CN115512098B - Bridge electronic inspection system and inspection method - Google Patents
Bridge electronic inspection system and inspection method Download PDFInfo
- Publication number
- CN115512098B CN115512098B CN202211177540.8A CN202211177540A CN115512098B CN 115512098 B CN115512098 B CN 115512098B CN 202211177540 A CN202211177540 A CN 202211177540A CN 115512098 B CN115512098 B CN 115512098B
- Authority
- CN
- China
- Prior art keywords
- data
- inspection
- layer
- target
- map
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Bridges Or Land Bridges (AREA)
Abstract
The application relates to the technical field of inspection management, in particular to a management method for bridge inspection, and specifically relates to an electronic inspection system and an inspection method for bridges; the system comprises a handheld terminal, a data processing terminal, a data storage terminal and a plurality of data marking points; the handheld terminal comprises a display module, a communication establishing module and a data acquisition module, wherein the communication establishing module receives a marking signal of the data marking point in a preset range and establishes communication, and the data acquisition module acquires real-time data in a preset range corresponding to the data marking point based on the establishment of the communication; the data processing terminal processes the real-time data acquired by the data acquisition module to obtain a processing result, and the data storage terminal is used for storing the processing result and the corresponding data processing terminal tag and the corresponding data marking point tag; the data processing terminal is internally provided with a data detection model, and the data detection model is used for detecting the real-time data to obtain a processing result.
Description
Technical Field
The application relates to the technical field of inspection management, in particular to a management method for bridge inspection, and specifically relates to an electronic inspection system and an inspection method for bridges.
Background
The number of bridges entering the service period in China is continuously increased, and the technical condition of the bridges is a main factor affecting the operation safety of the bridges. In order to ensure the safe operation of the bridge structure, the bridge maintenance management department standardizes the maintenance management work of the highway bridge by establishing a bridge inspection system. The bridge inspection work is mainly visual inspection, if necessary, a measuring tool is used in an auxiliary way, and professional inspection personnel inspect the bridge and fill in an inspection record table so as to facilitate bridge management and maintenance units to timely and accurately master the technical condition and disease development condition of each active bridge. The bridge inspection is a key and basic work for guaranteeing the safe operation of the bridge, and has important significance for bridge maintenance and management decision. The bridge has the characteristics of large construction volume and complex bridge position environment, the traditional bridge inspection is mainly based on visual inspection, inspection personnel are required to fill in forms on site and collect bridge disease photographs, the quality of inspection data is influenced by factors in various aspects such as personnel quality, environmental conditions and the like, and the accuracy is difficult to guarantee. If the collected data can not timely and effectively reflect the structural condition, the accurate evaluation of the structural operation condition of the bridge can be affected.
The traditional bridge inspection method can not meet the current management requirements of the bridge inspection for refinement, high efficiency and scientificity, and meanwhile, a bridge management unit also has supervision requirements on bridge inspection business. In order to truly grasp the presence inspection condition, inspection progress, bridge disease distribution and severity of inspection personnel, the bridge management and maintenance capability is further enhanced by means of informatization. Therefore, technologies such as the Internet of things, the mobile internet and an electronic map are required to be comprehensively utilized, a set of bridge inspection system capable of meeting higher management requirements is built, and bridge management and maintenance levels are comprehensively improved.
In the existing intelligent bridge inspection system, the functions of collection of inspection information, storage of inspection data, issuing of inspection instructions and the like are mainly realized through an inspection terminal. However, since the main object of inspection is an inspection personnel, the inspection cost of the inspection personnel is reduced, the inspection efficiency is improved, and the bridge abnormality recognition in the inspection process is still not realized in the prior art. Therefore, an electronic system capable of assisting inspection personnel in inspection is needed, bridge inspection work is standardized, and inspection efficiency and data reliability are improved.
Disclosure of Invention
In order to solve the technical problems, the application provides the bridge electronic inspection system and the inspection method, which can accurately obtain the inspection position and the inspection target in the inspection process through a computer technology, improve the inspection cost of inspection personnel for the inspection target point, realize the identification of potential abnormality in the bridge inspection point through an abnormal data identification method, reduce the neglect of the potential abnormality caused by subjective reasons of the inspection personnel, and realize the comprehensiveness of integral maintenance.
In order to achieve the above purpose, the technical scheme adopted by the embodiment of the application is as follows:
in a first aspect, an electronic inspection system for a bridge includes a handheld terminal, a data processing terminal, a data storage terminal and a plurality of data marking points; the handheld terminal comprises a display module, a communication establishing module and a data acquisition module, wherein the communication establishing module receives a marking signal of the data marking point in a preset range and establishes communication, and the data acquisition module acquires real-time data in a preset range corresponding to the data marking point based on the establishment of the communication; the data processing terminal processes the real-time data acquired by the data acquisition module to obtain a processing result, and the data storage terminal is used for storing the processing result and the corresponding data processing terminal tag and the corresponding data marking point tag; the data processing terminal is internally provided with a data detection model, and the data detection model is used for detecting the real-time data to obtain a processing result; the data storage terminal comprises a plurality of storage spaces corresponding to the data marking points, wherein the storage spaces are used for storing data corresponding to the data marking points, each storage space comprises at least two storage subspaces, and each storage subspace is used for correspondingly storing the processing result types.
In a first possible implementation manner of the first aspect, the method further includes a terminal platform, where the terminal platform is connected to the data storage terminal, and the stored data in the data storage terminal is collected based on an acquisition policy.
With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner, the acquisition policy includes a first acquisition policy and a second acquisition policy, where the first acquisition policy and the second acquisition policy are respectively configured in the two storage subspaces.
With reference to the second possible implementation manner of the first aspect, in a third possible implementation manner, the storage subspace includes a first storage subspace and a second storage subspace, which are configured with a first acquisition policy and a second acquisition policy, respectively; the first acquisition strategy is synchronous acquisition, and the second acquisition strategy is asynchronous acquisition.
In a fourth possible implementation manner of the first aspect, the handheld terminal further includes a map marking module, wherein a plurality of layer maps are configured in the map marking module, and codes of corresponding data marking points are set in any one of the layer maps; the data marking points are configured with NFC digital tags, the NFC digital tags and the communication establishing module establish communication within a preset range, the map layer map is determined based on unique codes of the NFC digital tags, and the map layer map and corresponding marking point coordinates are displayed on the display module.
In a second aspect, a bridge electronic inspection method is applied to the bridge electronic inspection system described in any one of the above, and the method includes: the communication establishing module collects communication signals of the data marking points in a preset range and establishes communication with the data marking points; the map marking module establishes a corresponding unique code of the data marking point based on communication to determine a corresponding map layer map, and carries out the map layer map and coordinates corresponding to the data marking point on the display module; the data acquisition module acquires real-time data in a preset range corresponding to the data mark point based on the establishment of communication, wherein the real-time data comprises real-time image data meeting the preset range; transmitting the real-time image data to the data processing terminal, and performing anomaly detection on the real-time image data based on a data detection model in the data processing terminal to obtain a detection result; transmitting the detection result to the data storage terminal based on a storage strategy; and the terminal platform collects the data in the data storage terminal based on the collection strategy.
