process, the sample image is input into the disease detection model to be trained for processing, and the disease detection result corresponding to the sample image is output. The error loss between the disease detection result corresponding to the sample image and the manhole cover disease label corresponding to the sample image is calculated. The optimization goal is to minimize the error loss. The parameters of the disease detection model to be trained are adjusted through the error back propagation algorithm. The above error calculation and parameter adjustment process is iterated until the disease detection model to be trained converges at the minimum error loss or the number of iterations reaches the preset number. The training is completed to obtain a trained disease recognition model.
[0074] The trained disease detection model outputs a disease detection frame (x, y, w, h, r, u, v), where (x, y) represents the center position of the disease detection frame, w represents the width of the disease detection frame, h represents the height of the disease detection frame, r represents the confidence of the disease detection frame, u represents the disease type in the disease detection frame, and v represents the degree of the disease in the disease detection frame.
[0075] The above-mentioned disease types can be determined by u of the disease detection frame, and the disease types may include manhole cover damage, manhole cover loss, manhole cover displacement, cracks around the manhole cover, manhole cover depression, manhole cover bulge and other diseases.
[0076] The above-mentioned defect position may be the position of the defect in the image to be identified, and the defect position may be represented by the center position of the defect detection frame. In a possible embodiment, the image to be identified has a corresponding shooting time and shooting position. After determining the shooting position of the image to be identified, the defect position in the image to be identified may be converted into the defect position in the real world through the conversion relationship between the camera coordinate system of the image shooting device and the real world coordinate system. In a possible embodiment, the position of the manhole cover in the image to be identified may be identified at the same time. Since the position of the manhole cover is known in the real world, the relative position of the defect position and the manhole cover position is calculated in the image to be identified, and the corresponding defect position may be mapped in the real world according to the relative position. Specifically, since the position of the manhole cover in the real world is known, when the position of the manhole cover in the image to be identified is detected, the mapping relationship between the manhole cover position in the real world and the manhole cover position in the image to be identified may be calculated, and the defect position in the image to be identified may be mapped to the real world through the mapping relationship to obtain the defect position in the real world.
[0077] The above-mentioned disease degree can be determined by v of the disease detection frame, and the disease degree can be expressed in the form of a score or a level. The higher the score of the disease degree, the more serious the disease, or the higher the disease level, the more serious the disease.
[0078] In an embodiment of the present invention, when the inspection vehicle inspects a predetermined manhole cover position, it determines whether the manhole cover position is blocked by a vehicle; if it is determined that the manhole cover position is blocked by a vehicle, an image to be identified of the manhole cover position is obtained based on the vehicle networking platform; based on the image to be identified, the manhole cover position is subjected to a defect detection to obtain a defect detection result of the manhole cover position. When the inspection vehicle inspects that a predetermined manhole cover position is blocked, the vehicle networking platform is used to obtain an image of the manhole cover position as an image to be identified, and a defect detection result of the manhole cover position can be obtained by performing a defect detection on the image to be identified. In the case where the manhole cover is blocked, the inspection vehicle does not need to stop and wait or re-inspect, which solves the problem that the existing inspection vehicle cannot perform defect detection on the blocked manhole cover, and improves the inspection efficiency of the inspection vehicle.
[0079] It is understandable that in the specific implementation of this application, the relevant data such as captured images, vehicle information, vehicle-mounted images or vehicle-mounted videos, when the embodiments in this application are applied to specific products or technologies, need to obtain the permission or consent of the relevant departments, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions.
[0080] Optionally, in the step of obtaining the image to be identified at the manhole cover position based on the Internet of Vehicles platform if it is determined that the manhole cover position is blocked by a vehicle, image information of the blocked vehicle can be obtained; target vehicle information of the blocked vehicle can be obtained in the Internet of Vehicles platform based on the image information of the blocked vehicle; corresponding in-vehicle video can be requested in the Internet of Vehicles platform based on the target vehicle information; and the image to be identified at the manhole cover position can be determined based on the requested in-vehicle video.
[0081] In an embodiment of the present invention, if it is determined that the manhole cover position is blocked by a vehicle, it can be determined that the inspection vehicle cannot capture an image of the manhole cover position during this inspection task, that is, the inspection vehicle cannot obtain the captured image of the manhole cover position, and thus cannot perform disease detection on the manhole cover position. The above-mentioned blocking vehicle refers to a vehicle at the manhole cover position, and the blocking vehicle can be a moving vehicle or a parked vehicle.
[0082] The above-mentioned Internet of Vehicles platform at least includes vehicle information of each vehicle connected to the network.
[0083] The image information of the vehicle may be a photographed image of the obstructing vehicle. Specifically, after determining that the manhole cover is obstructed by the vehicle, the obstructing vehicle may be photographed by an image capturing device on the inspection vehicle to obtain a vehicle image of the obstructing vehicle. The license plate information of the obstructing vehicle is obtained by performing license plate recognition on the vehicle image through the server or edge end of the inspection vehicle. The vehicle information corresponding to the obstructing vehicle is obtained in the Internet of Vehicles platform according to the vehicle information as the target vehicle information.
[0084] Furthermore, the license plate of the obstructed vehicle can be photographed by the image capturing device on the inspection vehicle to obtain the license plate image of the obstructed vehicle. The license plate image is recognized by the server or edge end of the inspection vehicle to obtain the license plate information of the obstructed vehicle. The vehicle information corresponding to the obstructed vehicle is obtained in the Internet of Vehicles platform according to the vehicle information and returned to the inspection vehicle. The inspection vehicle can use the returned vehicle information as the target vehicle information.
[0085] If the above-mentioned Internet of Vehicles platform also includes on-board videos uploaded by each connected vehicle, after the server or edge end of the inspection vehicle obtains the target vehicle information, it generates corresponding video request information based on the target vehicle information, and requests the corresponding on-board video from the Internet of Vehicles platform through the video request information. The Internet of Vehicles platform searches for the corresponding on-board video based on the video request information and returns it to the inspection vehicle. The server or edge end of the inspection vehicle filters out frame images that meet the image quality conditions from the received on-board video, and performs image preprocessing on the frame image to obtain the corresponding image to be identified.
[0086] If the above-mentioned Internet of Vehicles platform does not include the in-vehicle videos uploaded by each connected vehicle, after the server or edge end of the inspection vehicle obtains the target vehicle information, the corresponding video request information is generated according to the target vehicle information, and the video request information is sent to the Internet of Vehicles platform. The Internet of Vehicles platform sends the video request information to the corresponding connected vehicle. After the connected vehicle receives the video request, the owner can agree or refuse to return the corresponding in-vehicle video. If the owner of the connected vehicle agrees to return the corresponding in-vehicle video, the in-vehicle video will be returned to the Internet of Vehicles platform, and the in-vehicle video will be forwarded to the inspection vehicle through the Internet of Vehicles platform. The server or edge end of the inspection vehicle selects the frame image that meets the image quality conditions from the received in-vehicle video, and performs image preprocessing on the frame image to obtain the corresponding image to be identified. If the owner of the connected vehicle refuses to return the corresponding in-vehicle video, the rejection result will be returned to the Internet of Vehicles platform, and the Internet of Vehicles platform will return the rejection result to the inspection vehicle.
[0087] In a possible embodiment, if the owner of the networked vehicle refuses to return the corresponding in-vehicle video, the result of the refusal will be returned to the Internet of Vehicles platform. The Internet of Vehicles platform can find the networked vehicles that pass through the manhole cover position within the preset time based on the vehicle's driving trajectory, and send the vehicle information corresponding to the networked vehicles that pass through the manhole cover to the inspection vehicle. After receiving the new vehicle information, the inspection vehicle generates new video request information and continues to request the in-vehicle video until the corresponding in-vehicle video is requested. The above-mentioned preset time can be set manually. For example, the above-mentioned preset time can be set to within the day before the inspection by the inspection vehicle.
[0088] Optionally, in the step of obtaining target vehicle information of the obstructing vehicle in the Internet of Vehicles platform based on the image information of the obstructing vehicle, the driving speed of the obstructing vehicle can be determined based on the image information of the obstructing vehicle; if the driving speed of the obstructing vehicle is less than or equal to a preset first speed threshold, the license plate information of the obstructing vehicle is determined based on the image information of the obstructing vehicle; and the target vehicle information of the obstructing vehicle is obtained in the Internet of Vehicles platform based on the license plate information of the obstructing vehicle.
[0089] In an embodiment of the present invention, the image information of the obstructing vehicle may include two consecutive frames of captured images, and the speed of the obstructing vehicle may be determined by the two consecutive frames of captured images. Specifically, the speed of the obstructing vehicle may be determined based on the position of the obstructing vehicle in the two consecutive frames of captured images and the time difference between the two consecutive frames of captured images.
[0090] The image information of the obstructing vehicle may be a tracking video of the obstructing vehicle, and the speed of the obstructing vehicle may be determined based on the tracking video of the obstructing vehicle. Specifically, the speed of the obstructing vehicle may be determined based on the position of the obstructing vehicle in the first and last frames of the tracking video and the duration of the tracking video.
[0091] If the speed of the obstructing vehicle is less than or equal to the preset first speed threshold, it can be said that the quality of the vehicle video captured by the obstructing vehicle may be good, and the inspection vehicle can select the best quality image in the vehicle video of the obstructing vehicle as the image to be identified. If the speed of the obstructing vehicle is greater than the preset first speed threshold, it can be said that the quality of the vehicle video captured by the obstructing vehicle may be poor, which will affect the detection accuracy of the inspection vehicle.
[0092] Furthermore, when the driving speed of the obstructed vehicle is less than or equal to a preset first speed threshold, the corresponding vehicle information can be searched on the Internet of Vehicles platform based on the license plate information of the obstructed vehicle. If the vehicle information of the obstructed vehicle is found, it means that the obstructed vehicle is a networked vehicle. The vehicle information corresponding to the obstructed vehicle is returned to the inspection vehicle. The inspection vehicle generates a corresponding on-board video request based on the vehicle information of the obstructed vehicle and sends it to the Internet of Vehicles platform.
[0093] Optionally, after the step of determining the driving speed of the obstructing vehicle based on the image information of the obstructing vehicle, if the driving speed of the obstructing vehicle is greater than a preset first speed threshold, candidate vehicles passing through the manhole cover position can also be determined based on the Internet of Vehicles platform; the driving speed of candidate vehicles passing through the area where the manhole cover position is located is determined; the target vehicle is determined from the candidate vehicles whose driving speed is less than or equal to the preset second speed threshold, and the target vehicle information of the target vehicle is determined in the Internet of Vehicles.
[0094] In an embodiment of the present invention, if the driving speed of the obstructed vehicle is greater than a preset first speed threshold, it means that the on-board video of the obstructed vehicle may have a problem of poor image quality. At this time, the inspection vehicle can request the Internet of Vehicles platform for networked vehicles that pass the manhole cover position within a preset time as candidate vehicles. After receiving the request, the Internet of Vehicles platform can find out the networked vehicles that pass the manhole cover position within a preset time according to the vehicle driving trajectory, and calculate the driving speed of the candidate vehicle according to the vehicle driving trajectory of the candidate vehicle, and then select a slower vehicle from the candidate vehicles according to the driving speed of the candidate vehicle as the target vehicle, and determine the target vehicle information corresponding to the target vehicle, and return the target vehicle information to the inspection vehicle.
[0095] Specifically, after determining that the driving speed of the obstructing vehicle is greater than a preset first speed threshold, a corresponding search request information can be generated according to the manhole cover position and the preset time. After receiving the search request information, the Internet of Vehicles platform parses the manhole cover position in the search request information, and determines the networked vehicles with the manhole cover position in the vehicle driving trajectory within the preset time as candidate vehicles, and determines the corresponding driving speed of each candidate vehicle according to the vehicle driving trajectory of the candidate vehicle, and determines the candidate vehicles that are less than or equal to the preset second speed threshold as the vehicles to be determined, selects the vehicle with the smallest driving speed among the vehicles to be determined as the target vehicle, and extracts the target vehicle information corresponding to the target vehicle and returns it to the inspection vehicle.
[0096] It should be noted that the second speed threshold is smaller than the first speed threshold. This is because the second speed threshold needs to screen out the most suitable target vehicle from a large number of vehicles connected to the network. Therefore, it can be set smaller to narrow the screening range. The first speed threshold only needs to determine whether the obstructing vehicle meets the conditions. Therefore, it can be set larger to meet the image quality for disease detection.
[0097] In an embodiment of the present invention, when the speed of the blocking vehicle is too fast, the vehicle-mounted video corresponding to the vehicles passing the manhole cover position at other times can be requested on the Internet of Vehicles platform, so as to obtain the vehicle-mounted video with higher image quality, thereby improving the detection accuracy of manhole cover defects.
[0098] Optionally, in the step of determining the target vehicle from candidate vehicles whose driving speed is less than or equal to a preset second speed threshold, vehicles to be determined whose driving speed is less than or equal to the preset second speed threshold can be screened out from the candidate vehicles, and the number of vehicles to be determined is greater than or equal to 2; vehicle parameters and shooting equipment parameters of the vehicle to be determined are obtained, and the vehicle parameters and shooting equipment parameters of the vehicle to be determined are compared with the vehicle parameters and shooting equipment parameters of the inspection vehicle to obtain a comparison result between the vehicle to be determined and the inspection vehicle; and the target vehicle is determined from the vehicles to be determined based on the comparison result.
[0099] In the embodiment of the present invention, the above-mentioned vehicle parameters may include vehicle size, vehicle speed, etc., and the above-mentioned device parameters may include the height, resolution, frame rate, etc. of the vehicle-mounted image device.
[0100] If there are multiple vehicles to be determined whose driving speed is less than or equal to the preset second speed threshold, the vehicle to be determined that is closest to the inspection vehicle in vehicle parameters and shooting equipment parameters can be selected as the target vehicle. Considering that the defect detection model is trained and adjusted according to the specific situation of the inspection vehicle, the defect detection model has the most adaptable detection ability for the images captured by the inspection vehicle. Therefore, when the number of vehicles to be determined is greater than or equal to 2, the vehicle parameters and shooting equipment parameters of the vehicle to be determined can be obtained, and the vehicle parameters and shooting equipment parameters of the vehicle to be determined can be calculated to be similar to the vehicle parameters and shooting equipment parameters of the inspection vehicle to obtain the parameter similarity between the vehicle to be determined and the inspection vehicle, and the vehicle to be determined with the greatest parameter similarity to the inspection vehicle can be selected as the target vehicle. After the target vehicle is determined on the Internet of Vehicles platform, the target vehicle information corresponding to the target vehicle is returned to the inspection vehicle. After receiving the target vehicle information, the inspection vehicle generates corresponding video request information according to the target vehicle information and sends it to the Internet of Vehicles platform to request the vehicle-mounted video corresponding to the target vehicle information. The frame images in the vehicle-mounted video obtained in this way are closer to the images taken by the inspection vehicle, and are more suitable for the defect detection of the inspection vehicle, thereby improving the accuracy of manhole cover defects.
[0101] Optionally, in the step of requesting the corresponding in-vehicle video in the Internet of Vehicles platform according to the target vehicle information, the video request length can be determined according to the manhole cover position, and the video request object can be determined according to the target vehicle information; video request information is generated based on the video request length and the video request object; and the corresponding in-vehicle video is requested in the Internet of Vehicles platform based on the video request information.
[0102] In an embodiment of the present invention, the video request length may be the video length corresponding to N meters before and after the manhole cover position, or the video request length may be the video length corresponding to M seconds before and after the manhole cover position. The video request object may be the vehicle identification of the target vehicle.
[0103] The inspection vehicle sends the video request information to the Internet of Vehicles platform. After receiving the video request, the Internet of Vehicles platform parses the video request to obtain the video request object and the length of the video request. According to the video request object, the vehicle video uploaded by the video request object is searched. According to the video request length, the vehicle video corresponding to N meters before and after the manhole cover position or the vehicle video corresponding to M seconds before and after the manhole cover position is searched in the vehicle video uploaded by the video request object. After finding the vehicle video corresponding to the video request information, the Internet of Vehicles platform sends the vehicle video to the inspection vehicle. The server or edge end of the inspection vehicle selects the frame image that meets the image quality conditions from the received vehicle video, and performs image preprocessing on the frame image to obtain the corresponding image to be identified.
[0104] It should be noted that the manhole cover disease detection method provided in the embodiment of the present invention can be applied to photographing equipment, smart phones, computers, servers and other equipment that can perform manhole cover disease detection.
[0105] like Figure 2 As shown, an embodiment of the present invention provides a manhole cover defect detection device, the manhole cover defect detection device comprising:
[0106] The first determination module 201 is used to determine whether the manhole cover position is blocked by the vehicle when the inspection vehicle inspects the predetermined manhole cover position;
[0107] The first processing module 202 is used to obtain an image to be identified at the manhole cover position based on the Internet of Vehicles platform if it is determined that the manhole cover position is blocked by a vehicle;
[0108] The second processing module 203 is used to perform disease detection on the manhole cover position based on the image to be identified to obtain a disease detection result of the manhole cover position.
[0109] Optionally, the first processing module 202 includes:
[0110] A first processing submodule is used for obtaining image information of the blocking vehicle if it is determined that the manhole cover is blocked by the vehicle;
[0111] An acquisition submodule, used for acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle;
[0112] A second processing submodule is used to request a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information;
[0113] The first determination submodule is used to determine the image to be identified at the manhole cover position based on the requested vehicle-mounted video.
[0114] Optionally, the acquisition submodule includes:
[0115] A first determining unit, configured to determine a travel speed of the obstructing vehicle according to the image information of the obstructing vehicle;
[0116] a second determining unit, configured to determine the license plate information of the obstructing vehicle according to the image information of the obstructing vehicle if the traveling speed of the obstructing vehicle is less than or equal to a preset first speed threshold;
[0117] An acquisition unit is used to acquire the target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the license plate information of the obstructing vehicle.
[0118] Optionally, the device comprises:
[0119] A second determination module is used to determine a candidate vehicle passing the manhole cover position based on the Internet of Vehicles platform if the driving speed of the blocking vehicle is greater than a preset first speed threshold;
[0120] A third determination module is used to determine the travel speed of the candidate vehicle passing through the area where the manhole cover is located;
[0121] The third processing module is used to determine a target vehicle from candidate vehicles whose driving speed is less than or equal to a preset second speed threshold, and determine target vehicle information of the target vehicle in the Internet of Vehicles.
[0122] Optionally, the third processing module includes:
[0123] A second determination submodule is used to screen out to-be-determined vehicles whose travel speed is less than or equal to a preset second speed threshold from the candidate vehicles, and the number of the to-be-determined vehicles is greater than or equal to 2;
[0124] The third processing submodule is used to obtain vehicle parameters and shooting equipment parameters of the vehicle to be determined, and compare the vehicle parameters and shooting equipment parameters of the vehicle to be determined with the vehicle parameters and shooting equipment parameters of the inspection vehicle to obtain a comparison result between the vehicle to be determined and the inspection vehicle;
[0125] The third determination submodule is used to determine a target vehicle among the vehicles to be determined based on the comparison result.
[0126] Optionally, the second processing submodule includes:
[0127] A third determining unit, configured to determine a video request length according to the manhole cover position, and determine a video request object according to the target vehicle information;
[0128] A first processing unit, configured to generate video request information based on the video request length and the video request object;
[0129] The second processing unit is used to request the corresponding in-vehicle video in the Internet of Vehicles platform based on the video request information.
[0130] It should be noted that the manhole cover disease detection device provided in the embodiment of the present invention can be applied to photographing equipment, smart phones, computers, servers and other equipment that can perform manhole cover disease detection.
[0131] The manhole cover defect detection device provided in the embodiment of the present invention can implement each process implemented by the manhole cover defect detection method in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.
[0132] See also Figure 3 , Figure 3is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, it includes: a memory 302, a processor 301, and a computer program of a manhole cover disease detection method stored in the memory 302 and executable on the processor 301, wherein:
[0133] The processor 301 is used to call the computer program stored in the memory 302 and execute the following steps:
[0134] When the inspection vehicle inspects a predetermined manhole cover position, determining whether the manhole cover position is blocked by the vehicle;
[0135] If it is determined that the manhole cover position is blocked by a vehicle, an image to be identified of the manhole cover position is obtained based on the Internet of Vehicles platform;
[0136] Based on the image to be identified, the manhole cover position is subjected to disease detection to obtain a disease detection result of the manhole cover position.
[0137] Optionally, if it is determined that the manhole cover position is blocked by a vehicle, the processor 301 acquires an image to be identified at the manhole cover position based on the Internet of Vehicles platform, including:
[0138] If it is determined that the manhole cover is blocked by a vehicle, obtaining image information of the blocked vehicle;
[0139] Acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle;
[0140] Requesting a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information;
[0141] Based on the requested vehicle-mounted video, an image to be identified at the manhole cover location is determined.
[0142] Optionally, the processor 301 executes the step of acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle, including:
[0143] Determining a travel speed of the obstructing vehicle according to the image information of the obstructing vehicle;
[0144] If the travel speed of the obstructing vehicle is less than or equal to a preset first speed threshold, determining the license plate information of the obstructing vehicle according to the image information of the obstructing vehicle;
[0145] The target vehicle information of the obstructing vehicle is obtained in the Internet of Vehicles platform according to the license plate information of the obstructing vehicle.
[0146] Optionally, after determining the travel speed of the obstructing vehicle according to the image information of the obstructing vehicle, the method executed by the processor 301 includes:
[0147] If the speed of the blocking vehicle is greater than a preset first speed threshold, determining a candidate vehicle passing the manhole cover position based on the Internet of Vehicles platform;
[0148] Determine the travel speed of the candidate vehicle passing through the area where the manhole cover is located;
[0149] A target vehicle is determined from candidate vehicles whose driving speed is less than or equal to a preset second speed threshold, and target vehicle information of the target vehicle is determined in the connected vehicle network.
[0150] Optionally, the processor 301 determines the target vehicle from candidate vehicles whose driving speed is less than or equal to a preset second speed threshold, including:
[0151] Screening out to-be-determined vehicles whose travel speed is less than or equal to a preset second speed threshold from the candidate vehicles, wherein the number of the to-be-determined vehicles is greater than or equal to 2;
[0152] Acquire vehicle parameters and shooting equipment parameters of the vehicle to be determined, and compare the vehicle parameters and shooting equipment parameters of the vehicle to be determined with the vehicle parameters and shooting equipment parameters of the inspection vehicle to obtain a comparison result between the vehicle to be determined and the inspection vehicle;
[0153] A target vehicle is determined from the vehicles to be determined based on the comparison result.
[0154] Optionally, the requesting of a corresponding in-vehicle video in the Internet of Vehicles platform according to the target vehicle information executed by the processor 301 includes:
[0155] Determine the video request length according to the manhole cover position, and determine the video request object according to the target vehicle information;
[0156] Generate video request information based on the video request length and the video request object;
[0157] Based on the video request information, a corresponding vehicle video is requested in the Internet of Vehicles platform.
[0158] It should be noted that the electronic device provided in the embodiment of the present invention can be applied to smart phones, computers, servers and other devices that can perform the manhole cover disease detection method.
[0159] The electronic device provided in the embodiment of the present invention can implement each process implemented by the method for detecting manhole cover defects in the above method embodiment, and can achieve the same beneficial effects. To avoid repetition, it will not be described here.
[0160] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the manhole cover disease detection method or the application-end manhole cover disease detection method provided in the embodiment of the present invention are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
[0161] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.
[0162] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.
Claims
1. A method for detecting manhole cover defects, It is characterized in that The method comprises the following steps: When the inspection vehicle inspects a predetermined manhole cover position, determining whether the manhole cover position is blocked by the vehicle; If it is determined that the manhole cover is blocked by a vehicle, obtaining image information of the blocked vehicle; Acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle; Requesting a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information; Determine the image to be identified at the manhole cover location based on the requested vehicle-mounted video; Based on the image to be identified, the manhole cover position is subjected to disease detection to obtain a disease detection result of the manhole cover position.
2. The method for detecting manhole cover defects according to claim 1, It is characterized in that The step of acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle includes: Determining a travel speed of the obstructing vehicle according to the image information of the obstructing vehicle; If the travel speed of the obstructing vehicle is less than or equal to a preset first speed threshold, determining the license plate information of the obstructing vehicle according to the image information of the obstructing vehicle; The target vehicle information of the obstructing vehicle is obtained in the Internet of Vehicles platform according to the license plate information of the obstructing vehicle.
3. The method for detecting manhole cover defects as claimed in claim 2, It is characterized in that After determining the travel speed of the obstructing vehicle according to the image information of the obstructing vehicle, the method includes: If the speed of the blocking vehicle is greater than a preset first speed threshold, a candidate vehicle passing the manhole cover position is determined based on the Internet of Vehicles platform; Determining a travel speed of the candidate vehicle passing the manhole cover position; A target vehicle is determined from candidate vehicles whose driving speed is less than or equal to a preset second speed threshold, and target vehicle information of the target vehicle is determined in the connected vehicle network.
4. The method for detecting manhole cover defects as claimed in claim 3, It is characterized in that The step of determining a target vehicle from candidate vehicles whose travel speed is less than or equal to a preset second speed threshold comprises: Screening out to-be-determined vehicles whose travel speed is less than or equal to a preset second speed threshold from the candidate vehicles, wherein the number of the to-be-determined vehicles is greater than or equal to 2; Acquire vehicle parameters and shooting equipment parameters of the vehicle to be determined, and compare the vehicle parameters and shooting equipment parameters of the vehicle to be determined with the vehicle parameters and shooting equipment parameters of the inspection vehicle to obtain a comparison result between the vehicle to be determined and the inspection vehicle; A target vehicle is determined from the vehicles to be determined based on the comparison result.
5. The method for detecting manhole cover defects according to any one of claims 1 to 4, It is characterized in that The requesting a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information includes: Determine the video request length according to the manhole cover position, and determine the video request object according to the target vehicle information, wherein the video request length is the video length corresponding to N meters before and after the manhole cover position; Generate video request information based on the video request length and the video request object; Based on the video request information, a corresponding vehicle video is requested in the Internet of Vehicles platform.
6. A manhole cover disease detection device, It is characterized in that The manhole cover disease detection device comprises: A first determination module is used to determine whether the manhole cover position is blocked by the vehicle when the inspection vehicle inspects the predetermined manhole cover position; The first processing module is used to obtain image information of the obstructing vehicle if it is determined that the manhole cover position is blocked by the vehicle; obtain target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle; request a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information; and determine the image to be identified at the manhole cover position based on the requested vehicle video; The second processing module is used to perform disease detection on the manhole cover position based on the image to be identified to obtain a disease detection result of the manhole cover position.
7. A manhole cover disease detection system, It is characterized in that The manhole cover disease detection system includes: an inspection vehicle and a vehicle networking platform, wherein the inspection vehicle is connected to the vehicle networking platform through a vehicle networking protocol; The inspection vehicle is used to determine whether the manhole cover position is blocked by the vehicle when inspecting a predetermined manhole cover position; If it is determined that the manhole cover is blocked by a vehicle, obtaining image information of the blocked vehicle; Acquiring target vehicle information of the obstructing vehicle in the Internet of Vehicles platform according to the image information of the obstructing vehicle; Requesting a corresponding vehicle video in the Internet of Vehicles platform according to the target vehicle information; Determine the image to be identified at the manhole cover location based on the requested vehicle-mounted video; Based on the image to be identified, the manhole cover position is subjected to disease detection to obtain a disease detection result of the manhole cover position.
8. An electronic device, It is characterized in that include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the method for detecting manhole cover defects as described in any one of claims 1 to 5 when executing the computer program.
9. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the manhole cover defect detection method according to any one of claims 1 to 5 are implemented.
