CN117314917B - Bridge crack detection method and device, computer equipment and storage medium - Google Patents
Bridge crack detection method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN117314917B CN117314917B CN202311608863.2A CN202311608863A CN117314917B CN 117314917 B CN117314917 B CN 117314917B CN 202311608863 A CN202311608863 A CN 202311608863A CN 117314917 B CN117314917 B CN 117314917B
- Authority
- CN
- China
- Prior art keywords
- crack
- binary image
- skeleton
- bridge
- module
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 70
- 238000004364 calculation method Methods 0.000 claims abstract description 77
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 39
- 238000000605 extraction Methods 0.000 claims abstract description 28
- 230000011218 segmentation Effects 0.000 claims abstract description 23
- 239000013598 vector Substances 0.000 claims description 22
- 238000005260 corrosion Methods 0.000 claims description 21
- 230000007797 corrosion Effects 0.000 claims description 21
- 238000010586 diagram Methods 0.000 claims description 17
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 11
- 238000012545 processing Methods 0.000 abstract description 9
- 230000008569 process Effects 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000011426 transformation method Methods 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000007246 mechanism Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005674 electromagnetic induction Effects 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000036039 immunity Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30181—Earth observation
- G06T2207/30184—Infrastructure
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The application belongs to the technical field of image processing, and relates to a bridge crack detection method, a device, computer equipment and a storage medium, wherein the method comprises the following steps: receiving an original bridge picture sent by a user terminal; calling a trained UNet++ model, and inputting an original bridge picture into the trained UNet++ model to perform crack segmentation operation to obtain a crack binary image, wherein the UNet++ model is combined with a channel attention module and a space attention module; denoising the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoising binary image; performing skeleton extraction operation on the denoising binary image according to a skeletonization algorithm to obtain fracture skeleton data; performing crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method to obtain a crack calculation result; and outputting the crack calculation result to the user terminal. The method and the device can apply the crack detection technology based on the image to concrete crack detection, and realize automatic crack detection.
Description
Technical Field
The application relates to the technical field of image processing of artificial intelligence, in particular to a bridge crack detection method, a bridge crack detection device, computer equipment and a storage medium.
Background
Along with the acceleration of the urban process, bridge construction becomes an important component of modern urban construction. However, as the service life of bridges increases, the problem of cracking of bridges also becomes more and more pronounced. The appearance of bridge crack can not only influence the life of bridge, still can cause the threat to driving safety. Therefore, research and application of bridge crack detection technology become particularly important. At present, research on bridge crack detection technology at home and abroad has been advanced to a certain extent. The bridge cracks are mainly detected by adopting technologies such as visual detection, laser scanning, ultrasonic detection and the like in China. In contrast, foreign bridge crack detection techniques have been studied more intensively and extensively. In addition to the traditional technologies of visual inspection, laser scanning, ultrasonic inspection and the like, a series of novel inspection technologies such as infrared imaging technology, electromagnetic induction technology, microwave imaging technology and the like are also developed abroad.
In recent years, image-based crack detection has received increasing attention in the field of nondestructive detection. The main advantages of image segmentation detection are that the automatic identification and positioning of cracks obtained on the surface of an object are realized by utilizing computer vision and image processing technology, and compared with the traditional manual detection method, the detection precision and efficiency can be improved, and a large amount of manpower and material resources are saved. Among them, after deep convolutional neural networks (Deep Convolutional Neural Networks, DCNN) have been proposed, in solving many computer vision problems, such as image recognition, object detection, semantic image segmentation, etc., a number of practices have been widely applied, showing that image-based crack detection is superior to conventional crack detection methods. Convolutional neural networks are artificial intelligence and machine learning based techniques that can automate classification, recognition, segmentation, etc. of images. In the field of concrete crack detection, convolutional neural networks have also found widespread use.
However, applicants have found that conventional image-based crack detection methods are complex and variable, resulting in many network models not yet being applied to concrete crack detection.
Disclosure of Invention
The embodiment of the application aims to provide a bridge crack detection method, a bridge crack detection device, computer equipment and a storage medium, so as to solve the problem that an image-based crack detection technology is applied to concrete crack detection.
In order to solve the above technical problems, the embodiments of the present application provide a bridge crack detection method, which adopts the following technical scheme:
receiving an original bridge picture sent by a user terminal;
calling a trained UNet++ model, and inputting the original bridge picture to the trained UNet++ model to perform crack segmentation operation to obtain a crack binary image, wherein the UNet++ model combines a channel attention module and a space attention module;
denoising the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoising binary image;
performing skeleton extraction operation on the denoising binary image according to a skeletonization algorithm to obtain fracture skeleton data;
performing fracture parameter calculation operation on the fracture skeleton data according to a fracture geometric parameter calculation method to obtain a fracture calculation result;
And outputting the crack calculation result to the user terminal.
Further, after the step of denoising the crack binary image according to the image morphology algorithm and the skeletonizing algorithm to obtain a denoised binary image, the method specifically comprises the following steps:
performing corrosion operation on the crack binary image to obtain a corrosion binary image;
and performing expansion operation on the corrosion binary image to obtain the denoising binary image.
Further, the step of performing skeleton extraction operation on the denoising binary image according to a skeletonizing algorithm to obtain fracture skeleton data specifically comprises the following steps:
approximating the shape of the crack in the crack binary graph according to the method of the maximum inscribed sphere diameter, so that each maximum inscribed sphere is at least in contact with two points at the edge of the crack;
after traversing the cracks in the crack binary diagram, obtaining a sphere center data set of all maximum inscribed spheres;
and taking the sphere center data set as fracture skeleton data in the fracture binary image.
Further, the fracture calculation result includes a fracture length, and the step of performing a fracture parameter calculation operation on the fracture skeleton data according to a fracture geometry parameter calculation method to obtain the fracture calculation result specifically includes the following steps:
Dividing the fracture skeleton data according to a differential method to obtain N skeleton branch nodes, wherein N is an integer greater than or equal to 1;
respectively calculating the straight line distance between adjacent framework branch nodes to obtain N-1 distance data;
and summing the N-1 distance data to obtain the crack length of the crack calculation result.
Further, the fracture calculation result includes an average fracture width, and the step of performing fracture parameter calculation operation on the fracture skeleton data according to a fracture geometry parameter calculation method to obtain the fracture calculation result specifically includes the following steps:
dividing the fracture skeleton data according to a differential method to obtain M skeleton branch nodes, wherein M is an integer greater than or equal to 1;
calculating normal vectors of the M skeleton branch nodes according to a kd-tree algorithm and singular value decomposition, and extending the normal vectors to a first intersection point and a second intersection point of the crack edges of the crack skeleton data;
calculating a first width distance and a second width distance between each framework branch node and the first intersection point and the second intersection point respectively, and summing the first width distance and the second width distance to obtain the crack width of each framework branch node;
And calculating the average value of the crack widths of all framework branch nodes to obtain the average crack width of the crack calculation result.
In order to solve the technical problem, the embodiment of the application also provides a bridge crack detection device, which adopts the following technical scheme:
the image acquisition module is used for receiving the original bridge image sent by the user terminal;
the crack segmentation module is used for calling a trained UNet++ model, inputting the original bridge picture into the trained UNet++ model to perform crack segmentation operation to obtain a crack binary image, wherein the UNet++ model is combined with a channel attention module and a space attention module;
the denoising module is used for denoising the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoising binary image;
the skeleton extraction module is used for carrying out skeleton extraction operation on the denoising binary image according to a skeletonization algorithm to obtain fracture skeleton data;
the parameter calculation module is used for carrying out crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method to obtain a crack calculation result;
and the result output module is used for outputting the crack calculation result to the user terminal.
Further, the denoising module includes:
the corrosion submodule is used for carrying out corrosion operation on the crack binary image to obtain a corrosion binary image;
and the expansion sub-module is used for carrying out expansion operation on the corrosion binary image to obtain the denoising binary image.
Further, the skeleton extraction module includes:
a maximum inscribed sphere submodule, which is used for approximating the shape of the crack in the crack binary diagram according to the method of the maximum inscribed sphere diameter, so that each maximum inscribed sphere is at least in two-point contact with the edge of the crack;
the sphere center data acquisition sub-module is used for acquiring sphere center data sets of all maximum inscribed spheres after traversing the cracks in the crack binary diagram;
and the skeleton extraction sub-module is used for taking the sphere center data set as the fracture skeleton data in the fracture binary image.
In order to solve the above technical problems, the embodiments of the present application further provide a computer device, which adopts the following technical schemes:
the bridge crack detection method comprises a memory and a processor, wherein the memory stores computer readable instructions, and the processor executes the computer readable instructions to realize the steps of the bridge crack detection method.
In order to solve the above technical problems, embodiments of the present application further provide a computer readable storage medium, which adopts the following technical solutions:
the computer readable storage medium has stored thereon computer readable instructions which when executed by a processor implement the steps of the bridge crack detection method as described above.
The application provides a bridge crack detection method, which comprises the following steps: receiving an original bridge picture sent by a user terminal; calling a trained UNet++ model, and inputting the original bridge picture to the trained UNet++ model to perform crack segmentation operation to obtain a crack binary image, wherein the UNet++ model combines a channel attention module and a space attention module; denoising the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoising binary image; performing skeleton extraction operation on the denoising binary image according to a skeletonization algorithm to obtain fracture skeleton data; performing fracture parameter calculation operation on the fracture skeleton data according to a fracture geometric parameter calculation method to obtain a fracture calculation result; and outputting the crack calculation result to the user terminal. Compared with the prior art, the method and the device realize effective segmentation of the cracks by utilizing the improved UNet++ network, thereby achieving the purpose of efficiently and accurately detecting the cracks of the bridge. By constructing a deep learning framework, a PyTorch-based UNet++ network model is applied to a bridge crack data set, and the model is improved to realize the identification and detection of cracks; meanwhile, by combining the channel attention module and the spatial attention module, the CBAM module can help the neural network to better understand the characteristics in the image, so that the performance of the model is improved.
Drawings
For a clearer description of the solution in the present application, a brief description will be given below of the drawings that are needed in the description of the embodiments of the present application, it being obvious that the drawings in the following description are some embodiments of the present application, and that other drawings may be obtained from these drawings without inventive effort for a person of ordinary skill in the art.
FIG. 1 is an exemplary system architecture diagram in which the present application may be applied;
fig. 2 is a flowchart of an implementation of a bridge crack detection method according to an embodiment of the present application;
FIG. 3 is a schematic structural diagram of a CBAM module structure according to an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of a channel attention module structure according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of a spatial attention module structure according to an embodiment of the present disclosure;
FIG. 6 is a schematic diagram of a crack noise point provided in accordance with an embodiment of the present application;
fig. 7 is a schematic structural diagram of a bridge crack detection device according to a second embodiment of the present disclosure;
FIG. 8 is a schematic structural diagram of one embodiment of a computer device according to the present application.
Detailed Description
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having" and any variations thereof in the description and claims of the present application and in the description of the figures above are intended to cover non-exclusive inclusions. The terms first, second and the like in the description and in the claims or in the above-described figures, are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.
In order to better understand the technical solutions of the present application, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings.
As shown in fig. 1, a system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
The user may interact with the server 105 via the network 104 using the terminal devices 101, 102, 103 to receive or send messages or the like. Various communication client applications, such as a web browser application, a shopping class application, a search class application, an instant messaging tool, a mailbox client, social platform software, etc., may be installed on the terminal devices 101, 102, 103.
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, electronic book readers, MP3 players (Moving Picture Experts Group Audio Layer III, dynamic video expert compression standard audio plane 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic video expert compression standard audio plane 4) players, laptop and desktop computers, and the like.
The server 105 may be a server providing various services, such as a background server providing support for pages displayed on the terminal devices 101, 102, 103.
It should be noted that, the bridge crack detection method provided in the embodiments of the present application is generally executed by a server/terminal device, and accordingly, the bridge crack detection device is generally disposed in the server/terminal device.
It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
With continued reference to fig. 2, a flow chart of one embodiment of a bridge crack detection method according to the present application is shown. The bridge crack detection method comprises the following steps: step S201, step S202, step S203, step S204, and step S205.
In step S201, an original bridge picture sent by a user terminal is received.
In the embodiment of the present application, the user terminal refers to a terminal device for performing the image processing method for preventing document abuse provided in the present application, and the user terminal may be a mobile terminal such as a mobile phone, a smart phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet personal computer), a PMP (portable multimedia player), a navigation device, etc., and a fixed terminal such as a digital TV, a desktop computer, etc., and it should be understood that the examples of the user terminal herein are merely for convenience of understanding and are not intended to limit the present application.
In the embodiment of the application, the crack detection system is a user interaction interface (GUI) designed based on PyQt5, and Pycharms are used as a development platform and QT Designer is used as a development tool. PyQt5 provides a plurality of practical tools and classes, can rapidly develop GUI application programs with complete functions and easy maintenance, and the PyQt5 can directly call various functions of Qt based on a Qt framework, and the specific development environment is shown in the following table:
in step S202, a trained unet++ model is invoked, and an original bridge image is input to the trained unet++ model to perform a crack segmentation operation, so as to obtain a crack binary image, wherein the unet++ model combines a channel attention module and a spatial attention module.
In the embodiment of the application, in order to alleviate the problems of uneven crack pixels and background pixels, unstructured crack morphology and the like of the segmentation result of the unet++ network, the conventional segmentation algorithm model is difficult to accurately acquire the crack edge and detail information, and the convolution attention module (Convolutional Block Attention Module, CBAM) combining the channel and the spatial attention is adopted. The CBAM module consists of two sub-modules: a channel attention module (Channel Attention Module, CAM) and a spatial attention module (Spatial Attention Module, SAM). The channel attention module mainly focuses on the importance of each channel in the image, and the weight of each channel is determined by calculating the average value and the maximum value of each channel. The spatial attention module then focuses on the importance of each pixel in the image, and the weight of each pixel is determined by calculating the average and maximum value of each pixel. By combining these two sub-modules, the CBAM module can help the neural network to better understand the features in the image, thereby improving the performance of the model.
In the embodiment of the application, since the CBAM is a lightweight universal module, the CBAM can be conveniently integrated into a UNet++ model. The structure of the system is shown in figures 3, 4 and 5, and mainly comprises a channel (channel) attention mechanism module and a space (spatial) attention mechanism module, so that better effect can be achieved compared with the attention mechanism which only focuses on the channel.
In the embodiment of the present application, the CBAM overall flow is as follows:
given an intermediate feature map (F.epsilon.RC.times.H.times.W) as input, for each feature map its channel attention and spatial attention are calculated separately. Firstly, calculating importance scores of all channels, and then carrying out weighted summation on all the channels to obtain a channel scalar weight; and then calculating importance scores of each pixel under different spatial scales, and calculating scores of different scales for weighted summation to obtain a spatial scalar weight. Multiplying the channel attention and the spatial attention yields the final attention weight. The attention weight is multiplied by the feature map to obtain a feature map that emphasizes important features. In the channel attention module, the dimension of the input feature map is assumed to be H multiplied by W multiplied by C, the dimension of the input feature map is firstly reduced from three-dimensional height, width and channel number to two-dimensional height, width (H multiplied by W), then each channel is respectively subjected to maximum pooling and average pooling to obtain two vectors of 1 multiplied by C, then the two output vectors are respectively sent to two layers of full-connection layers, the obtained two vectors are added, a weight coefficient between 0 and 1 is obtained through a Sigmoid function, and the weight coefficient of each channel is multiplied by the feature map corresponding to the channel to obtain the weighted feature map. After the channel attention module outputs, the space attention module is introduced, the input image dimension is H x W x C, the input image is divided into a plurality of small blocks, the vector of each small block is calculated to represent the characteristics of the small block, and then the weight of the vector is calculated to represent the importance degree of the small block to the whole image. The feature vector of the patch is multiplied by its weight to obtain a weighted feature vector. And adding all the feature vectors to obtain the feature vector of the whole image, and further obtaining a final feature map.
The combination of channel attention and spatial attention of the CBAM attention mechanism allows the model to better understand the characteristics of the input data. Channel attention can help models learn the correlation between different channels, thereby better capturing the high-level semantic information of the input data. While spatial attention may help the model focus on important spatial locations in the input data, thereby better capturing local features of the input data. The combination can enable the model to process input data more flexibly, and improve the performance and generalization capability of the model. Meanwhile, the CBAM attention mechanism can be applied to various visual tasks.
According to the network model provided by the invention, the CBAM attention module is added into a fracture detection backbone network convolution stage, and the detection network can better analyze the context information of the model. In addition, the feature map of each convolution stage is more compact, the noise immunity of the network is improved, the false detection rate is reduced, and the accuracy and the robustness of crack detection are improved.
In step S203, denoising the crack binary image according to the image morphology algorithm and the skeletonizing algorithm to obtain a denoised binary image.
In this embodiment of the present application, after the binary image of the crack is obtained, the binary image also has small noise points except for the crack, as shown in fig. 6, and some problems affect the calculation of the geometric parameters of the subsequent crack, so that the binary image of the crack needs to be denoised.
In step S204, skeleton extraction operation is performed on the denoising binary image according to a skeletonizing algorithm, so as to obtain fracture skeleton data.
In the present embodiments, detecting only cracks is not sufficient to make structural health assessment and maintenance decisions. On the basis of crack detection, the skeletonizing treatment is further carried out on the crack image, so that the center line of the crack can be extracted, the form and the characteristics of the crack, such as the length, the width and the like of the crack, are more accurately described and quantized, the efficiency and the accuracy of crack detection are improved, and therefore, the maintenance and the management of infrastructures such as bridges and the like are better carried out.
In the embodiment of the application, skeleton extraction is also called binary image refinement, and as a result, the original cracks are represented by cracks with single pixel width, so that noise and redundant information in the images are reduced, and the shape and the outline of an object are recognized by a computer; the method has the advantages that the skeleton extraction processing is carried out on the crack image, the crack region can be converted into a group of line segments, the complexity of an image structure is reduced, the morphological information of the crack is extracted, the image processing efficiency is improved, the subsequent data analysis and processing are convenient, meanwhile, the important characteristics of the crack image are reserved, the recognition, classification and judgment of the crack are facilitated, and more reliable data support is provided for the maintenance and management of the bridge. The skeleton extraction method is generally Hilditch, rosenfeld and the medial axis transformation method, and it should be understood that the skeleton extraction method is merely illustrated herein for convenience of understanding and is not limited to the present application.
In step S205, a fracture parameter calculation operation is performed on the fracture skeleton data according to the fracture geometry parameter calculation method, so as to obtain a fracture calculation result.
In step S206, the crack calculation result is output to the user terminal.
In an embodiment of the application, a bridge crack detection method is provided, and an original bridge picture sent by a user terminal is received; calling a trained UNet++ model, and inputting an original bridge picture into the trained UNet++ model to perform crack segmentation operation to obtain a crack binary image, wherein the UNet++ model is combined with a channel attention module and a space attention module; denoising the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoising binary image; performing skeleton extraction operation on the denoising binary image according to a skeletonization algorithm to obtain fracture skeleton data; performing crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method to obtain a crack calculation result; and outputting the crack calculation result to the user terminal. Compared with the prior art, the method and the device have the advantages that the improved UNet++ network is utilized to realize effective segmentation of the cracks, so that the purpose of efficiently and accurately detecting the bridge cracks is achieved. By constructing a deep learning framework, a PyTorch-based UNet++ network model is applied to a bridge crack data set, and the model is improved to realize the identification and detection of cracks; meanwhile, by combining the channel attention module and the spatial attention module, the CBAM module can help the neural network to better understand the characteristics in the image, so that the performance of the model is improved.
In some optional implementations of the present embodiment, step S203 specifically includes the following steps:
performing corrosion operation on the crack binary image to obtain a corrosion binary image;
and performing expansion operation on the corrosion binary image to obtain a denoising binary image.
In the embodiment of the present application, the purpose of the denoising is to eliminate the white noise point in fig. 6, and reduce the influence on the calculation of the geometry parameters of the crack as much as possible. The binary image denoising algorithm is an image morphological open operation.
In the embodiment of the application, for the binary image extracted by the algorithm of the application, firstly, etching operation is performed to reduce the edge area of the object so that noise points or small objects cannot disappear; and then performing expansion operation to restore the edge area of the object, so as to achieve the purpose of smoothing the edge of the object. In this process, noise or small objects are eliminated, while large objects remain almost unchanged. Because the binary image is processed, different convolution kernel sizes and shapes can be set to adapt to different scenes and requirements.
In some optional implementations of the present embodiment, step S204 specifically includes the following steps:
approximating the shape of the crack in the crack binary map according to the method of the maximum inscribed sphere diameter, so that each maximum inscribed sphere is at least in contact with two points at the edge of the crack;
After traversing the cracks in the crack binary diagram, obtaining a sphere center data set of all the maximum inscribed spheres;
and taking the sphere center data set as fracture skeleton data in a fracture binary image.
In the embodiment of the application, the skeleton of the crack is extracted by adopting a medial axis transformation method, and the algorithm is simple and efficient to realize and can well extract the skeleton of the image. The Medial axis function is an image function based on a Medial axis transformation method, the binary image of the crack is input, then edge detection is carried out on the binary image of the crack to obtain crack edges, and then distance transformation is carried out on the edges to obtain the distance from each pixel to the nearest edge pixel. And marking pixels with the distance less than or equal to a certain threshold value as a foreground, and marking the rest pixels as a background. And finally, carrying out skeleton extraction on the binarized image to obtain the central axis of the input image. The principle of the medial axis transformation method is to approach the shape of the crack by a method of maximum inscribed sphere diameter, and the inscribed sphere must be in contact with at least two points of the crack edge. After traversing the complete crack, the central axis in the central axis transformation method is the set of the sphere centers of all inner spheres. The method can well approximate the geometric structure of the crack and can completely retain details.
In some optional implementations of the present embodiment, the fracture calculation result includes a fracture length, and step S205 specifically includes the following steps:
dividing the fracture skeleton data according to a differential method to obtain N skeleton branch nodes, wherein N is an integer greater than or equal to 1;
respectively calculating the straight line distance of adjacent skeleton branch nodes to obtain N-1 distance data;
and summing the N-1 distance data to obtain the crack length of the crack calculation result.
In the embodiment of the application, since the crack has a bending process, the length of the crack cannot be directly obtained by using a geometric method, so the idea of differentiating the length is applied to the calculation of the crack length, and the principle is that each section of length is calculated from one end of the crack along the skeleton until the whole crack is traversed, and the calculated lengths are accumulated to obtain the pixel length of the crack.
In the embodiment of the present application, the process of calculating the crack length may be:
1. traversing the debranched skeleton to obtain coordinates of n sets of target points between the start point and the end point;/>=1,2,…,n。
2. Calculating the linear distance between adjacent points:
3. adding each linear distance:
repeating the above 1-3 processes until the whole crack skeleton is traversed, and obtaining the pixel length of the crack.
In some optional implementations of the present embodiment, the crack calculation result includes an average crack width, and step S205 specifically includes the following steps:
dividing the fracture skeleton data according to a differential method to obtain M skeleton branch nodes, wherein M is an integer greater than or equal to 1;
calculating normal vectors of M skeleton branch nodes according to a kd-tree algorithm and singular value decomposition, and extending the normal vectors to a first intersection point and a second intersection point of the crack edges of the crack skeleton data;
calculating a first width distance and a second width distance between each framework branch node and the first intersection point and the second intersection point respectively, and summing the first width distance and the second width distance to obtain the crack width of each framework branch node;
and calculating the average value of the crack widths of all the framework branch nodes to obtain the average crack width of the crack calculation result.
In the embodiment of the application, the width recognition of the bridge cracks is very important, so that engineers can be helped to evaluate the safety of the bridge, and corresponding measures are taken to repair and strengthen the cracks so as to ensure the stability and safety of the bridge. In general, a larger crack width indicates a more severe crack, and more urgent and efficient repair measures are required. Timely detection and evaluation of the width of the crack are needed, and corresponding repair measures are adopted to ensure the safety and stability of the structure.
The crack skeleton is extracted by an algorithm of a medial axis transformation method, and the crack width is further calculated on the basis of the extracted crack skeleton, wherein the specific calculation steps are as follows:
(1) Normal vector estimation: firstly, searching for adjacent points of a given point on a crack image, calculating a normal vector of a skeleton line by using a kd-tree algorithm and then using Singular Value Decomposition (SVD), and calculating the width of the crack in the direction of the normal vector;
(2) Width calculation: for a given position, according to the previous part, estimating the square vector of the skeleton line at the position, taking the square vector as the y-axis direction of a local coordinate system, and taking the y-axis direction as the x-axis direction, then transforming the edge line of the crack into the local coordinate system, searching two crack edge line points closest to the skeleton line normal vector in the local coordinate system, and calculating the intersection point of a line segment formed by the two points and the skeleton line normal vector. And respectively calculating the distance between the intersection point and the central axis for the two crack edge lines on the left side and the right side, obtaining the width of the crack by adding the left distance and the right distance, traversing all points on the central axis, and repeating the algorithm to obtain the maximum width information and the average width information.
In the embodiment of the present application, the process of calculating the crack width and the average crack width may be:
1. Traversing the debranching skeleton to obtain coordinates (x) of n sets of target points between the start point and the end point i ,y i ),i=1,2,…,n;。
2. The direction of the skeleton can be obtained from the coordinates of the points on the skeleton, and the normal thereof can be determined, and the normal is lengthened and intersected with the crack edge to obtain two intersection points A, B. Its coordinate point is (x) ai ,y ai )、(x bi ,y bi )
3. At this time, the distance between the target point and the A, B point is calculated:
the width information of the crack at the position of the target point can be obtained by adding the distances on two sides:
4. this operation is repeated until the crack width at the last point is calculated.
5. Comparing the maximum crack width for each location:
6. average crack width was calculated:
in practical application, in order to verify the accuracy of the method for calculating the parameters of the crack of the present chapter, 100 crack images under different backgrounds are selected in an experiment, the true values of the length and width pixels of the crack in the images are calculated first, then the crack in the images is identified by using the algorithm of the present disclosure to obtain a binary image of the crack, the binary image of the crack is subjected to skeletonizing treatment to obtain a skeleton image of the crack, and the length and width pixels of the crack are calculated. And comparing the difference between the true value pixel and the predicted value pixel, and analyzing the cause of the error. The fracture geometry calculation results are shown in the following table:
Test results show that the identification capability of the seal algorithm on the transverse crack, the vertical crack and the oblique crack is relatively strong, and the cracks can be extracted well. The predicted value of the geometric parameter is smaller than the true value, but the error is smaller, and in the allowable range, the analysis reasons may not identify the tiny cracks, so that the crack tips or the crack connection positions cannot be identified, and the numerical value is smaller in calculation.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by computer readable instructions stored in a computer readable storage medium that, when executed, may comprise the steps of the embodiments of the methods described above. The storage medium may be a nonvolatile storage medium such as a magnetic disk, an optical disk, a Read-Only Memory (ROM), or a random access Memory (Random Access Memory, RAM).
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated in the present application. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.
Example two
With further reference to fig. 7, as an implementation of the method shown in fig. 2, the present application provides an embodiment of a bridge crack detection apparatus, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus may be specifically applied to various electronic devices.
As shown in fig. 7, the bridge crack detection device 200 of the present embodiment includes: a picture acquisition module 210, a crack segmentation module 220, a denoising module 230, a skeleton extraction module 240, a parameter calculation module 250, and a result output module 260, wherein:
the image obtaining module 210 is configured to receive an original bridge image sent by the user terminal;
the crack segmentation module 220 is used for calling the trained unet++ model, inputting the original bridge picture into the trained unet++ model for crack segmentation operation, and obtaining a crack binary image, wherein the unet++ model combines the channel attention module and the space attention module;
the denoising module 230 is configured to denoise the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoised binary image;
the skeleton extraction module 240 is configured to perform skeleton extraction operation on the denoising binary image according to a skeletonization algorithm, so as to obtain fracture skeleton data;
The parameter calculation module 250 is configured to perform a crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method, so as to obtain a crack calculation result;
and the result output module 260 is configured to output the crack calculation result to the user terminal.
In this embodiment, there is provided a bridge crack detection device 200, including: the image obtaining module 210 is configured to receive an original bridge image sent by the user terminal; the crack segmentation module 220 is used for calling the trained unet++ model, inputting the original bridge picture into the trained unet++ model for crack segmentation operation, and obtaining a crack binary image, wherein the unet++ model combines the channel attention module and the space attention module; the denoising module 230 is configured to denoise the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoised binary image; the skeleton extraction module 240 is configured to perform skeleton extraction operation on the denoising binary image according to a skeletonization algorithm, so as to obtain fracture skeleton data; the parameter calculation module 250 is configured to perform a crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method, so as to obtain a crack calculation result; and the result output module 260 is configured to output the crack calculation result to the user terminal. Compared with the prior art, the method and the device realize effective segmentation of the cracks by utilizing the improved UNet++ network, thereby achieving the purpose of efficiently and accurately detecting the cracks of the bridge. By constructing a deep learning framework, a PyTorch-based UNet++ network model is applied to a bridge crack data set, and the model is improved to realize the identification and detection of cracks; meanwhile, by combining the channel attention module and the spatial attention module, the CBAM module can help the neural network to better understand the characteristics in the image, so that the performance of the model is improved.
In some optional implementations of this embodiment, the denoising module 230 includes:
the corrosion submodule is used for carrying out corrosion operation on the crack binary image to obtain a corrosion binary image;
and the expansion sub-module is used for carrying out expansion operation on the corrosion binary image to obtain a denoising binary image.
In some optional implementations of this embodiment, the skeleton extraction module 240 includes:
the maximum inscribed sphere submodule is used for approximating the shape of the crack in the crack binary diagram according to the method of the maximum inscribed sphere diameter, so that each maximum inscribed sphere is at least in contact with two points at the edge of the crack;
the sphere center data acquisition sub-module is used for acquiring sphere center data sets of all maximum inscribed spheres after traversing the cracks in the crack binary diagram;
and the skeleton extraction submodule is used for taking the sphere center data set as the crack skeleton data in the crack binary image.
In order to solve the technical problems, the embodiment of the application also provides computer equipment. Referring specifically to fig. 8, fig. 8 is a basic structural block diagram of a computer device according to the present embodiment.
The computer device 300 includes a memory 310, a processor 320, and a network interface 330 communicatively coupled to each other via a system bus. It should be noted that only computer device 300 having components 310-330 is shown in the figures, but it should be understood that not all of the illustrated components need be implemented, and that more or fewer components may be implemented instead. It will be appreciated by those skilled in the art that the computer device herein is a device capable of automatically performing numerical calculations and/or information processing in accordance with predetermined or stored instructions, the hardware of which includes, but is not limited to, microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASICs), programmable gate arrays (fields-Programmable Gate Array, FPGAs), digital processors (Digital Signal Processor, DSPs), embedded devices, etc.
The computer equipment can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing equipment. The computer equipment can perform man-machine interaction with a user through a keyboard, a mouse, a remote controller, a touch pad or voice control equipment and the like.
The memory 310 includes at least one type of readable storage medium including flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), programmable Read Only Memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 310 may be an internal storage unit of the computer device 300, such as a hard disk or a memory of the computer device 300. In other embodiments, the memory 310 may also be an external storage device of the computer device 300, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 300. Of course, the memory 310 may also include both internal storage units and external storage devices of the computer device 300. In this embodiment, the memory 310 is generally used to store an operating system and various application software installed on the computer device 300, such as computer readable instructions of a bridge crack detection method. In addition, the memory 310 may also be used to temporarily store various types of data that have been output or are to be output.
The processor 320 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 320 is generally used to control the overall operation of the computer device 300. In this embodiment, the processor 320 is configured to execute computer readable instructions stored in the memory 310 or process data, such as computer readable instructions for executing the bridge crack detection method.
The network interface 330 may include a wireless network interface or a wired network interface, the network interface 330 typically being used to establish communication connections between the computer device 300 and other electronic devices.
According to the computer equipment, the improved UNet++ network is utilized to realize effective segmentation of the cracks, so that the purpose of efficiently and accurately detecting the bridge cracks is achieved. By constructing a deep learning framework, a PyTorch-based UNet++ network model is applied to a bridge crack data set, and the model is improved to realize the identification and detection of cracks; meanwhile, by combining the channel attention module and the spatial attention module, the CBAM module can help the neural network to better understand the characteristics in the image, so that the performance of the model is improved.
The present application also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions executable by at least one processor to cause the at least one processor to perform the steps of the bridge crack detection method as described above.
The computer readable storage medium provided by the application realizes effective segmentation of the cracks by utilizing the improved UNet++ network, thereby achieving the purpose of efficiently and accurately detecting the cracks of the bridge. By constructing a deep learning framework, a PyTorch-based UNet++ network model is applied to a bridge crack data set, and the model is improved to realize the identification and detection of cracks; meanwhile, by combining the channel attention module and the spatial attention module, the CBAM module can help the neural network to better understand the characteristics in the image, so that the performance of the model is improved.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method described in the embodiments of the present application.
It is apparent that the embodiments described above are only some embodiments of the present application, but not all embodiments, the preferred embodiments of the present application are given in the drawings, but not limiting the patent scope of the present application. This application may be embodied in many different forms, but rather, embodiments are provided in order to provide a more thorough understanding of the present disclosure. Although the present application has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described in the foregoing, or equivalents may be substituted for elements thereof. All equivalent structures made by the specification and the drawings of the application are directly or indirectly applied to other related technical fields, and are also within the protection scope of the application.
Claims (10)
1. The bridge crack detection method is characterized by comprising the following steps of:
receiving an original bridge picture sent by a user terminal;
calling a trained UNet++ model, and inputting the original bridge picture to the trained UNet++ model to perform crack segmentation operation to obtain a crack binary image, wherein the UNet++ model combines a channel attention module and a space attention module;
Denoising the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoising binary image;
performing skeleton extraction operation on the denoising binary image according to a skeletonization algorithm to obtain fracture skeleton data;
performing fracture parameter calculation operation on the fracture skeleton data according to a fracture geometric parameter calculation method to obtain a fracture calculation result;
and outputting the crack calculation result to the user terminal.
2. The bridge crack detection method according to claim 1, wherein after the step of denoising the crack binary image according to the image morphology algorithm and the skeletonizing algorithm to obtain a denoised binary image, the method specifically comprises the following steps:
performing corrosion operation on the crack binary image to obtain a corrosion binary image;
and performing expansion operation on the corrosion binary image to obtain the denoising binary image.
3. The bridge crack detection method according to claim 1, wherein the step of performing skeleton extraction operation on the denoising binary image according to a skeletonizing algorithm to obtain crack skeleton data specifically comprises the following steps:
approximating the shape of the crack in the crack binary graph according to the method of the maximum inscribed sphere diameter, so that each maximum inscribed sphere is at least in contact with two points at the edge of the crack;
After traversing the cracks in the crack binary diagram, obtaining a sphere center data set of all maximum inscribed spheres;
and taking the sphere center data set as fracture skeleton data in the fracture binary image.
4. The bridge crack detection method according to claim 1, wherein the crack calculation result includes a crack length, and the step of performing a crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method to obtain a crack calculation result specifically includes the following steps:
dividing the fracture skeleton data according to a differential method to obtain N skeleton branch nodes, wherein N is an integer greater than or equal to 1;
respectively calculating the straight line distance between adjacent framework branch nodes to obtain N-1 distance data;
and summing the N-1 distance data to obtain the crack length of the crack calculation result.
5. The bridge crack detection method according to claim 1, wherein the crack calculation result includes an average crack width, and the step of performing a crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method to obtain a crack calculation result specifically includes the following steps:
Dividing the fracture skeleton data according to a differential method to obtain M skeleton branch nodes, wherein M is an integer greater than or equal to 1;
calculating normal vectors of the M skeleton branch nodes according to a kd-tree algorithm and singular value decomposition, and extending the normal vectors to a first intersection point and a second intersection point of the crack edges of the crack skeleton data;
calculating a first width distance and a second width distance between each framework branch node and the first intersection point and the second intersection point respectively, and summing the first width distance and the second width distance to obtain the crack width of each framework branch node;
and calculating the average value of the crack widths of all framework branch nodes to obtain the average crack width of the crack calculation result.
6. Bridge crack detection device, characterized by includes:
the image acquisition module is used for receiving the original bridge image sent by the user terminal;
the crack segmentation module is used for calling a trained UNet++ model, inputting the original bridge picture into the trained UNet++ model to perform crack segmentation operation to obtain a crack binary image, wherein the UNet++ model is combined with a channel attention module and a space attention module;
The denoising module is used for denoising the crack binary image according to an image morphology algorithm and a skeletonizing algorithm to obtain a denoising binary image;
the skeleton extraction module is used for carrying out skeleton extraction operation on the denoising binary image according to a skeletonization algorithm to obtain fracture skeleton data;
the parameter calculation module is used for carrying out crack parameter calculation operation on the crack skeleton data according to a crack geometric parameter calculation method to obtain a crack calculation result;
and the result output module is used for outputting the crack calculation result to the user terminal.
7. The bridge crack detection device of claim 6, wherein the denoising module comprises:
the corrosion submodule is used for carrying out corrosion operation on the crack binary image to obtain a corrosion binary image;
and the expansion sub-module is used for carrying out expansion operation on the corrosion binary image to obtain the denoising binary image.
8. The bridge crack detection device of claim 6, wherein the skeleton extraction module comprises:
a maximum inscribed sphere submodule, which is used for approximating the shape of the crack in the crack binary diagram according to the method of the maximum inscribed sphere diameter, so that each maximum inscribed sphere is at least in two-point contact with the edge of the crack;
The sphere center data acquisition sub-module is used for acquiring sphere center data sets of all maximum inscribed spheres after traversing the cracks in the crack binary diagram;
and the skeleton extraction sub-module is used for taking the sphere center data set as the fracture skeleton data in the fracture binary image.
9. A computer device comprising a memory having stored therein computer readable instructions which when executed by the processor implement the steps of the bridge fracture detection method of any one of claims 1 to 5.
10. A computer readable storage medium having stored thereon computer readable instructions which when executed by a processor implement the steps of the bridge crack detection method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311608863.2A CN117314917B (en) | 2023-11-29 | 2023-11-29 | Bridge crack detection method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311608863.2A CN117314917B (en) | 2023-11-29 | 2023-11-29 | Bridge crack detection method and device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117314917A CN117314917A (en) | 2023-12-29 |
| CN117314917B true CN117314917B (en) | 2024-02-06 |
Family
ID=89255698
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311608863.2A Active CN117314917B (en) | 2023-11-29 | 2023-11-29 | Bridge crack detection method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117314917B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745477A (en) * | 2014-01-22 | 2014-04-23 | 汕头大学 | Machine vision-based building structure crack detection and repair method |
| KR101922831B1 (en) * | 2018-01-31 | 2018-11-27 | 서울시립대학교 산학협력단 | Image analysis apparatus and method for discriminating concrete status |
| CN112308826A (en) * | 2020-10-23 | 2021-02-02 | 南京航空航天大学 | Bridge structure surface defect detection method based on convolutional neural network |
-
2023
- 2023-11-29 CN CN202311608863.2A patent/CN117314917B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745477A (en) * | 2014-01-22 | 2014-04-23 | 汕头大学 | Machine vision-based building structure crack detection and repair method |
| KR101922831B1 (en) * | 2018-01-31 | 2018-11-27 | 서울시립대학교 산학협력단 | Image analysis apparatus and method for discriminating concrete status |
| CN112308826A (en) * | 2020-10-23 | 2021-02-02 | 南京航空航天大学 | Bridge structure surface defect detection method based on convolutional neural network |
Non-Patent Citations (1)
| Title |
|---|
| 基于深度卷积神经网络融合模型的路面裂缝识别方法;孙朝云;马志丹;李伟;郝雪丽;申浩;;长安大学学报(自然科学版)(第04期);5-17 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117314917A (en) | 2023-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110738207B (en) | Character detection method for fusing character area edge information in character image | |
| CN111008597B (en) | Space identification method and device for CAD drawing, electronic equipment and storage medium | |
| CN113657390B (en) | Training method of text detection model and text detection method, device and equipment | |
| WO2022105125A1 (en) | Image segmentation method and apparatus, computer device, and storage medium | |
| CN110084238B (en) | Finger vein image segmentation method and device based on LadderNet network and storage medium | |
| CN111444807B (en) | Target detection method, device, electronic equipment and computer readable medium | |
| CN111899270B (en) | Card frame detection method, device, equipment and readable storage medium | |
| CN110414649B (en) | DM code positioning method, device, terminal and storage medium | |
| CN112989995B (en) | Text detection method and device and electronic equipment | |
| CN112861575A (en) | Pedestrian structuring method, device, equipment and storage medium | |
| CN110751154A (en) | Complex environment multi-shape text detection method based on pixel-level segmentation | |
| CN116311214B (en) | License plate recognition method and device | |
| CN111260564A (en) | Image processing method and device and computer storage medium | |
| CN111951283A (en) | Medical image identification method and system based on deep learning | |
| CN111709387A (en) | Building segmentation method and system for high-resolution remote sensing image | |
| CN111815748B (en) | Animation processing method and device, storage medium and electronic equipment | |
| CN111192312B (en) | Depth image acquisition method, device, equipment and medium based on deep learning | |
| CN117314917B (en) | Bridge crack detection method and device, computer equipment and storage medium | |
| CN111583417A (en) | Method and device for constructing indoor VR scene with combined constraint of image semantics and scene geometry, electronic equipment and medium | |
| CN112381118A (en) | Method and device for testing and evaluating dance test of university | |
| CN114511862B (en) | Form identification method and device and electronic equipment | |
| CN113610856B (en) | Method and device for training image segmentation model and image segmentation | |
| CN114936395A (en) | Household type graph recognition method and device, computer equipment and storage medium | |
| CN112084874A (en) | Object detection method and device and terminal equipment | |
| CN115797633B (en) | Remote sensing image segmentation method, remote sensing image segmentation system, storage medium and electronic equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |