CN116012764A - Road disease identification and positioning method and device, electronic equipment and storage medium - Google Patents
Road disease identification and positioning method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116012764A CN116012764A CN202310089453.5A CN202310089453A CN116012764A CN 116012764 A CN116012764 A CN 116012764A CN 202310089453 A CN202310089453 A CN 202310089453A CN 116012764 A CN116012764 A CN 116012764A
- Authority
- CN
- China
- Prior art keywords
- target
- road surface
- disease
- image
- inspection image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Image Analysis (AREA)
Abstract
The application discloses a road disease identification and positioning method, a device, electronic equipment and a storage medium. Firstly, acquiring a target inspection image, and preprocessing the target inspection image; dividing a road surface area and a non-road surface area from the preprocessed target inspection image through a semantic segmentation model, and adjusting the size of the divided road surface area to obtain a road surface image; the road surface image is processed through the target detection model, each disease target in the road surface is determined, the repeated targets in the front frame and the rear frame in the image can be removed according to the characteristics of each disease target, and finally the disease targets are marked according to the position data when each frame of inspection image is read and collected. The invention utilizes the existing low-cost and mature hardware environment to facilitate deployment; the repeated targets are removed through GNSS positioning information, so that a target tracking algorithm is greatly simplified, and the calculation efficiency is effectively improved.
Description
Technical Field
The invention relates to the field of intelligent traffic, in particular to a road disease identification and positioning method, a device, electronic equipment and a storage medium.
Background
Along with the high-speed construction of traffic roads, a large number of roads can experience the influence of overload, natural environment and the like, and diseases such as cracks, pits and the like can appear, so that the later-stage road maintenance is important for guaranteeing the safe running of vehicles and the service life of the roads, and a large amount of manpower and material resources are required to be input for road inspection.
At present, a manual inspection mode is mainly adopted, namely an inspection person drives a road administration maintenance inspection vehicle to conduct naked eye observation on the road surface of the administered road, and when the road surface is damaged, the inspection person needs to get off the vehicle and take a photo by using a handheld terminal to report. However, inspection personnel can easily feel tired and can easily leak after long-term observation on the vehicle; and traffic hidden danger exists in taking photos of getting off, and normal operation of traffic is affected. In addition, a small number of specialized inspection vehicles are put into the industry, and are equipped with a high-precision differential positioning module, a three-dimensional laser scanning radar and the like, an industrial computer and the like to collect and analyze road surface data, but the equipment has high cost (about tens to millions of special vehicles in the market), is mainly applied to road detection before major, medium and minor repair of roads, and is difficult to popularize in daily maintenance inspection.
Disclosure of Invention
Based on the above, the embodiment of the application provides a road disease identification and positioning method, a device, electronic equipment and a storage medium, which are convenient to deploy by utilizing the existing low-cost and mature hardware environment; the GNSS information is utilized to remove the repeated targets, so that a target tracking algorithm is greatly simplified, and the calculation efficiency is effectively improved.
In a first aspect, a method for identifying and locating a road disease is provided, the method comprising:
acquiring a target inspection image, and preprocessing the target inspection image;
dividing a road surface area and a non-road surface area from the preprocessed target inspection image through a semantic segmentation model, and adjusting the size of the divided road surface area to obtain a road surface image;
and processing the road surface image through a target detection model to determine each disease target in the road surface.
Optionally, the method further comprises:
obtaining a patrol video stream, wherein the patrol video stream comprises a plurality of patrol images;
determining each disease target in each inspection image, and determining the characteristics of each disease target; wherein the features include coordinates, size, direction of motion in the image;
and calculating the probability that the corresponding disease targets in the front frame and the rear frame belong to the same target according to the similarity, and eliminating the repeated targets.
Optionally, the method further comprises:
when the inspection video stream is acquired, reading and acquiring position data of each frame of inspection image; wherein the coordinate data at least comprises longitude and latitude, course and speed;
and marking each disease target in the current frame according to the position data read by each frame of inspection image.
Optionally, marking each disease target in the current frame according to the position data read by each frame of inspection image includes:
and (3) placing the real-time detected disease target information and the patrol image into a queue cache pool of the FIFO type, and asynchronously processing the disease target in the cache pool in another thread by a target tracking algorithm.
Optionally, when the inspection video stream is acquired, reading position data when each frame of inspection image is acquired, including:
calculating a vehicle running state using the position data:
C diff ＝C now -C pre
wherein C is now The current driving direction angle; c (C) pre The last read driving direction angle; c (C) diff For the current driving direction and the lastDifference in driving direction.
In a second aspect, there is provided a road disease recognition and location device, the device comprising:
the first acquisition module is used for acquiring a target inspection image and preprocessing the target inspection image;
the segmentation module is used for segmenting a road surface area and a non-road surface area from the preprocessed target inspection image through a semantic segmentation model, and performing size adjustment on the segmented road surface area to obtain a road surface image;
and the determining module is used for processing the road surface image through the target detection model and determining each disease target in the road surface.
Optionally, the apparatus further comprises:
the second acquisition module is used for acquiring a patrol video stream, wherein the patrol video stream comprises a plurality of patrol images;
the determining module is also used for determining each disease target in each inspection image and determining the characteristics of each disease target; wherein the features include coordinates, size, direction of motion in the image;
and calculating the probability that the corresponding disease targets in the front frame and the rear frame belong to the same target according to the similarity, and eliminating the repeated targets.
Optionally, the apparatus further comprises:
when the inspection video stream is acquired, reading and acquiring position data of each frame of inspection image; wherein the coordinate data at least comprises longitude and latitude, course and speed;
and marking each disease target in the current frame according to the position data read by each frame of inspection image.
In a third aspect, there is provided an electronic device comprising a memory storing a computer program and a processor implementing the road fault identification positioning method according to any one of the first aspects above when the processor executes the computer program.
In a fourth aspect, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the road disease identification positioning method of any one of the above first aspects.
In the technical scheme provided by the embodiment of the application, a target inspection image is firstly obtained, and the target inspection image is preprocessed; dividing a road surface area and a non-road surface area from the preprocessed target inspection image through a semantic segmentation model, and adjusting the size of the divided road surface area to obtain a road surface image; and processing the road surface image through a target detection model to determine each disease target in the road surface. And further, repeated targets in the front frame and the rear frame can be removed according to the determined characteristics of each disease target. It can be seen that the beneficial effects brought by the technical scheme provided by the embodiment of the application at least include:
(1) The whole set of disease identification and positioning method based on the camera and the positioning module is provided, so that the method is convenient to deploy into a portable inspection instrument;
(2) The existing low-cost and mature hardware environment is utilized to facilitate deployment; the GNSS information is utilized to remove the repeated targets, so that a target tracking algorithm is greatly simplified, and the calculation efficiency is effectively improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It will be apparent to those of ordinary skill in the art that the drawings in the following description are exemplary only and that other implementations can be obtained from the extensions of the drawings provided without inventive effort.
Fig. 1 is a step flowchart of a road disease identifying and positioning method provided in an embodiment of the present application;
FIG. 2 is a schematic flow chart of performing object detection according to an embodiment of the present disclosure;
FIG. 3 is a schematic diagram of an image segmentation and resizing process provided in an embodiment of the present application;
FIG. 4 is a schematic diagram of multi-target tracking recognition provided in an embodiment of the present application;
fig. 5 is a block diagram of a road disease identifying and positioning device according to an embodiment of the present application;
fig. 6 is a schematic diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
In the description of the present invention, unless otherwise indicated, "a plurality" means two or more. The terms first and second in the description and claims of the invention and in the above-mentioned figures are intended to distinguish between the objects referred to. For schemes with time sequence flows, such term expressions are not necessarily to be understood as describing a specific order or sequence, nor are such term expressions to distinguish between importance levels, positional relationships, etc. for schemes with device structures.
Furthermore, the terms "comprises," "comprising," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements that are expressly listed but may include other steps or elements not expressly listed but inherent to such process, method, article, or apparatus or steps or elements that may be added based on a further optimization of the inventive concept.