In a first possible implementation manner of the second aspect, the data acquisition module acquires real-time data within a preset range of corresponding data marking points based on establishment of communication, including: the data acquisition module determines a target acquisition frame and a target acquisition standard image based on the unique code of the data mark point; based on the target acquisition frame and the target acquisition standard image, image acquisition is carried out on the object to be patrolled and examined corresponding to the data mark point, and the real-time image data is obtained, and the method specifically comprises the following steps: identifying the real-time image data based on the target acquisition frame to obtain an identification result; based on the recognition result, judging whether the real-time image data meets a preset condition or not, specifically comprising: binarizing the real-time image data to obtain preprocessed image data; dividing the preprocessed image data to obtain a target processed image; and extracting edge coordinates of the target processing image, comparing the edge coordinates with preset standard edge coordinates, and determining whether the real-time image data meets preset conditions or not based on a comparison threshold.
In a second possible implementation manner of the second aspect, the detecting result is obtained by performing anomaly detection on the real-time image data based on a data detection model in the data processing terminal, where the data detection model includes a feature extraction network, a region recommendation network, and a classification network, and the method includes: extracting features in the real-time image data based on a feature extraction network to obtain a target feature map; positioning a plurality of candidate areas containing detection targets in the target feature map based on an area recommendation network; and acquiring the spatial characteristics of the candidate areas, modeling, judging whether the candidate areas contain probability distribution of target features based on the model, and determining whether the candidate areas contain the target features based on the probability distribution.
With reference to the second possible implementation manner of the second aspect, in a third possible implementation manner, the feature extraction network includes five sets of convolutional neural networks, and any one of the convolutional neural networks includes at least one convolutional layer, an activation function layer, and a pooling layer; the activation function in the activation function layer is a rule function.
With reference to the second possible implementation manner of the second aspect, in a fourth possible implementation manner, the area recommendation network includes a first convolution layer, and a second convolution layer and a third convolution layer connected to an output end of the first convolution layer, where the output end of the second convolution layer is connected to a normalized exponential function layer, and an output end of the normalized exponential function layer and an output end of the third convolution layer are connected to a combination layer.
In a third aspect, there is provided an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the method of any one of the preceding claims when executing the computer program.
In a fourth aspect, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program which, when executed by a processor, implements a method as claimed in any one of the preceding claims.
According to the technical scheme provided by the embodiment of the application, the accurate acquisition of the inspection point in the inspection process and the identification of whether the inspection point contains an abnormality or not are realized by setting a plurality of functional modules in the system. The inspection efficiency is improved, and the recognition accuracy of the abnormality in the inspection is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
The methods, systems, and/or programs in the accompanying drawings will be described further in terms of exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein the exemplary numbers represent like mechanisms throughout the various views of the drawings.
Fig. 1 is a schematic structural diagram of an electronic inspection system for bridges according to an embodiment of the present application.
Fig. 2 is a flow chart of a method of electronic inspection of a bridge according to some embodiments of the present application.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to better understand the above technical solutions, the following detailed description of the technical solutions of the present application is made by using the accompanying drawings and specific embodiments, and it should be understood that the specific features of the embodiments and the embodiments of the present application are detailed descriptions of the technical solutions of the present application, and not limiting the technical solutions of the present application, and the technical features of the embodiments and the embodiments of the present application may be combined with each other without conflict.
In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant teachings. It will be apparent, however, to one skilled in the art that the application can be practiced without these details. In other instances, well known methods, procedures, systems, components, and/or circuits have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.
The present application uses a flowchart to illustrate the execution of a system according to an embodiment of the present application. It should be clearly understood that the execution of the flowcharts may be performed out of order. Rather, these implementations may be performed in reverse order or concurrently. Additionally, at least one other execution may be added to the flowchart. One or more of the executions may be deleted from the flowchart.
Before describing embodiments of the present application in further detail, the terms and terminology involved in the embodiments of the present application will be described, and the terms and terminology involved in the embodiments of the present application will be used in the following explanation.
(1) In response to a condition or state that is used to represent the condition or state upon which the performed operation depends, the performed operation or operations may be in real-time or with a set delay when the condition or state upon which it depends is satisfied; without being specifically described, there is no limitation in the execution sequence of the plurality of operations performed.
(2) Based on the conditions or states that are used to represent the operations that are being performed, one or more of the operations that are being performed may be in real-time or with a set delay when the conditions or states that are being relied upon are satisfied; without being specifically described, there is no limitation in the execution sequence of the plurality of operations performed.
(3) Convolutional neural networks are a class of feedforward neural networks that involve convolutional computations and have a deep structure. Convolutional neural networks are proposed by biological Receptive Field (fielded) mechanisms. Convolutional neural networks are dedicated to neural networks that process data having a grid-like structure. For example, time-series data (which may be regarded as a one-dimensional grid formed by regularly sampling on a time axis) and image data (which may be regarded as a two-dimensional grid of pixels), the convolutional neural network employed in the present embodiment processes the image data.
According to the technical scheme provided by the embodiment of the application, the main application scene is the inspection of the bridge, and in a real situation, the inspection is mainly based on the judgment of the inspection target and abnormal conditions by manpower. The abnormal conditions of the bridge include structural abnormality of the bridge, which is mainly reflected in the defect that the key main body structure of the bridge has cracks or structures, and daily road surface abnormality includes road surface casting matters or other pollutants, wherein the detection of the structural defect of the key main body structure of the bridge, such as a segmented road surface connection position or a beam body, is a key necessary detection, and the detection of the structural abnormality of the bridge belongs to random inspection. The application focus of the inspection system in this embodiment is mainly based on inspection for the abnormal condition of the bridge, but inspection for the normal road surface abnormality can be configured separately in the scheme in this embodiment. The main part to the inspection is mainly the personnel of patrolling and examining, and the personnel of patrolling and examining mainly based on subjective mode in current inspection in-process carries out, and has following problem to this kind of mode of patrolling and examining: first, the management of the inspection personnel is mainly based on subjectivity of the inspection personnel, the inspection omission is possibly caused by the inexperience of the inspection personnel for the key inspection points, and the inspection omission is more easily caused under the inexperienced operation condition of the inspection personnel on the new duty. Moreover, specific judgment conclusion cannot be given to the inspection result aiming at subjectivity of inspection personnel, and particularly the problem that the accuracy of the inspection result is low due to insufficient judgment of a key main body structure of a bridge is easy to cause.
Therefore, in view of the above, there is a need to provide a targeted inspection system, which can improve the efficiency of the inspection process and the inspection accuracy.
The solution to the problems is mainly based on configuration of a routing inspection system, by configuring marks of key target points in the routing inspection system, accurate positioning of the key mark points is achieved through the marks, routing inspection coordinates are generated based on extraction of routing inspection maps corresponding to the key mark points, and routing inspection personnel can conduct routing inspection operation on the key mark points through the routing inspection coordinates. And the corresponding inspection detection module is configured in the system to realize the abnormal identification of the inspection target, so that the accuracy of the inspection target is improved. In addition, the abnormal identification of the inspection target is processed through a remote data processing module which is separated from the inspection terminal, so that the problem of high operation cost of the inspection terminal caused by processing is solved. In addition, in the embodiment, through configuration of the acquisition strategy for the identification result and the data correspondence in the inspection process, the problems of storage redundancy and high data interaction cost caused by large data volume are solved.
Referring to fig. 1, based on the above technical background, an embodiment of the present application provides an electronic inspection system 100 for a bridge, including a handheld terminal 110, a data processing terminal 120, a data storage terminal 130, and a plurality of data marking points 140; the handheld terminal comprises a display module 111, a communication establishing module 112 and a data acquisition module 113, wherein the communication establishing module receives a marking signal of the data marking point in a preset range and establishes communication, and the data acquisition module acquires real-time data in a preset range corresponding to the data marking point based on the establishment of the communication; the data processing terminal processes the real-time data acquired by the data acquisition module to obtain a processing result, and the data storage terminal is used for storing the processing result and the corresponding data processing terminal tag and the corresponding data marking point tag; the data processing terminal is internally provided with a data detection model, and the data detection model is used for detecting the real-time data to obtain a processing result; the data storage terminal comprises a plurality of storage spaces corresponding to the data marking points, wherein the storage spaces are used for storing data corresponding to the data marking points, each storage space comprises at least two storage subspaces, and each storage subspace is used for correspondingly storing the processing result types.
In this embodiment, the bridge electronic inspection system further includes a terminal platform 150, where the terminal platform is connected to the data storage terminal, and collects storage data in the data storage terminal based on an acquisition policy.
Specifically, in this embodiment, the acquisition policy includes a first acquisition policy and a second acquisition policy, where the first acquisition policy and the second acquisition policy are respectively configured in the two storage subspaces. The storage subspace comprises a first storage subspace 131 and a second storage subspace 132, and a first acquisition strategy and a second acquisition strategy are respectively configured; the first acquisition strategy is synchronous acquisition, and the second acquisition strategy is asynchronous acquisition. In this embodiment, different manners of collecting different data are implemented by dividing the storage subspace into a first storage subspace and a second storage subspace and configuring corresponding collection strategies, specifically, abnormal data is stored in the first storage subspace, and normal data is stored in the second storage subspace. Because the importance of the abnormal data for the subsequent analysis of the data is higher than that of the normal data, the abnormal data needs to be acquired in real time, so that decision errors caused by data transmission delay are reduced, the normal data can be acquired to the terminal platform based on preset acquisition frequency, and in the embodiment, the time period for the abnormal generation of the bridge is longer because the application scene is the inspection of the bridge, so that the acquisition frequency of the normal data can be set longer, for example, the acquisition can be performed in a month or quarter mode. The above acquisition strategy is an automatic acquisition strategy, and a third acquisition strategy can be set as an active acquisition strategy, so that the terminal platform user can actively perform self-defined real-time acquisition on the data of the second storage subspace at any time.
In this embodiment, the handheld terminal further includes a map marking module 114, where a plurality of layer maps are configured in the map marking module 114, and codes corresponding to data marking points are set in any one of the layer maps; the data marking points are configured with NFC digital tags, the NFC digital tags and the communication establishing module establish communication within a preset range, the map layer map is determined based on unique codes of the NFC digital tags, and the map layer map and corresponding marking point coordinates are displayed on the display module. The communication port or the communication protocol of the communication establishing module in the handheld terminal matched with the preset can be received based on the preset signal receiving range for the NFC digital tag to communicate, so that a communication link is established, a corresponding command is sent through the port and displayed on a corresponding display module, wherein the display mode is based on the mode of marking the marking point in the map layer corresponding to the NFC digital tag, and the display mode is used for acquiring the patrol inspector for the patrol inspection area and acquiring the key patrol inspection point in the patrol inspection area. The above arrangement is mainly based on missed detection of key inspection points, and inspection for daily inspection including road surface casting or contamination is based on random event generation, which is not described in this embodiment.
Referring to fig. 2, the present embodiment provides a method for inspecting a bridge electronic inspection system, which is applied to any one of the above-mentioned bridge electronic inspection systems, and the method includes:
and S210, the communication establishing module collects communication signals of the data marking points in a preset range and establishes communication with the data marking points.
In this embodiment, the data marking point is configured with an NFC digital tag, and the NFC digital tag and the communication establishing module establish communication within a preset range. The NFC digital tag has a communication function, and a communication protocol or a communication port corresponding to the NFC digital tag is configured in the communication establishment module. The NFC digital tag has a certain communication range, and sends a real-time communication connection request through the NFC digital tag, wherein a communication protocol matched with the handheld terminal is arranged in the communication connection request, and the communication connection is realized by matching based on the same communication protocol. In this embodiment, the establishment of the communication connection includes active connection and passive connection, where the active connection is an NFC digital tag for performing a communication connection request, and performs active connection based on an inspector; and the passive connection is that the NFC digital tag carries out communication connection request, and the communication establishment module directly carries out connection based on the communication connection request aiming at the communication connection request. The determination of the active connection and the passive connection modes can be set based on the patrol personnel, and the method defaults to active connection.
And S220, the map marking module establishes the corresponding unique code of the data marking point based on communication to determine the corresponding map layer map, and the map layer map and the coordinates corresponding to the data marking point are carried out on the display module.
In this embodiment, a unique code is set in the NFC digital tag, and an acquisition of a layer map corresponding to a corresponding inspection target may be acquired through an association relationship of the unique code, where a coordinate point of the inspection target point is configured in the layer map and the coordinate point is displayed in the layer map, where the inspection point for establishing communication connection is displayed in a visual jumping manner, so that an inspection person for displaying the inspection target point for establishing communication connection to the inspection person may determine an inspection target point closest to the inspection person, and the determination of the inspection target point by the inspection person is reduced by displaying a plurality of inspection target points. The association relation in the embodiment is realized on the basis of a preset database, wherein the association relation is constructed among a plurality of inspection target points through a hierarchical relation in the database. The method comprises the following steps: and establishing a list, and setting the list in a multi-level mode, wherein the highest level is a patrol target, namely, a plurality of patrol targets are stored in a database, and the patrol targets are target bridges. The next level is a target inspection point, and the last level is a unique code and coordinate point of the NFC digital tag. The method comprises the steps of determining a target inspection point through unique codes in NFC digital tags, determining an inspection target based on the target inspection point, and determining all the target inspection points, a plurality of NFC digital tags corresponding to the target inspection points and a plurality of corresponding coordinate points based on the inspection target.
In this embodiment, the determination of the inspection target point can be realized through the process, so that the possibility of missed inspection of an inspection personnel is reduced, and in other embodiments, an optimal path planning method can be configured based on coordinate points, and an optimal inspection path can be obtained through the relation among a plurality of coordinate points and the coordinates of the current inspection target point, wherein the path planning method can be realized by adopting a method in the prior art, and is not described in this embodiment.
And S230, the data acquisition module acquires real-time data in a preset range corresponding to the data mark point based on the establishment of communication.
In this embodiment, because the main application scenario is to identify the abnormality in the bridge, and the identification manner is mainly based on identifying the abnormality in the bridge key structure, the main embodiment of the abnormality for the bridge key structure is that the surface of the bridge generates a crack or a crack. The abnormal recognition for the bridge is mainly based on whether the surface of the bridge has a crack or not, and the acquisition for the crack is mainly recognized by the image of the bridge, so the real-time data is real-time image data in the embodiment.
The key points of the bridge can be arranged on the road surface of the bridge and the beam column of the bridge, the areas of the road surface of the bridge and the beam column of the bridge are large, and in order to enable the image to be clearer and more complete, the acquisition mode of the image data needs to be determined, so that the image can display the information of the crack as much as possible, other noise images are prevented from being mixed in the image, wherein the noise images are other environment background images except for related images of the bridge. Because bridge pavement has a large area and is less likely to be mixed into other environmental background images, the processing for this step is mainly used for image acquisition at the edges of the beam columns or pavement. In order to achieve the object, the processing logic in the embodiment sets a target acquisition frame, wherein the target acquisition frame has preset limiting conditions such as length, width, height and the like, and acquires the main image and the background image in the target image through the target acquisition frame, so that the target image is completely positioned in the target acquisition frame, and the problem of higher subsequent processing caused by introducing the background environment image is solved.
In this embodiment, this step specifically includes the following steps:
and the data acquisition module determines a target acquisition frame and a target acquisition standard image based on the unique code of the data mark point. In this embodiment, the areas of the target acquisition frames adopted based on the difference in the areas of the targets are also different, wherein the determination of the target acquisition frames is achieved through the correspondence.
And acquiring the real-time image data by carrying out image acquisition on the object to be patrolled and examined corresponding to the data mark point based on the target acquisition frame and the target acquisition standard image. The method specifically comprises the following steps:
and identifying the real-time image data based on the target acquisition frame to obtain an identification result.
Based on the recognition result, judging whether the real-time image data meets a preset condition, wherein the method specifically comprises the following steps:
and performing binarization processing on the real-time image data to obtain preprocessed image data.
And dividing the preprocessed image data to obtain a target processed image.
And extracting edge coordinates of the target processing image, comparing the edge coordinates with preset standard edge coordinates, and determining whether the real-time image data meets preset conditions or not based on a comparison threshold.
In this embodiment, when a preset condition is satisfied, the real-time image data is acquired to obtain a real-time image, where the real-time image is an acquired target image. The extraction method for image segmentation and coordinate points may be obtained based on an existing method in existing machine vision, and will not be described in the present embodiment.
And S340, sending the real-time image data to the data processing terminal, and detecting abnormality of the real-time image data based on a data detection model in the data processing terminal to obtain a detection result.
In the present embodiment, the data detection model includes a feature extraction network, a region recommendation network, and a classification network. The feature extraction network comprises five groups of convolutional neural networks, wherein any convolutional neural network comprises at least one convolutional layer, an activation function layer and a pooling layer; the activation function in the activation function layer is a rule function. The regional recommendation network comprises a first convolution layer, a second convolution layer and a third convolution layer, wherein the second convolution layer and the third convolution layer are connected with the output end of the first convolution layer, the output end of the second convolution layer is connected with a normalized index function layer, and the output end of the normalized index function layer and the output end of the third convolution layer are connected with a combination layer. The classification network comprises a mapping layer, a first convolution neural network, a second convolution neural network and a CRF layer, wherein the mapping layer is used for taking an image block to be detected and the field thereof output by the regional recommendation network as input, mapping is carried out through the mapping layer to obtain an image matrix, the image matrix is input into the first convolution neural network layer to obtain feature vectors, the feature vectors are respectively input into the second convolution neural network layer and the CRF layer, position regression processing is carried out through the second convolution neural network layer, and whether the probability of a target feature is judged based on the CRF layer, namely whether the probability of a crack is judged.
The method comprises the following specific processes:
the feature extraction is realized through five times of convolution processing in the feature extraction network, wherein in the convolution process, the size of each convolution kernel is 3 multiplied by 3, and the combination of a plurality of convolution kernels with small sizes through the arrangement can reduce the calculated amount and achieve the same receptive field as large convolutions of 5 multiplied by 5 and 7 multiplied by 7. In this embodiment, the step size of the convolution operation is 1, and the size of the boundary filling is also 1, so that the width and the height of the image will not change after each convolution. Each pooling layer adopts the maximum pooling, the size of the pooling layer is 2 multiplied by 2, the step length is also 2, and thus, the width and the height of the image are reduced to half of the original width and the height of the image after each pooling. Thus, after passing through the feature extraction network, the RGB color image of original size (h, w, 3) will become a feature map of size (h/16, w/16, 5112).
The feature map is first divided into two paths after being convolved with step size and boundary filling of 1 by 512 3×3 in the region recommendation network. The first path is operated on by 18 1 x 1 convolutions and softmax. The purpose of the 1×1 convolution is to change the dimension of the feature map, i.e. the number of channels of the feature map becomes 18, corresponding to the probability (Pfg, pbg) that the preset 9 anchor points at each point of the feature map belong to the foreground and the background, respectively. The second path is operated by 36 1×1 convolutions, the number of channels of the feature map becomes 36, the 4 position coordinate correction values (tx, ty, tw, th) corresponding to the 9 anchor points, and the coordinates and width and height of the original preset anchor point are set to be (px, py, pw, ph), and the positions of the anchor points after correction are output through the second path as follows:
Wherein g x ，g y ,g w ,g h The center coordinates of the anchor points after correction are wide and high. The combination layer is a set of operations, and is used for combining the results obtained by the two convolutions. The layer firstly cuts off the size of the anchor points exceeding the size of the original image to ensure that all the anchor points are in the original image, then eliminates some anchor points with undersize and oversized aspect ratio, sorts the anchor points from high to low by utilizing the output of the first path, reserves the first K anchor points, and finally screens through a non-maximum suppression algorithm to reserve M candidate areas.
The classification and network takes each generated candidate region as input, and needs to perform classification judgment on whether the candidate region is a crack or not and regression correction on position coordinates of the candidate region. The method comprises the steps of dividing a classification network into two paths, wherein the first path is a classification part formed by a convolutional neural network and a CRF, in the classification process, firstly taking each candidate region and surrounding image blocks with the same size as the region, wherein the number of the image blocks is 9, taking 9 image blocks as input, firstly carrying out feature extraction through the convolutional neural network to obtain 9 feature vectors X, then taking the 9 feature vectors as input to send the 9 feature vectors into the CRF, and calculating the conditional probability of each image block label Y, wherein the label Y is a crack, thereby judging the category of the candidate region. The second path is similar to the region recommendation network, the position coordinate correction value of each candidate region is directly obtained through the convolutional neural network, the probability of containing the Y tag is obtained through a variation inference mode, and whether a crack exists or not is determined based on the probability. The method comprises the following steps:
Based on the above-described set conditional probability distribution model P (y|x), this probability distribution model is used to represent the conditional probability distribution of the other set of output random variables Y given one set of input random variables X. In this embodiment, this probability distribution is represented by an undirected graph, the nodes in the undirected graph represent random variables, and the edges in the graph represent the dependency relationships between the random variables. Acquiring a joint probability distribution about the undirected graph based on the undirected graph, wherein the joint probability distribution can be factored on the largest clique thereof, and the joint probability distribution can be represented as follows:
wherein, the liquid crystal display device comprises a liquid crystal display device,as a normalization factor, the effect is to ensure that the resulting P (y|x) is a reasonable probability distribution. X is defined by a series of random variables (X 1 ,X 2 ,L,X N ) Composition, representing a given image block, in this embodiment, where the number of image blocks is 9, n=9 and x represents a 3×3 image block. Y is formed from a series of random variables (Y 1 ,Y 2 ,L,Y N )，Y i The value range of the label of the ith image block is {0,1},0 indicates no crack, and 1 indicates crack. E (Y, X) is a potential function representing the cost of Y taking a certain set of specific values given X, in this embodiment a binary potential function, in this embodiment a cosine similarity is used to define the binary potential function in the form:
Wherein u (Y) i ,Y j ) As a tag compatible function, when yi=yj, u (Y i ,Y j ) =1, otherwise 0.
And obtaining a result of whether the crack exists or not based on the joint probability distribution.
S350, sending the detection result to the data storage terminal based on a storage strategy; and the terminal platform collects the data in the data storage terminal based on the collection strategy.
According to the technical scheme provided by the embodiment of the application, the accurate acquisition of the inspection point in the inspection process and the identification of whether the inspection point contains an abnormality or not are realized by setting a plurality of functional modules in the system. The inspection efficiency is improved, and the recognition accuracy of the abnormality in the inspection is improved.
Referring to fig. 3, the above method may be integrated into an electronic device 300 comprising a memory 310, a processor 320 and a computer program stored in the memory and executable on the processor, wherein the processor performs the data sharing method.
In this embodiment, the memory, the processor and each element of the communication unit are electrically connected directly or indirectly to each other, so as to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory is used for storing specific information and programs, and the communication unit is used for sending the processed information to the corresponding user side.
In this embodiment, the storage module is divided into two storage areas, where one storage area is a program storage unit and the other storage area is a data storage unit. The program storage unit is equivalent to a firmware area, the read-write authority of the area is set to be in a read-only mode, and the data stored in the area can not be erased and changed. And the data in the data storage unit can be erased or read and written, and when the capacity of the data storage area is full, the newly written data can cover the earliest historical data.
The Memory may be, but is not limited to, random access Memory (Random Access Memory, RAM), read Only Memory (ROM), programmable Read Only Memory (Programmable Read-Only Memory, PROM), erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable Read Only Memory (Ele ultrasound ric Erasable Programmable Read-Only Memory, EEPROM), etc.
The processor may be an integrated circuit chip having signal processing capabilities. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also Digital Signal Processors (DSPs)), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It is to be understood that the terminology which is not explained by terms of nouns in the foregoing description is not intended to be limiting, as those skilled in the art can make any arbitrary deduction from the foregoing disclosure.
The person skilled in the art can undoubtedly determine technical features/terms of some preset, reference, predetermined, set and preference labels, such as threshold values, threshold value intervals, threshold value ranges, etc., from the above disclosure. For some technical feature terms which are not explained, a person skilled in the art can reasonably and unambiguously derive based on the logical relation of the context, so that the technical scheme can be clearly and completely implemented. The prefixes of technical feature terms, such as "first", "second", "example", "target", etc., which are not explained, can be unambiguously deduced and determined from the context. Suffixes of technical feature terms, such as "set", "list", etc., which are not explained, can also be deduced and determined unambiguously from the context.
The foregoing disclosure of embodiments of the present application will be apparent to and complete in light of the foregoing disclosure to those skilled in the art. It should be appreciated that the development and analysis of technical terms not explained based on the above disclosure by those skilled in the art is based on the description of the present application, and thus the above is not an inventive judgment of the overall scheme.
While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements and adaptations of the application may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within the present disclosure, and therefore, such modifications, improvements, and adaptations are intended to be within the spirit and scope of the exemplary embodiments of the present disclosure.
Meanwhile, the present application uses specific terms to describe embodiments of the present application. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the application. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various portions of this specification are not necessarily all referring to the same embodiment. Furthermore, certain features, structures, or characteristics of at least one embodiment of the present application may be combined as suitable.
In addition, those skilled in the art will appreciate that the various aspects of the application are illustrated and described in the context of a number of patentable categories or conditions, including any novel and useful processes, machines, products, or materials, or any novel and useful improvements thereof. Accordingly, aspects of the application may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "unit," component, "or" system. Furthermore, aspects of the application may be embodied as a computer product in at least one computer-readable medium, the product comprising computer-readable program code.
The computer readable signal medium may comprise a propagated data signal with computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take on a variety of forms, including electro-magnetic, optical, etc., or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code located on a computer readable signal medium may be propagated through any suitable medium including radio, electrical, fiber optic, RF, or the like, or any combination of the foregoing.
Computer program code required for carrying out aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming such as Java, scala, smalltalk, eiffel, JADE, emerald, C ++, c#, vb net, python and the like, or similar conventional programming languages such as the "C" programming language, visual Basic, fortran 2003,Perl,COBOL 2002,PHP,ABAP, dynamic programming languages such as Python, ruby and Groovy or other programming languages. The programming code may execute entirely on the user's computer, or as a stand-alone software package, or partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any form of network, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as software as a service (SaaS).
Furthermore, the order in which the processing elements and sequences are described, the use of numerical letters, or other designations are used is not intended to limit the order in which the processes and methods of the application are performed unless specifically recited in the claims. While in the foregoing disclosure there has been discussed, by way of various examples, some embodiments of the application which are presently considered to be useful, it is to be understood that this detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments of the application. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.
It should also be appreciated that in the foregoing description of at least one embodiment of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of at least one embodiment of the application. This method of disclosure, however, is not intended to imply that more features than are required by the subject application. Indeed, less than all of the features of a single embodiment disclosed above.
Claims (5)
1. The electronic inspection method for the bridge is characterized by comprising a bridge electronic inspection system, wherein the bridge electronic inspection system comprises a handheld terminal, a data processing terminal, a data storage terminal and a plurality of data marking points; the handheld terminal comprises a display module, a communication establishing module and a data acquisition module, wherein the communication establishing module receives a marking signal of the data marking point in a preset range and establishes communication, and the data acquisition module acquires real-time data in a preset range corresponding to the data marking point based on the establishment of the communication; the data processing terminal processes the real-time data acquired by the data acquisition module to obtain a processing result, and the data storage terminal is used for storing the processing result and the corresponding data processing terminal tag and the corresponding data marking point tag; the data processing terminal is internally provided with a data detection model, and the data detection model is used for detecting the real-time data to obtain a processing result; the data storage terminal comprises a plurality of storage spaces corresponding to the data marking points, wherein the storage spaces are used for storing data corresponding to the data marking points, each storage space comprises at least two storage subspaces, and each storage subspace is used for correspondingly storing the processing result types;
The handheld terminal further comprises a map marking module, wherein a plurality of map layer maps are configured in the map marking module, and codes corresponding to data marking points are arranged in any map layer map; the data marking points are configured with NFC digital tags, the NFC digital tags and the communication establishing module establish communication within a preset range, the map layer map is determined based on unique codes of the NFC digital tags, and the map layer map and corresponding marking point coordinates are displayed on the display module;
the NFC digital tag is provided with a unique code, the map layer map corresponding to the corresponding inspection target can be obtained through the association relation of the unique code, wherein the map layer map is provided with coordinate points of the inspection target points and the coordinate points are displayed in the map layer map, the inspection points for establishing communication connection are displayed in a visual jumping mode, the inspection target points for establishing communication connection are displayed to inspection personnel, the inspection personnel can determine the inspection target point closest to the inspection target point, and the inspection personnel can determine the inspection target points through the display of a plurality of inspection target points, so that the omission of inspection is reduced; the association relation is realized on the basis of a preset database, wherein the association relation is constructed among a plurality of inspection target points in the database through a hierarchical relation, and the association relation is specifically as follows: establishing a list, and setting the list in a multi-level mode, wherein the highest level is a patrol target, namely, a plurality of patrol targets are stored in a database, and the patrol targets are target bridges; the next level is a target inspection point, and the last level is a unique code and a coordinate point of the NFC digital label; determining a target inspection point through unique codes in the NFC digital tags, determining an inspection target based on the target inspection point, and determining all the target inspection points, a plurality of NFC digital tags corresponding to the target inspection points and a plurality of corresponding coordinate points based on the inspection target;
The data detection model comprises a feature extraction network, a region recommendation network and a classification network; the feature extraction network comprises five groups of convolutional neural networks, wherein any convolutional neural network comprises at least one convolutional layer, an activation function layer and a pooling layer; the activation function in the activation function layer is a rule function; the regional recommendation network comprises a first convolution layer, a second convolution layer and a third convolution layer, wherein the second convolution layer and the third convolution layer are connected with the output end of the first convolution layer, the output end of the second convolution layer is connected with a normalized index function layer, and the output ends of the normalized index function layer and the third convolution layer are connected with a combination layer; the method comprises the steps that a mapping layer, a first convolutional neural network, a second convolutional neural network and a CRF layer are used for mapping an image block to be detected and the field of the image block to be detected, which are output by a regional recommendation network, serving as input, the mapping layer to obtain an image matrix, the image matrix serving as input to the first convolutional neural network layer to obtain feature vectors, the feature vectors are respectively input to the second convolutional neural network layer and the CRF layer, position regression processing is conducted through the second convolutional neural network layer, and whether the probability of a target feature is achieved or not, namely whether the probability of a crack is achieved or not is judged based on the CRF layer;
The bridge electronic inspection system further comprises a terminal platform, wherein the terminal platform is connected with the data storage terminal, and stored data in the data storage terminal are collected based on an acquisition strategy;
the method comprises the following steps:
the communication establishing module collects communication signals of the data marking points in a preset range and establishes communication with the data marking points;
the map marking module establishes a corresponding unique code of the data marking point based on communication to determine a corresponding map layer map, and displays the map layer map and coordinates of the corresponding data marking point on the display module;
the data acquisition module acquires real-time data in a preset range corresponding to the data mark point based on the establishment of communication, wherein the real-time data comprises real-time image data meeting the preset range;
transmitting the real-time image data to the data processing terminal, and performing anomaly detection on the real-time image data based on a data detection model in the data processing terminal to obtain a detection result;
transmitting the detection result to the data storage terminal based on a storage strategy;
the terminal platform collects data in the data storage terminal based on a collection strategy;
The method comprises the following specific processes:
the feature extraction is realized through five times of convolution processing in a feature extraction network, wherein in the convolution process, the size of each convolution kernel is 3 multiplied by 3, and the combination of a plurality of convolution kernels with small sizes can reduce the calculated amount and achieve the same receptive field as large convolutions of 5 multiplied by 5 and 7 multiplied by 7; the step length of the convolution operation is 1, and the size of boundary filling is 1, so that the width and the height of the image are not changed after each convolution; each pooling layer adopts the maximum pooling, the size of the pooling layer is 2 multiplied by 2, and the step length is also 2, so that the width and the height of the image are reduced to half of the original width and the height of the image after each pooling; thus, after passing through the feature extraction network, the RGB color image of original size (h, w, 3) will become a feature map of size (h/16, w/16, 5112);
the feature map is divided into two paths after being convolved with step length and boundary filling of 1 through 512 3×3 in the regional recommendation network; the first path is operated on by 18 1 x 1 convolutions and softmax; the purpose of the 1×1 convolution is to change the dimension of the feature map, i.e. the number of channels of the feature map becomes 18, corresponding to the probability (Pfg, pbg) that the 9 anchor points preset on each point of the feature map belong to the foreground and the background respectively; the second path is operated by 36 1×1 convolutions, the number of channels of the feature map becomes 36, the 4 position coordinate correction values (tx, ty, tw, th) corresponding to the 9 anchor points, and the coordinates and width and height of the original preset anchor point are set to be (px, py, pw, ph), and the positions of the anchor points after correction are output through the second path as follows: Wherein gx, gy, gw, gh are the center coordinates and width and height of the corrected anchor points; the combination layer is a set of a series of operations and has the function of combining the results obtained by the two convolutions; the method comprises the steps of firstly cutting off anchors exceeding the original image size to ensure that all anchors are in the original image, rejecting anchors with undersize and oversized aspect ratio, sequencing the anchors from high to low by using the output of a first path, reserving the first K anchors, and finally screening by a non-maximum suppression algorithm to reserve M candidate areas; classifying and taking each generated candidate area as input by the network, and judging whether the candidate area is a crack or not and carrying out regression correction on the position coordinates of the candidate area; wherein the classification network is equally classified intoThe method comprises the steps of two paths, wherein the first path is a classification part formed by a convolutional neural network and a CRF, in the classification process, firstly taking each candidate region and image blocks with the same size as the region around the candidate region, wherein the number of the image blocks is 9, taking 9 image blocks as input, firstly carrying out feature extraction through the convolutional neural network to obtain 9 feature vectors X, and then sending the 9 feature vectors as input into the CRF, and calculating the conditional probability of each image block label Y, wherein the label Y is a crack, so that the category of the candidate region is judged; the second path is similar to the region recommendation network, the position coordinate correction value of each candidate region is directly obtained through a convolutional neural network, the probability of containing the Y tag is obtained through a variation inference mode, and whether a crack exists or not is determined based on the probability; the method comprises the following steps: based on the above-described set conditional probability distribution model P (y|x) for representing the conditional probability distribution of the other set of output random variables Y given one set of input random variables X; representing the probability distribution by adopting an undirected graph, wherein nodes in the undirected graph represent random variables, and edges in the graph represent the dependency relationship among the random variables; acquiring a joint probability distribution about the undirected graph based on the undirected graph, wherein the joint probability distribution can be factored on the largest clique thereof, and the joint probability distribution can be represented as follows:
Wherein (1)>Is a normalization factor, whose function is to ensure that the resulting P (Y|X) is a reasonable probability distribution; x consists of a series of random variables (X1, X2, L, XN) representing a given image block, the number of image blocks being 9, then n=9, X representing a 3X 3 image block; y is a label of an ith image block by a series of random variables (Y1, Y2, L and YN), the value range of the label is {0,1},0 represents no crack, and 1 represents crack; e (Y, X) is a potential function representing the cost of Y taking a certain set of specific values given X, the potential function is a binary potential function, and cosine is adoptedSimilarity defines a binary potential function in the form:
where u (Yi, yj) is a tag compatible function, and when yi=yj, u (Yi, yj) =1, otherwise, 0; and obtaining a result of whether the crack exists or not based on the joint probability distribution.
2. The bridge electronic inspection method of claim 1, wherein the acquisition strategy comprises a first acquisition strategy and a second acquisition strategy, the first acquisition strategy and the second acquisition strategy being respectively configured within two of the storage subspaces.
3. The bridge electronic inspection method of claim 2, wherein the storage subspace comprises a first storage subspace and a second storage subspace, and a first acquisition strategy and a second acquisition strategy are respectively configured; the first acquisition strategy is synchronous acquisition, and the second acquisition strategy is asynchronous acquisition.
4. The electronic inspection method of claim 3, wherein the data acquisition module acquires real-time data within a preset range of corresponding data mark points based on establishment of communication, and comprises:
the data acquisition module determines a target acquisition frame and a target acquisition standard image based on the unique code of the data mark point;
based on the target acquisition frame and the target acquisition standard image, image acquisition is carried out on the object to be patrolled and examined corresponding to the data mark point, and the real-time image data is obtained, and the method specifically comprises the following steps:
identifying the real-time image data based on the target acquisition frame to obtain an identification result;
based on the recognition result, judging whether the real-time image data meets a preset condition or not, specifically comprising:
binarizing the real-time image data to obtain preprocessed image data;
dividing the preprocessed image data to obtain a target processed image;
and extracting edge coordinates of the target processing image, comparing the edge coordinates with preset standard edge coordinates, and determining whether the real-time image data meets preset conditions or not based on a comparison threshold.
5. The bridge electronic inspection method according to claim 4, wherein the real-time image data is subjected to anomaly detection based on a data detection model in the data processing terminal to obtain a detection result, the data detection model being composed of a feature extraction network, a region recommendation network and a classification network, the method comprising:
extracting features in the real-time image data based on a feature extraction network to obtain a target feature map;
positioning a plurality of candidate areas containing detection targets in the target feature map based on an area recommendation network;
and acquiring the spatial characteristics of the candidate areas, modeling, judging whether the candidate areas contain probability distribution of target features based on the model, and determining whether the candidate areas contain the target features based on the probability distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211177540.8A CN115512098B (en) | 2022-09-26 | 2022-09-26 | Bridge electronic inspection system and inspection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211177540.8A CN115512098B (en) | 2022-09-26 | 2022-09-26 | Bridge electronic inspection system and inspection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115512098A CN115512098A (en) | 2022-12-23 |
| CN115512098B true CN115512098B (en) | 2023-09-01 |
Family
ID=84506426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211177540.8A Active CN115512098B (en) | 2022-09-26 | 2022-09-26 | Bridge electronic inspection system and inspection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115512098B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503779A (en) * | 2023-04-26 | 2023-07-28 | 中国公路工程咨询集团有限公司 | Pavement casting object identification system and method |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105913183A (en) * | 2016-04-11 | 2016-08-31 | 南京学安网络科技有限公司 | Safety tour inspection management system |
| CN107832770A (en) * | 2017-11-08 | 2018-03-23 | 浙江国自机器人技术有限公司 | A kind of equipment routing inspection method, apparatus, system, storage medium and crusing robot |
| CN108564109A (en) * | 2018-03-21 | 2018-09-21 | 天津大学 | A kind of Remote Sensing Target detection method based on deep learning |
| CN109765462A (en) * | 2019-03-05 | 2019-05-17 | 国家电网有限公司 | Fault detection method, device and the terminal device of transmission line of electricity |
| JP2019117105A (en) * | 2017-12-27 | 2019-07-18 | ウエブテック株式会社 | Print quality inspection apparatus |
| CN112580858A (en) * | 2020-12-04 | 2021-03-30 | 北京京能高安屯燃气热电有限责任公司 | Equipment parameter prediction analysis method and system |
| CN113160022A (en) * | 2021-04-27 | 2021-07-23 | 广东天濠建设工程有限公司 | Municipal bridge maintenance management system, method and equipment and readable storage medium |
| WO2021195563A1 (en) * | 2020-03-26 | 2021-09-30 | Digimarc Corporation | Arrangements for digital marking and reading of items, useful in recycling |
| CN113963277A (en) * | 2021-10-26 | 2022-01-21 | 广西壮族自治区通信产业服务有限公司工程分公司 | Aerial photography-based railway power through line hidden danger target detection method |
| CN114267082A (en) * | 2021-09-16 | 2022-04-01 | 南京邮电大学 | Bridge side falling behavior identification method based on deep understanding |
| WO2022095616A1 (en) * | 2020-11-03 | 2022-05-12 | 国网智能科技股份有限公司 | On-line intelligent inspection system and method for transformer substation |
| WO2022121186A1 (en) * | 2020-12-11 | 2022-06-16 | 平安科技（深圳）有限公司 | Method and apparatus for routing inspection of oil and gas pipelines on the basis of target matching |
| CN114665608A (en) * | 2022-05-19 | 2022-06-24 | 国网智能科技股份有限公司 | Intelligent sensing inspection system and method for transformer substation |
-
2022
- 2022-09-26 CN CN202211177540.8A patent/CN115512098B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105913183A (en) * | 2016-04-11 | 2016-08-31 | 南京学安网络科技有限公司 | Safety tour inspection management system |
| CN107832770A (en) * | 2017-11-08 | 2018-03-23 | 浙江国自机器人技术有限公司 | A kind of equipment routing inspection method, apparatus, system, storage medium and crusing robot |
| JP2019117105A (en) * | 2017-12-27 | 2019-07-18 | ウエブテック株式会社 | Print quality inspection apparatus |
| CN108564109A (en) * | 2018-03-21 | 2018-09-21 | 天津大学 | A kind of Remote Sensing Target detection method based on deep learning |
| CN109765462A (en) * | 2019-03-05 | 2019-05-17 | 国家电网有限公司 | Fault detection method, device and the terminal device of transmission line of electricity |
| WO2021195563A1 (en) * | 2020-03-26 | 2021-09-30 | Digimarc Corporation | Arrangements for digital marking and reading of items, useful in recycling |
| WO2022095616A1 (en) * | 2020-11-03 | 2022-05-12 | 国网智能科技股份有限公司 | On-line intelligent inspection system and method for transformer substation |
| CN112580858A (en) * | 2020-12-04 | 2021-03-30 | 北京京能高安屯燃气热电有限责任公司 | Equipment parameter prediction analysis method and system |
| WO2022121186A1 (en) * | 2020-12-11 | 2022-06-16 | 平安科技（深圳）有限公司 | Method and apparatus for routing inspection of oil and gas pipelines on the basis of target matching |
| CN113160022A (en) * | 2021-04-27 | 2021-07-23 | 广东天濠建设工程有限公司 | Municipal bridge maintenance management system, method and equipment and readable storage medium |
| CN114267082A (en) * | 2021-09-16 | 2022-04-01 | 南京邮电大学 | Bridge side falling behavior identification method based on deep understanding |
| CN113963277A (en) * | 2021-10-26 | 2022-01-21 | 广西壮族自治区通信产业服务有限公司工程分公司 | Aerial photography-based railway power through line hidden danger target detection method |
| CN114665608A (en) * | 2022-05-19 | 2022-06-24 | 国网智能科技股份有限公司 | Intelligent sensing inspection system and method for transformer substation |
Non-Patent Citations (1)
| Title |
|---|
| L. Zhang等.Application of Internet of Things Technology and Convolutional Neural Network Model in Bridge Crack Detection.2018,第6卷第39442-39451页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115512098A (en) | 2022-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ji et al. | An integrated approach to automatic pixel-level crack detection and quantification of asphalt pavement | |
| Sajedi et al. | Uncertainty‐assisted deep vision structural health monitoring | |
| CN111862067B (en) | Welding defect detection method and device, electronic equipment and storage medium | |
| Hoang et al. | A novel method for asphalt pavement crack classification based on image processing and machine learning | |
| Eisenbach et al. | How to get pavement distress detection ready for deep learning? A systematic approach | |
| Chen et al. | Pavement crack detection and classification based on fusion feature of LBP and PCA with SVM | |
| CN115908424A (en) | Building health detection method, system and medium based on three-dimensional laser scanning | |
| CN110781885A (en) | Text detection method, device, medium and electronic equipment based on image processing | |
| CN105825169B (en) | A kind of pavement crack recognition methods based on road image | |
| CN109886122B (en) | Lane line detection method, lane line detection device, computer device, and storage medium | |
| CN115512098B (en) | Bridge electronic inspection system and inspection method | |
| CN111274926B (en) | Image data screening method, device, computer equipment and storage medium | |
| CN113240623B (en) | Pavement disease detection method and device | |
| CN112749653A (en) | Pedestrian detection method, device, electronic equipment and storage medium | |
| Chatterjee et al. | Smart infrastructure monitoring: Development of a decision support system for vision-based road crack detection | |
| Ha et al. | Assessing severity of road cracks using deep learning-based segmentation and detection | |
| Ibrahim et al. | Characterization of cracking in pavement distress using image processing techniques and k-Nearest neighbour | |
| Guerin et al. | Evaluation of runtime monitoring for UAV emergency landing | |
| König et al. | What's Cracking? A Review and Analysis of Deep Learning Methods for Structural Crack Segmentation, Detection and Quantification | |
| CN115171045A (en) | YOLO-based power grid operation field violation identification method and terminal | |
| US20230314169A1 (en) | Method and apparatus for generating map data, and non-transitory computer-readable storage medium | |
| CN110738208A (en) | efficient scale-normalized target detection training method | |
| US20220245782A1 (en) | Method and apparatus for classifying image of displaying base plate | |
| CN115409839A (en) | Road sound barrier hidden danger identification method and device based on pixel analysis model | |
| CN115376106A (en) | Vehicle type identification method, device, equipment and medium based on radar map |
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 | ||
| CB02 | Change of applicant information |
Address after: 550001 26th floor, No.1 Yan'an Middle Road, Yunyan District, Guiyang City, Guizhou Province Applicant after: GUIZHOU BRIDGE CONSTRUCTION GROUP Co.,Ltd. Applicant after: Chongqing University Address before: 400044 No. 174, positive street, Shapingba District, Chongqing Applicant before: Chongqing University Applicant before: GUIZHOU BRIDGE CONSTRUCTION GROUP Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 400044 No. 174, positive street, Shapingba District, Chongqing Applicant after: Chongqing University Applicant after: GUIZHOU BRIDGE CONSTRUCTION GROUP Co.,Ltd. Address before: 550001 26th floor, No.1 Yan'an Middle Road, Yunyan District, Guiyang City, Guizhou Province Applicant before: GUIZHOU BRIDGE CONSTRUCTION GROUP Co.,Ltd. Applicant before: Chongqing University |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |