CN113888481A - Bridge deck disease detection method, system, equipment and storage medium - Google Patents
Bridge deck disease detection method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN113888481A CN113888481A CN202111085252.5A CN202111085252A CN113888481A CN 113888481 A CN113888481 A CN 113888481A CN 202111085252 A CN202111085252 A CN 202111085252A CN 113888481 A CN113888481 A CN 113888481A
- Authority
- CN
- China
- Prior art keywords
- disease
- image
- bridge deck
- bridge
- original
- 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
- 201000010099 disease Diseases 0.000 title claims abstract description 365
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 title claims abstract description 365
- 238000001514 detection method Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 11
- 238000007689 inspection Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 6
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000002035 prolonged effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 208000037919 acquired disease Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a bridge deck disease detection method, a bridge deck disease detection system, bridge deck disease detection equipment and a storage medium, wherein the bridge deck disease detection method comprises the following steps: acquiring an original bridge deck image through a camera and sending the original bridge deck image to an FTP server; the FTP server transmits the original bridge deck image to the GPU server; the GPU server carries out disease marking on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image; identifying the type and the confidence coefficient of the diseases in the image of the disease bridge floor according to the disease identification model, and calculating the actual disease information of the bridge according to the ratio of the diseases in the image of the disease bridge floor; the processor generates a disease information table of the bridge according to the actual disease information and stores the disease information table into the database, wherein the disease information table comprises a plurality of pieces of disease information; and the Web front end acquires corresponding disease information according to the query conditions input by the user. The bridge maintenance system improves the bridge maintenance efficiency, meets the requirements on bridge monitoring and maintenance, and reduces the detection risk and the cost.
Description
Technical Field
The invention relates to the technical field of bridge detection, in particular to a bridge deck disease detection method, a bridge deck disease detection system, bridge deck disease detection equipment and a bridge deck disease detection storage medium.
Background
The bridge is an important component of a road, and the total number of the bridge in China currently exceeds one million seats. With the extension of the service period of the bridge, the bridge inevitably has damage on the appearance under the influence of the erosion of long-term sand wind and rain and snow and the overload phenomenon caused by the continuous increase of the vehicle flow. And once the damage on the appearance structure of the bridge is generated to be more than 1/3, the bridge is damaged to different degrees, and even the hidden trouble of functional failure is generated. Monitoring and maintenance of bridges is therefore also becoming increasingly important.
However, in the prior art, the disease is mainly observed by a detection person in a short distance by means of a mechanical platform such as a ladder, a scaffold or a bridge inspection vehicle, or the disease image is shot by an unmanned aerial vehicle, and then the disease record is manually updated, so that the detection efficiency is low, the time consumption is long, the cost is high, and the requirements for monitoring and maintaining the bridge are difficult to meet.
Disclosure of Invention
In view of the above, it is necessary to provide a bridge deck damage detection method, system, device and storage medium for solving the above technical problems.
A bridge deck disease detection method comprises the following steps: acquiring an original bridge deck image through the camera and sending the original bridge deck image to an FTP server, wherein the original bridge deck image is provided with a unique image identifier; the FTP server transmits the original bridge deck image to the GPU server; the GPU server carries out disease marking on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image; identifying the type and the confidence coefficient of the diseases in the disease bridge floor image according to the disease identification model, and calculating the actual disease information of the bridge according to the ratio of the diseases in the disease bridge floor image, wherein the actual disease information comprises the disease area, the disease width and the disease length; the processor generates a disease information table of the bridge according to the actual disease information and stores the disease information table into the database, wherein the disease information table comprises a plurality of pieces of disease information, and the disease information comprises an original bridge deck image name, an image identifier, shooting time, an image storage path, a disease name, a disease description and actual disease information; and the Web front end acquires corresponding disease information according to the query condition input by the user.
In one embodiment, the acquiring an original bridge deck image specifically includes: the method comprises the steps of collecting original bridge deck images through a camera arranged on an automatic bridge inspection vehicle, and uploading the original bridge deck images to an FTP server in real time.
In one embodiment, the target detection algorithm is the YOLOv4 algorithm.
In one embodiment, the identifying the disease type and obtaining the confidence level of the disease bridge floor image according to the disease identification model specifically includes: collecting a sample bridge deck data set, wherein the sample bridge deck data set comprises a sample bridge deck image and a sample bridge deck disease type, and performing disease marking on the sample bridge deck image; establishing an initial disease identification model according to a neural network algorithm, inputting the sample bridge deck data set into the initial disease identification model, and training the initial disease identification model to obtain a disease identification model; and inputting the disease bridge image into the disease identification model, and acquiring the disease type and the confidence coefficient of the disease bridge image.
In one embodiment, after the GPU server performs disease labeling on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image, before calculating actual disease information of the bridge according to a ratio of diseases in the disease bridge deck image, the method further includes: carrying out binarization processing on the disease bridge floor image, and determining the ratio of diseases in the disease bridge floor image according to the pixel level; and acquiring the actual disease information of the bridge by combining a scale between the actual bridge deck and the disease bridge deck image.
In one embodiment, the obtaining, by the Web front end, the corresponding disease information according to the query condition input by the user specifically includes: acquiring the disease information in the disease information table according to a query condition, wherein the query condition comprises at least one of a bridge number, a disease name, a start time and an end time; and performing priority ranking on the inquired disease information according to the set attributes, and displaying the disease information according to a ranking result.
A bridge deck disease detection system, comprising: the system comprises a camera, an FTP server, a GPU server, a processor, a database and a Web front end; the camera is connected with the FTP server, the GPU server and the processor are mutually connected, and the processor is respectively connected with the database and the Web front end; the camera is used for acquiring an original bridge deck image and sending the original bridge deck image to the FTP server, and the original bridge deck image is provided with a unique image identifier; the FTP server is used for storing the original image and transmitting the original bridge deck image to the GPU server; the GPU server is used for marking diseases on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image; the GPU server is also used for identifying the type and the confidence coefficient of the diseases in the disease bridge floor image through a disease identification model, calculating the actual disease information of the bridge according to the ratio of the diseases in the disease bridge floor image, and sending the actual disease information to the processor, wherein the actual disease information comprises the disease area, the disease width and the disease length; the processor is used for generating a disease information table of the bridge according to the actual disease information, the disease information table comprises a plurality of pieces of disease information, and the disease information comprises an original bridge deck image name, an image identifier, shooting time, an image storage path, a disease name, disease description and actual disease information; the database 50 is used for storing the disease information table; and the Web front end is used for acquiring corresponding disease information according to the query condition input by the user.
An apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of a bridge deck disease detection method as described in the above embodiments when executing the program.
A storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a bridge deck disease detection method as described in the various embodiments above.
Compared with the prior art, the invention has the advantages and beneficial effects that: the method can analyze the actual damage degree of the bridge according to the acquired damage information, improve the bridge overhaul efficiency and meet the requirements on bridge monitoring and maintenance; meanwhile, the detection risk and cost are reduced, the disease information is accurately positioned, bridge diseases are conveniently found and processed in time, and the service life of the bridge is prolonged.
Drawings
FIG. 1 is a schematic flow chart of a bridge deck disease detection method according to an embodiment;
FIG. 2 is a schematic structural diagram of a bridge deck disease detection system according to an embodiment;
FIG. 3 is a schematic view of an application scenario of a bridge deck disease detection system in an embodiment;
fig. 4 is a schematic diagram of the internal structure of the apparatus in one embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings by way of specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In one embodiment, as shown in fig. 1, a bridge deck disease detection method is provided, which includes the following steps:
and S101, acquiring an original bridge deck image through a camera and sending the original bridge deck image to an FTP server, wherein the original bridge deck image is provided with a unique image identifier.
Specifically, the camera can be installed on an automatic bridge inspection vehicle, and the camera acquires images of the side surface and the bottom surface of the bridge through the movement of the automatic bridge inspection vehicle, acquires an original bridge deck image, and uploads the acquired original bridge deck image to the FTP server in real time. An original image folder can be set in the FTP server, and the original bridge deck image is saved in the original image folder.
And step S102, the FTP server transmits the original bridge deck image to the GPU server.
Specifically, the GPU server sends a request to the FTP server at regular time, inquires whether a newly uploaded original bridge deck image exists, the FTP server responds to the timing request of the GPU server, scans whether a new original bridge deck image exists in an original image folder, if the new original bridge deck image exists, the FTP protocol is used for downloading the new original bridge deck image to the GPU server for storage, and the original bridge deck image in the FTP server is deleted.
And S103, the GPU server performs disease marking on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image.
Specifically, after acquiring an original bridge deck image, the GPU server performs disease labeling on the original bridge deck image by using a single-stage target detection algorithm, such as the YOLOv4 algorithm, so as to acquire a diseased bridge deck image and filter out a normal bridge deck image. The object detection problem is converted into a regression problem of a space-separated boundary box and related probability through a YOLOv4 algorithm, and the boundary box and class probability of an object in an original bridge deck image are directly predicted in one-time evaluation, so that end-to-end optimization is directly performed on the detection performance.
And step S104, identifying the type and the confidence coefficient of the diseases in the image of the disease bridge floor according to the disease identification model, and calculating the actual disease information of the bridge according to the ratio of the diseases in the image of the disease bridge floor, wherein the actual disease information comprises the area, the width and the length of the diseases.
Specifically, a pre-trained disease identification model is arranged in the GPU server, the disease type and the confidence coefficient in the disease bridge floor image are determined according to the disease identification model, and the format of returned data can be { x }1The horizontal coordinate of the upper left point; y is1The vertical coordinate of the upper left point; x is the number of2The horizontal coordinate of the right lower point; y is2The vertical coordinate of the right lower point; cls _ id: a category; cls _ conf: confidence of category }.
After the disease type and the confidence coefficient are determined, the disease bridge deck image is subjected to binarization processing, the ratio of a disease part to the disease bridge deck image is counted on a pixel level, conversion is carried out according to a scale between the disease image and an actual bridge deck, and actual disease information is obtained, wherein the actual disease information can comprise a disease area, a disease span and a disease length. Finally, uploading the disease bridge deck image with the disease and completed marking to a disease image folder of the FTP server for storage; and meanwhile, encapsulating the disease information corresponding to each disease bridge deck image, and sending the disease information to the processor through an HTTP (hyper text transport protocol), wherein the disease information can comprise an original bridge deck image name, an image identifier, a disease name and a disease description.
And S105, generating a disease information table of the bridge according to the actual disease information by the processor, and storing the disease information table into a database, wherein the disease information table comprises a plurality of pieces of disease information, and the disease information comprises an original bridge deck image name, an image identifier, shooting time, an image storage path, a disease name, a disease description and actual disease information.
Specifically, after receiving the disease information, the processor processes the disease information into a whole, extracts the shooting time of the image from the original bridge deck image, extracts a plurality of diseases contained in one disease bridge deck image and corresponding descriptions and confidence degrees of the diseases, forms a disease information table, and stores the disease information table in the database.
And step S106, the Web front end acquires corresponding disease information according to the query condition input by the user.
Specifically, the Web front end inputs a query condition through a call interface, and queries corresponding disease information, where the query condition may be at least one of a bridge number, a disease name, a start time, and an end time.
Specifically, the processor responds to a query request of a Web front end, performs format conversion on a received query condition, and sends a data query request to a database; the database responds to the query request of the processor, queries according to the query conditions, acquires corresponding disease information from the disease information table, and returns the corresponding disease information to the processor; the processor returns the inquired disease information to the Web front end; after acquiring the disease data, the Web front end calls a processor interface through an image storage address to send an image acquisition request, performs priority sequencing according to attributes determined by a user, such as the time of occurrence of the disease or the severity of the disease, and intelligently sequences and displays returned disease information.
In the embodiment, an original bridge deck image is collected through a camera and sent to an FTP server, the FTP server stores the original bridge deck image and transmits the original bridge deck image to a GPU server, and the GPU server performs disease marking on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image; calculating actual disease information of the bridge according to the disease type and confidence coefficient in the disease bridge floor image and the ratio of the diseases in the disease bridge floor image, generating a disease information table of the bridge according to the actual disease information by a processor, storing the disease information table into a database, wherein the disease information table comprises a plurality of disease information, acquiring corresponding disease information by a Web front end according to query conditions input by a user, sending the corresponding disease information to a user terminal, analyzing the actual disease degree of the bridge by the user according to the acquired disease information, improving the bridge overhauling efficiency and meeting the requirements on bridge monitoring and maintenance; meanwhile, the detection risk and cost are reduced, the disease information is accurately positioned, bridge diseases are conveniently found and processed in time, and the service life of the bridge is prolonged.
Wherein, step S101 specifically includes: the method comprises the steps of collecting an original bridge deck image through a camera arranged on an automatic bridge inspection vehicle, and uploading the original bridge deck image to an FTP server in real time.
Specifically, the automatic bridge inspection vehicle is used for carrying a camera, images are captured at the detection frequency of the bridge floor, the side face and the bottom face of the bridge can be subjected to image acquisition, and the shot original bridge floor images are uploaded to an FTP server in real time.
Wherein, step S104 specifically includes: collecting a sample bridge deck data set, wherein the sample bridge deck data set comprises a sample bridge deck image and a sample bridge deck disease type, and performing disease marking on the sample bridge deck image; establishing an initial disease identification model according to a neural network algorithm, inputting a sample bridge deck data set into the initial disease identification model, training the initial disease identification model, and obtaining a disease identification model; and inputting the disease bridge image into a disease identification model, and acquiring the disease type and the confidence coefficient of the disease bridge image.
Specifically, collecting a sample data set, wherein the sample data set comprises a sample bridge deck image and a sample bridge deck disease type, the sample bridge deck image comprises corresponding diseases, and the sample bridge deck image is subjected to disease marking through a target detection algorithm; establishing an initial disease identification model according to a neural network algorithm, training the initial disease identification model through a sample data set, and obtaining a disease identification model; and inputting the disease bridge image subjected to disease marking into a disease identification model, and acquiring the disease type and the confidence coefficient of the disease bridge image.
The confidence coefficient is the credibility of the disease type judged by the disease identification model, and when a user checks the original bridge deck image and the disease type, the disease type in the original bridge deck image can be comprehensively judged and analyzed by combining the confidence coefficient, so that the accuracy of bridge deck disease judgment is improved.
After step S103 and before step S104, the method further includes: carrying out binarization processing on the disease bridge floor image, and determining the ratio of diseases in the disease bridge floor image according to the pixel level; and acquiring actual disease information of the bridge by combining a scale between the actual bridge deck image and the disease bridge deck image.
Specifically, after determining the type and confidence of the diseases in the disease bridge floor image, performing binarization processing on the disease bridge floor image, determining the ratio of the diseases in the disease bridge floor according to the pixel level, combining the ratio and the scale between the disease bridge floor image and the actual bridge floor, obtaining actual disease information of the bridge, and judging the disease degree of the bridge according to the actual disease information.
Wherein, step S106 specifically includes: acquiring disease information in a disease information table according to query conditions, wherein the query conditions comprise at least one of bridge numbers, disease names, start time and end time; and performing priority ranking on the inquired disease information according to the set attributes, and displaying the disease information according to a ranking result.
Specifically, a user inputs a query condition according to the Web front end, the processor determines disease information meeting the query condition in a disease information table according to the query condition, and returns the disease information to the Web front end; and the Web front end carries out priority ranking on the inquired disease information according to attributes set by a user, such as disease occurrence time, disease severity and the like, and displays the disease information in sequence according to a ranking result. The processor can also log in an FTP server according to the image storage address in the Web request, acquire the disease bridge image from the disease image folder, cache the disease bridge image in the memory through the input and output stream, and return the disease bridge image cached in the memory to the Web front end for the user to check.
As shown in fig. 2, there is provided a bridge deck disease detection system, including: the camera 10, FTP server 20, GPU server 30, processor 40, database 50 and Web front end 60; the camera 10 is connected with the FTP server 20, the GPU server 30 and the processor 40 are connected with each other, and the processor 40 is respectively connected with the database 50 and the Web front end 60; the camera 10 is used for acquiring an original bridge deck image and sending the original bridge deck image to the FTP server 20, wherein the original bridge deck image is provided with a unique image identifier; the GPU server 30 is used for marking diseases on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image; the GPU server 30 is further configured to identify a disease type and a confidence level in the disease bridge floor image through the disease identification model, calculate actual disease information of the bridge according to a ratio of the diseases in the disease bridge floor image, and send the actual disease information to the processor 40, where the actual disease information includes a disease area, a disease width, and a disease length; the processor 40 is configured to generate a disease information table of the bridge according to the actual disease information, where the disease information table includes a plurality of pieces of disease information, and the disease information includes an original bridge deck image name, an image identifier, shooting time, an image storage path, a disease name, a disease description, and actual disease information; the database 50 is used for storing a disease information table; the Web front end 60 is configured to obtain corresponding disease information according to a query condition input by a user.
In the embodiment, the FTP server 20 is connected with the camera 10 and the processor 40, and stores the processed damaged bridge floor image for the user to inquire; the GPU server 30 carries out intelligent disease detection and identification on the original bridge deck image, and simultaneously backs up all the original bridge deck images uploaded by the camera 10 to ensure the integrity of original data; the database 50 stores corresponding disease information; the system has strong coupling and high safety, is convenient for detection personnel to quickly determine the positions and types of diseases, and timely overhauls, improves the detection efficiency and reduces the detection cost.
As shown in fig. 3, which is a schematic view of an application scenario of a bridge deck disease detection system, a camera 10 collects an original bridge deck image and uploads the original bridge deck image to an FTP server 20 in real time, the FTP server 20 transmits the original bridge deck image to a GPU server 30, and the GPU server 30 processes the original bridge deck image to obtain a bridge disease information table and stores the bridge disease information table in a database 50; a user enters the Web front end 60 through cloud server connection based on various terminals, sends a query request to the Web front end 60, wherein the query request carries query conditions, and the Web front end 60 requests the database 50 to acquire disease information according to the query conditions and returns the disease information to the terminal equipment; meanwhile, Web can also request the FTP server to obtain the original bridge floor image corresponding to the disease information, and return to the terminal equipment; the user can fully analyze the bridge disease condition according to the disease information and the original bridge deck image, and accurately position the disease information, so that the bridge diseases are timely treated, and the service life of the bridge is prolonged.
Specifically, when a user needs to query for bridge diseases, the user can enter the Web front end 60 according to the terminal, input query conditions, send a query request to the processor 40 after the Web front end 60 receives the query conditions, and query the database 50 and the FTP server after the processor 40 receives the request of the Web front end 60, and return a query result to the Web front end 60 for display. And the user can quickly position the bridge deck diseases according to the information displayed by the Web for overhauling and maintaining.
In one embodiment, a device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 4. The device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the device is configured to provide computing and control capabilities. The memory of the device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the device is used for storing configuration templates and also can be used for storing target webpage data. The network interface of the device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a bridge deck disease detection method.
Those skilled in the art will appreciate that the configuration shown in fig. 3 is a block diagram of only a portion of the configuration associated with the present application and does not constitute a limitation on the devices to which the present application may be applied, and that a particular device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, a storage medium may also be provided, the storage medium storing a computer program comprising program instructions which, when executed by a computer, may be part of a bridge deck disease detection system as mentioned above, cause the computer to perform the method according to the preceding embodiment.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
It will be apparent to those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be centralized on a single computing device or distributed across a network of computing devices, and optionally they may be implemented in program code executable by a computing device, such that they may be stored on a computer storage medium (ROM/RAM, magnetic disks, optical disks) and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The foregoing is a more detailed description of the present invention that is presented in conjunction with specific embodiments, and the practice of the invention is not to be considered limited to those descriptions. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.
Claims (9)
1. A bridge deck disease detection method is characterized by comprising the following steps:
acquiring an original bridge deck image through the camera and sending the original bridge deck image to an FTP server, wherein the original bridge deck image is provided with a unique image identifier;
the FTP server transmits the original bridge deck image to the GPU server;
the GPU server carries out disease marking on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image;
identifying the type and the confidence coefficient of the diseases in the disease bridge floor image according to the disease identification model, and calculating the actual disease information of the bridge according to the ratio of the diseases in the disease bridge floor image, wherein the actual disease information comprises the disease area, the disease width and the disease length;
the processor generates a disease information table of the bridge according to the actual disease information and stores the disease information table into the database, wherein the disease information table comprises a plurality of pieces of disease information, and the disease information comprises an original bridge deck image name, an image identifier, shooting time, an image storage path, a disease name, a disease description and actual disease information;
and the Web front end acquires corresponding disease information according to the query condition input by the user.
2. The bridge deck disease detection method according to claim 1, wherein the acquiring of the original bridge deck image specifically comprises:
the method comprises the steps of collecting original bridge deck images through a camera arranged on an automatic bridge inspection vehicle, and uploading the original bridge deck images to an FTP server in real time.
3. The bridge deck disease detection method according to claim 1, wherein the target detection algorithm is a YOLOv4 algorithm.
4. The bridge deck disease detection method according to claim 1, wherein the identifying the disease type of the disease bridge deck image according to the disease identification model and obtaining the confidence coefficient specifically comprises:
collecting a sample bridge deck data set, wherein the sample bridge deck data set comprises a sample bridge deck image and a sample bridge deck disease type, and performing disease marking on the sample bridge deck image;
establishing an initial disease identification model according to a neural network algorithm, inputting the sample bridge deck data set into the initial disease identification model, and training the initial disease identification model to obtain a disease identification model;
and inputting the disease bridge image into the disease identification model, and acquiring the disease type and the confidence coefficient of the disease bridge image.
5. The bridge deck disease detection method according to claim 1, wherein after the GPU server performs disease labeling on the original bridge deck image according to a target detection algorithm to obtain a diseased bridge deck image, before calculating actual disease information of the bridge according to a ratio of diseases in the diseased bridge deck image, the method further comprises:
carrying out binarization processing on the disease bridge floor image, and determining the ratio of diseases in the disease bridge floor image according to the pixel level;
and acquiring the actual disease information of the bridge by combining a scale between the actual bridge deck and the disease bridge deck image.
6. The bridge deck disease detection method according to claim 1, wherein the Web front end obtains corresponding disease information according to a query condition input by a user, and specifically comprises:
acquiring the disease information in the disease information table according to a query condition, wherein the query condition comprises at least one of a bridge number, a disease name, a start time and an end time;
and performing priority ranking on the inquired disease information according to the set attributes, and displaying the disease information according to a ranking result.
7. A bridge floor disease detection system, characterized by, includes: the system comprises a camera, an FTP server, a GPU server, a processor, a database and a Web front end;
the camera is connected with the FTP server, the GPU server and the processor are mutually connected, and the processor is respectively connected with the database and the Web front end;
the camera is used for acquiring an original bridge deck image and sending the original bridge deck image to the FTP server, and the original bridge deck image is provided with a unique image identifier;
the FTP server is used for storing the original image and transmitting the original bridge deck image to the GPU server;
the GPU server is used for marking diseases on the original bridge deck image according to a target detection algorithm to obtain a disease bridge deck image;
the GPU server is also used for identifying the type and the confidence coefficient of the diseases in the disease bridge floor image through a disease identification model, calculating the actual disease information of the bridge according to the ratio of the diseases in the disease bridge floor image, and sending the actual disease information to the processor, wherein the actual disease information comprises the disease area, the disease width and the disease length;
the processor is used for generating a disease information table of the bridge according to the actual disease information, the disease information table comprises a plurality of pieces of disease information, and the disease information comprises an original bridge deck image name, an image identifier, shooting time, an image storage path, a disease name, disease description and actual disease information;
the database 50 is used for storing the disease information table;
and the Web front end is used for acquiring corresponding disease information according to the query condition input by the user.
8. An apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 6 are implemented when the computer program is executed by the processor.
9. A storage medium having a computer program stored thereon, the computer program, when being executed by a processor, realizing the steps of the method of any one of claims 1 to 6.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110656952 | 2021-06-11 | ||
| CN2021106569529 | 2021-06-11 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113888481A true CN113888481A (en) | 2022-01-04 |
Family
ID=79009556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111085252.5A Pending CN113888481A (en) | 2021-06-11 | 2021-09-16 | Bridge deck disease detection method, system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113888481A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115082802A (en) * | 2022-08-18 | 2022-09-20 | 深圳市城市交通规划设计研究中心股份有限公司 | Road disease identification method, device, equipment and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120033851A1 (en) * | 2010-04-22 | 2012-02-09 | Shen-En Chen | Spatially integrated aerial photography for bridge, structure, and environmental monitoring |
| CN109300126A (en) * | 2018-09-21 | 2019-02-01 | 重庆建工集团股份有限公司 | A kind of bridge defect high-precision intelligent detection method based on spatial position |
| CN109342447A (en) * | 2018-11-30 | 2019-02-15 | 重庆红岩建设机械制造有限责任公司 | A kind of bridge tower concrete reinforcement rust detection and the intelligent trolley of surface defect scanning |
| CN110070008A (en) * | 2019-04-04 | 2019-07-30 | 中设设计集团股份有限公司 | Bridge disease identification method adopting unmanned aerial vehicle image |
| CN111191714A (en) * | 2019-12-28 | 2020-05-22 | 浙江大学 | Intelligent identification method for bridge appearance damage diseases |
-
2021
- 2021-09-16 CN CN202111085252.5A patent/CN113888481A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120033851A1 (en) * | 2010-04-22 | 2012-02-09 | Shen-En Chen | Spatially integrated aerial photography for bridge, structure, and environmental monitoring |
| CN109300126A (en) * | 2018-09-21 | 2019-02-01 | 重庆建工集团股份有限公司 | A kind of bridge defect high-precision intelligent detection method based on spatial position |
| CN109342447A (en) * | 2018-11-30 | 2019-02-15 | 重庆红岩建设机械制造有限责任公司 | A kind of bridge tower concrete reinforcement rust detection and the intelligent trolley of surface defect scanning |
| CN110070008A (en) * | 2019-04-04 | 2019-07-30 | 中设设计集团股份有限公司 | Bridge disease identification method adopting unmanned aerial vehicle image |
| CN111191714A (en) * | 2019-12-28 | 2020-05-22 | 浙江大学 | Intelligent identification method for bridge appearance damage diseases |
Non-Patent Citations (2)
| Title |
|---|
| 韩晓健等: "卷积神经网络在桥梁结构表面病害检测中的应用研究", 结构工程师, no. 02 * |
| 龚尚福: "《计算机网络及应用》", 31 March 2008, 中国矿业大学出版社, pages: 210 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115082802A (en) * | 2022-08-18 | 2022-09-20 | 深圳市城市交通规划设计研究中心股份有限公司 | Road disease identification method, device, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109255288A (en) | A kind of road surface breakage detection method, device and terminal device | |
| CN110784628B (en) | Image data acquisition processing method and system, intelligent camera and server | |
| CN106771856B (en) | Electric power transmission line lightning stroke point determination method based on unmanned aerial vehicle technology | |
| CN107014827B (en) | Transmission line defect analysis method, device and system based on image processing | |
| CN109214280B (en) | Shop identification method and device based on street view, electronic equipment and storage medium | |
| CN108319926A (en) | A kind of the safety cap wearing detecting system and detection method of building-site | |
| CN111918039B (en) | Artificial intelligence high risk operation management and control system based on 5G network | |
| CN108986448B (en) | Traffic facility management method and terminal equipment | |
| CN112613569B (en) | Image recognition method, training method and device for image classification model | |
| CN112749589B (en) | Method and device for determining routing inspection path and storage medium | |
| CN113470374A (en) | Vehicle overspeed monitoring method and device, computer equipment and storage medium | |
| CN114140999A (en) | Engineering supervision system based on communication of Internet of things | |
| CN113888481A (en) | Bridge deck disease detection method, system, equipment and storage medium | |
| CN115169602A (en) | Maintenance method and device for power equipment, storage medium and computer equipment | |
| CN112800837B (en) | Bird recognition system and method | |
| CN115187880A (en) | Communication optical cable defect detection method and system based on image recognition and storage medium | |
| CN115880631A (en) | Power distribution station fault identification system, method and medium | |
| CN114038040A (en) | Machine room inspection monitoring method, device and equipment | |
| CN114966631A (en) | Fault diagnosis and processing method and device for vehicle-mounted laser radar, medium and vehicle | |
| CN113705442A (en) | Outdoor large-board advertising picture monitoring and identifying system and method | |
| CN113554610A (en) | Photovoltaic module operation state detection method and application device thereof | |
| CN114169404A (en) | Method for intelligently acquiring quantitative information of slope diseases based on images | |
| CN108363985B (en) | Target object perception system testing method and device and computer readable storage medium | |
| CN112307120A (en) | Information management server, information management method, and information management system | |
| CN116912721B (en) | Power distribution network equipment body identification method and system based on monocular stereoscopic vision |
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 |