In recent years, with the popularization of rear-loading vehicle-mounted equipment such as a driver monitoring system and a vehicle recorder, the equipment which is easy to install and deploy and has low cost can be modified into a portable vehicle-mounted road inspection instrument. Such devices are typically only equipped with a GNSS positioning/beidou positioning module, a camera module and a calculation module. The invention mainly aims at the inspection equipment, provides a light, efficient and accurate disease identification and positioning method, and is convenient to deploy into a portable inspection instrument.
Referring to fig. 1, a flowchart of a road disease identifying and positioning method provided in an embodiment of the present application is shown, and the method may include the following steps:
and 101, acquiring a target inspection image, and preprocessing the target inspection image.
As in fig. 2, a pipeline from a single picture algorithm is shown, where semantic segmentation, object detection is mainly inferred by a deep neural network model-based approach. Specifically:
in the embodiment of the application, the portable inspection equipment with low cost adopts a camera, and a common optical lens and a consumer-grade image sensor are often adopted, so that the collected inspection image is inevitably influenced by external environment and noise exists, and therefore, the image preprocessing is extremely important. The background difference method and the detail enhancement method are main methods for image preprocessing, and are mainly used for filtering out the reflection of a windshield, the reflection of articles in a vehicle and the shielding part of a vehicle head for a vehicle-mounted camera, optimizing sharpness and contrast and further modifying a picture. In addition, in the case where the vehicle speed is high, the influence of camera shake may be increased, and the preprocessing section may perform video anti-shake by reducing the exposure time and the motion compensation and motion estimation algorithms.
In the embodiment of the application, aiming at the picture after image preprocessing, road surface and non-road surface areas are segmented through a semantic segmentation algorithm, dynamic elements such as vehicles, pedestrians and non-motor vehicles are distinguished, objects in the image are segmented based on pixel level, and clean input picture data is provided for next target detection. Because the semantic segmentation model consumes relatively much computational resources, the semantic segmentation model can be trained and inferred based on lower resolution in order to improve model operation efficiency.
After the picture with only road surface pixels is extracted, further preprocessing is needed, and the size of the road surface target is calculated first. After the processing of the step, the size of the input picture of the next-stage target detection model is reduced, and the reasoning speed is faster. As shown in fig. 3, a schematic diagram of the image segmentation and resizing process provided in this embodiment is provided, in which the left side is an original image, the middle is a semantically segmented picture, and the right side is a resized picture.
And 103, processing the road surface image through the target detection model to determine each disease target in the road surface.
In the embodiment of the application, the classified pixels after semantic segmentation are enlarged back to the original size and are AND-processed on the original image. After the segmented image passes through the object detection model, various disease objects are framed by bounding boxes, and the categories, the confidence and the positions of the disease objects in the original image are marked.
In recent years, there are great progress and breakthrough in the target detection algorithm, and the more mature algorithms are mainly divided into two types: one class is: the two-stage algorithm, R-CNN algorithm based on RegionPropos (R-CNN, fastR-CNN, fasterR-CNN, etc.);
another class is one-stage algorithms such as yolo, SDD, etc. The first type of algorithm has higher accuracy than the second type of algorithm, but the second type of algorithm has very high speed, is suitable for real-time detection, and selects an improved yolov5 algorithm in the second type of algorithm as a target recognition algorithm in the application.
The identification of individual disease targets in the road surface in this application may result in, for example, roads having pits, sinkings, etc.
Another alternative embodiment is given below:
Wherein the features include coordinates in the image, size of the dimensions, direction of movement.
And 203, calculating the probability that the corresponding disease targets in the front frame and the rear frame belong to the same target according to the similarity, and eliminating the repeated targets.
The algorithms in steps 101-103 are all performed for each frame of picture, and the frame rate of the video stream in the real-time shooting process of the camera is often up to 30fps. In this embodiment, under the condition that the inspection speed is slower, the same disease target may be continuously photographed by the camera for about 3 seconds, which means that the front end target detection algorithm will output 90 targets; if traffic lights or traffic jams are encountered, the target will be photographed longer and the number of repeated detections is greater. If all the targets and pictures are uploaded to the background service, huge pressure and resource waste are caused to the background. There is a need for efficient deduplication of front-end detected ethical targets using a multi-target tracking algorithm.
For each detected target, calculating various characteristics of the target, such as coordinates, size, movement direction and the like in a picture by a multi-target tracking algorithm, and then calculating the probability that each object in the front frame and the rear frame belongs to the same target according to the similarity; and finally, associating the same target object and distributing a digital ID, and if the target ID is unchanged, changing the target object into the same disease target, so as to avoid repeated uploading.
As the accuracy of the front-end target detection model is generally only 60%, the multi-target tracking algorithm can perform temporal association by analyzing the front and rear frames of the video, and the recognition rate of the final target can be improved, and the co-workers can reject the repeated target.
Step 204, when the inspection video stream is acquired, the position data of each frame of inspection image is read and acquired.
The coordinate data at least comprises longitude and latitude, course and speed.
And 205, marking each disease target in the current frame according to the position data read by each frame of inspection image.
When the camera acquires each frame of picture, the equipment synchronously reads the coordinate data of the current positioning module, including longitude and latitude, heading, speed and the like. In the object detection algorithm, each detected object is assigned such positioning information. This information can be used as a feature of target deduplication in addition to locating the geographic location of the target.
The image segmentation and target detection algorithm model is based on a deep neural network model and mainly depends on GPU resources of a computing module. And if the multi-target tracking and de-duplication algorithm also adopts GPU real-time calculation, the hardware configuration of the equipment needs to be continuously improved, and the deployment cost is increased. But the road-defect target does not always appear in the video stream, but intermittently appears. The invention provides a multi-target tracking algorithm integrating positioning information, which is used for placing disease target information and pictures detected in real time into a queue cache pool of a first-in first-out (FIFO) type, wherein the target tracking algorithm asynchronously processes the disease target in the cache pool in another thread, so that the disease target can be reported in a delayed manner. Because the road disease detection does not need second-level detection reporting, the system function is not influenced by minute-level delay, and the throughput of the whole system to video streams can be improved.
As shown in fig. 4, a schematic diagram of multi-target tracking recognition provided in the present application is provided. Specifically:
the GNSS speed information is used to track the object, and since the device is mounted on a vehicle and the detected object is a road surface, the relative movement of the road surface and the vehicle is relatively simple, the device is roughly divided into two cases of forward movement and turning.
Calculating a vehicle running state by using GNSS:
C diff ＝C now -C pre
C now the current driving direction angle;
C pre the last read driving direction angle;
C diff the difference value between the current running direction and the last running direction is stored in a buffer pool with the length of N, and the value of N is 10;
utilizing C in a cache pool diff Calculating a mean value and a direction trend of the data:
n is the size of the cache pool;
C i i is more than or equal to 0 and less than or equal to N, which are the running direction difference values stored in the buffer pool;
C m is the average value of the running direction difference values in the buffer pool;
when C m T is the right turn trend, C m T is a left turning trend, and T is less than or equal to C m T is not more than T and is a straight trend, and the T value and the speed are related:
v is the running speed measured by the equipment, T is set to 3 when the speed is relatively small and is set to 2 when the speed is increased to 30-60, and is set to 1 when the speed is greater than 60.
Disease target information identified by the model is x, y, width and height:
x: upper left abscissa of target
y: upper left ordinate of target
width: target width
height: target height
Due to the perspective principle of the near-far small of the camera lens, the area size of a frame of the same target nearest to the camera will be the largest. So in the FIFO buffer queue, the algorithm will choose the location information marked by the frame with the largest area as the location information of the target.
In summary, the multi-target tracking algorithm integrating the positioning information makes full use of the positioning module information, can more efficiently track and de-duplicate disease targets, and is convenient to deploy in vehicle-mounted portable inspection equipment.
Referring to fig. 5, a block diagram of a road disease identifying and positioning device 200 according to an embodiment of the present application is shown. As shown in fig. 5, the apparatus 200 may include:
a first obtaining module 201, configured to obtain a target inspection image, and pre-process the target inspection image;
the segmentation module 202 is configured to segment a road surface area and a non-road surface area according to the preprocessed target inspection image through a semantic segmentation model, and adjust the size of the segmented road surface area to obtain a road surface image;
and the determining module 203 is configured to process the road surface image through the target detection model, and determine each disease target in the road surface.
In an optional embodiment of the present application, the system further includes a second acquisition module, configured to acquire a patrol video stream, where the patrol video stream includes a plurality of patrol images;
the determining module is also used for determining each disease target in each inspection image and determining the characteristics of each disease target; wherein the features include coordinates, size, direction of motion in the image;
and calculating the probability that the corresponding disease targets in the front frame and the rear frame belong to the same target according to the similarity, and eliminating the repeated targets.
In an alternative embodiment of the present application, the apparatus further comprises:
when the inspection video stream is acquired, reading and acquiring position data of each frame of inspection image; the coordinate data at least comprises longitude and latitude, course and speed;
and marking each disease target in the current frame according to the position data read by each frame of inspection image.
The specific limitation of the road disease recognition and positioning device can be referred to the limitation of the road disease recognition and positioning method hereinabove, and the description thereof will not be repeated. The above-mentioned various modules in the road disease recognition and positioning device can be implemented in whole or in part by software, hardware and a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
In one embodiment, an electronic device is provided, which may be a patrol device, and an internal structure thereof may be as shown in fig. 6. The electronic device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the device is configured to provide computing and control capabilities. The memory of the device includes a non-volatile storage medium, an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer equipment is used for identifying and positioning data of road diseases. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program when executed by the processor is used for realizing a road disease identification and positioning method.
It will be appreciated by those skilled in the art that the structure shown in fig. 6 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In one embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the steps of the road disease identification positioning method described above.
The computer readable storage medium provided in this embodiment has similar principles and technical effects to those of the above method embodiment, and will not be described herein.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in M forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SyMchlimk) DRAM (SLDRAM), memory bus (RaMbus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples only represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the claims. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (10)
1. A method for identifying and locating road diseases, which is characterized by comprising the following steps:
acquiring a target inspection image, and preprocessing the target inspection image;
dividing a road surface area and a non-road surface area from the preprocessed target inspection image through a semantic segmentation model, and adjusting the size of the divided road surface area to obtain a road surface image;
and processing the road surface image through a target detection model to determine each disease target in the road surface.
2. The method according to claim 1, wherein the method further comprises:
obtaining a patrol video stream, wherein the patrol video stream comprises a plurality of patrol images;
determining each disease target in each inspection image, and determining the characteristics of each disease target; wherein the features include coordinates, size, direction of motion in the image;
and calculating the probability that the corresponding disease targets in the front frame and the rear frame belong to the same target according to the similarity, and eliminating the repeated targets.
3. The method according to claim 2, wherein the method further comprises:
when the inspection video stream is acquired, reading and acquiring position data of each frame of inspection image; wherein the coordinate data at least comprises longitude and latitude, course and speed;
and marking each disease target in the current frame according to the position data read by each frame of inspection image.
4. A method according to claim 3, wherein marking the disease targets in the current frame based on the position data read by each frame of inspection image comprises:
and (3) placing the real-time detected disease target information and the patrol image into a queue cache pool of the FIFO type, and asynchronously processing the disease target in the cache pool in another thread by a target tracking algorithm.
5. A method according to claim 3, wherein reading the position data at the time of acquisition of each frame of inspection image while the inspection video stream is acquired, comprises:
calculating a vehicle running state using the position data:
C diff ＝C now -C pre
wherein C is now The current driving direction angle; c (C) pre The last read driving direction angle; c (C) diff The difference between the current driving direction and the previous driving direction.
6. A road disease identification and location device, the device comprising:
the first acquisition module is used for acquiring a target inspection image and preprocessing the target inspection image;
the segmentation module is used for segmenting a road surface area and a non-road surface area from the preprocessed target inspection image through a semantic segmentation model, and performing size adjustment on the segmented road surface area to obtain a road surface image;
and the determining module is used for processing the road surface image through the target detection model and determining each disease target in the road surface.
7. The apparatus of claim 6, wherein the apparatus further comprises:
the second acquisition module is used for acquiring a patrol video stream, wherein the patrol video stream comprises a plurality of patrol images;
the determining module is also used for determining each disease target in each inspection image and determining the characteristics of each disease target; wherein the features include coordinates, size, direction of motion in the image;
and calculating the probability that the corresponding disease targets in the front frame and the rear frame belong to the same target according to the similarity, and eliminating the repeated targets.
8. The apparatus of claim 7, wherein the apparatus further comprises:
when the inspection video stream is acquired, reading and acquiring position data of each frame of inspection image; wherein the coordinate data at least comprises longitude and latitude, course and speed;
and marking each disease target in the current frame according to the position data read by each frame of inspection image.
9. An electronic device comprising a memory and a processor, the memory storing a computer program which when executed by the processor implements the road fault identification and localization method of any one of claims 1 to 5.
10. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, implements the road condition identifying and locating method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310089453.5A CN116012764A (en) | 2023-02-09 | 2023-02-09 | Road disease identification and positioning method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310089453.5A CN116012764A (en) | 2023-02-09 | 2023-02-09 | Road disease identification and positioning method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116012764A true CN116012764A (en) | 2023-04-25 |
Family
ID=86021559
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310089453.5A Pending CN116012764A (en) | 2023-02-09 | 2023-02-09 | Road disease identification and positioning method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116012764A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117395378A (en) * | 2023-12-07 | 2024-01-12 | 北京道仪数慧科技有限公司 | Road product acquisition method and acquisition system |
-
2023
- 2023-02-09 CN CN202310089453.5A patent/CN116012764A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117395378A (en) * | 2023-12-07 | 2024-01-12 | 北京道仪数慧科技有限公司 | Road product acquisition method and acquisition system |
| CN117395378B (en) * | 2023-12-07 | 2024-04-09 | 北京道仪数慧科技有限公司 | Road product acquisition method and acquisition system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7430277B2 (en) | Obstacle detection method and apparatus, computer device, and computer program | |
| CN109977782B (en) | Cross-store operation behavior detection method based on target position information reasoning | |
| CN111899227A (en) | Automatic railway fastener defect acquisition and identification method based on unmanned aerial vehicle operation | |
| WO2013186662A1 (en) | Multi-cue object detection and analysis | |
| CN103413444A (en) | Traffic flow surveying and handling method based on unmanned aerial vehicle high-definition video | |
| CN109489724A (en) | A kind of tunnel safe train operation environment comprehensive detection device and detection method | |
| CN107194946B (en) | FPGA-based infrared salient object detection method | |
| JP2015090679A (en) | Vehicle trajectory extraction method, vehicle region extraction method, vehicle speed estimation method, vehicle trajectory extraction program, vehicle region extraction program, vehicle speed estimation program, vehicle trajectory extraction system, vehicle region extraction system, and vehicle speed estimation system | |
| CN111967396A (en) | Processing method, device and equipment for obstacle detection and storage medium | |
| CN113903008A (en) | Ramp exit vehicle violation identification method based on deep learning and trajectory tracking | |
| CN112613387A (en) | Traffic sign detection method based on YOLOv3 | |
| CN116012764A (en) | Road disease identification and positioning method and device, electronic equipment and storage medium | |
| CN111079675A (en) | Driving behavior analysis method based on target detection and target tracking | |
| CN110276318A (en) | Nighttime road rains recognition methods, device, computer equipment and storage medium | |
| CN116228756B (en) | Method and system for detecting bad points of camera in automatic driving | |
| CN112101134A (en) | Object detection method and device, electronic device and storage medium | |
| CN115482483A (en) | Traffic video target tracking device, method and storage medium | |
| Kathirvel et al. | Systematic number plate detection using improved yolov5 detector | |
| CN110427911B (en) | Road detection method, device, equipment and storage medium | |
| Kolcheck et al. | Visual counting of traffic flow from a car via vehicle detection and motion analysis | |
| CN117422669A (en) | Road disease detection method based on artificial intelligent visual recognition | |
| Chen et al. | A hierarchical accident recognition method for highway traffic systems | |
| Hoogendoorn et al. | Traffic data collection from aerial imagery | |
| CN112906424B (en) | Image recognition method, device and equipment | |
| Tsai et al. | Four Categories Vehicle Detection in Hsuehshan Tunnel via Single Shot Multibox Detector |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |