CN114775457A - Detection device for repairing ancient bridge and detection method thereof - Google Patents
Detection device for repairing ancient bridge and detection method thereof Download PDFInfo
- Publication number
- CN114775457A CN114775457A CN202210256905.XA CN202210256905A CN114775457A CN 114775457 A CN114775457 A CN 114775457A CN 202210256905 A CN202210256905 A CN 202210256905A CN 114775457 A CN114775457 A CN 114775457A
- Authority
- CN
- China
- Prior art keywords
- ancient
- bridge
- layer
- crack
- health
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 71
- 230000036541 health Effects 0.000 claims abstract description 65
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000012544 monitoring process Methods 0.000 claims abstract description 17
- 238000003062 neural network model Methods 0.000 claims abstract description 17
- 238000004891 communication Methods 0.000 claims abstract description 14
- 238000003745 diagnosis Methods 0.000 claims abstract description 14
- 238000011156 evaluation Methods 0.000 claims abstract description 14
- 238000013145 classification model Methods 0.000 claims abstract description 11
- 230000008439 repair process Effects 0.000 claims description 60
- 238000012549 training Methods 0.000 claims description 36
- 239000011159 matrix material Substances 0.000 claims description 25
- 230000011218 segmentation Effects 0.000 claims description 23
- 239000002994 raw material Substances 0.000 claims description 8
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 3
- 239000007921 spray Substances 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000003862 health status Effects 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
- E01D—CONSTRUCTION OF BRIDGES, ELEVATED ROADWAYS OR VIADUCTS; ASSEMBLY OF BRIDGES
- E01D22/00—Methods or apparatus for repairing or strengthening existing bridges ; Methods or apparatus for dismantling bridges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2323—Non-hierarchical techniques based on graph theory, e.g. minimum spanning trees [MST] or graph cuts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q9/00—Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20161—Level set
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q2209/00—Arrangements in telecontrol or telemetry systems
- H04Q2209/80—Arrangements in the sub-station, i.e. sensing device
- H04Q2209/84—Measuring functions
Abstract
The invention belongs to the technical field of detection for repairing an ancient-built bridge, and discloses a detection device for repairing the ancient-built bridge and a detection method thereof, wherein the detection device for repairing the ancient-built bridge comprises: the device comprises a case, a storage bin, a lifting pipe, an ejector, a camera, a controller, a crack detection module, a health assessment module, a wireless communication module and a fault diagnosis module. According to the method, the crack detection module is used for detecting and classifying the cracks of the ancient-constructed bridge by adopting the bridge crack classification model, so that the crack detection precision and efficiency of the ancient-constructed bridge can be obviously improved; meanwhile, newly acquired health monitoring data of the ancient bridge is input into the neural network model through the health evaluation module, and the predicted health state grade of the ancient bridge is output, so that the health state of the ancient bridge is accurately and effectively evaluated and predicted.
Description
Technical Field
The invention belongs to the technical field of detection for repairing an ancient bridge, and particularly relates to a detection device for repairing the ancient bridge and a detection method thereof.
Background
The bridge is generally a structure which is erected on rivers, lakes and seas and allows vehicles, pedestrians and the like to smoothly pass through. In order to adapt to the modern high-speed development traffic industry, bridges are also extended to be built for spanning mountain stream, unfavorable geology or meeting other traffic requirements, so that the buildings which are more convenient to pass are constructed. The bridge generally comprises an upper structure, a lower structure, a support and an auxiliary structure, wherein the upper structure is also called a bridge span structure and is a main structure for spanning obstacles; the lower structure comprises a bridge abutment, a bridge pier and a foundation; the support is a force transmission device arranged at the supporting positions of the bridge span structure and the bridge pier or the bridge abutment; the attached structure refers to bridge end lapping plate, taper slope protection, bank protection, diversion engineering, etc. However, the existing detection device and detection method for repairing the ancient bridge mainly depend on manual detection for detecting and maintaining the cracks of the ancient bridge; the manual detection method is time-consuming and needs a large amount of manpower, material resources and financial resources, not only is the detection precision low and the human influence factors large, but also the cracks cannot be detected visually due to the inaccessibility of the area or the microscopic size of the cracks in many cases; meanwhile, the health of the ancient bridge cannot be accurately evaluated.
In summary, the problems of the prior art are: the existing detection device and the detection method for repairing the ancient bridge mainly depend on manual detection on the detection and maintenance of the ancient bridge cracks; the manual detection method is time-consuming and needs a large amount of manpower, material resources and financial resources, not only is the detection precision low and the human influence factors large, but also the cracks cannot be detected visually due to the inaccessibility of the area or the microscopic size of the cracks in many cases; meanwhile, the health of the ancient bridge cannot be accurately evaluated.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a detection device for repairing an ancient building bridge and a detection method thereof.
The invention is realized in this way, a detection device for repairing an ancient bridge comprises:
the system comprises a case, a storage bin, a lifting pipe, an ejector, a camera, a controller, a crack detection module, a health evaluation module, a wireless communication module and a fault diagnosis module;
a storage bin is fixed on the left side in the case through screws; the top end of the storage bin is sleeved with a lifting pipe; the top end of the lifting pipe is provided with an ejector; the top end of the ejector is fixed with a camera through a screw; the right side of the front side of the case is embedded with a controller; the controller is internally provided with a crack detection module, a health evaluation module, a wireless communication module and a fault diagnosis module;
the crack detection module is used for detecting crack information of the ancient bridge;
the health evaluation module is used for evaluating the health of the ancient bridge;
the wireless communication module is used for connecting a wireless network through a wireless signal to carry out remote control;
and the fault diagnosis module is used for diagnosing the device fault. Detecting the repair state of the ancient bridge repair equipment through a repair state sensor to obtain first signal data comprising the repair state value of the ancient bridge repair equipment;
collecting first signal data through a 4-to-6 24-point repair state sensor, wherein the first signal data comprise a left repair state sensor signal of the left repair state of the ancient bridge repair equipment and a right repair state sensor signal of the right repair state of the ancient bridge repair equipment, and the first signal data are classified according to a normal state and a fault state;
includes preprocessing the first signal data, combining a plurality of 4 x 6 signals, and synthesizing a 48 x 48 signal matrix by using 48 sets of signals, wherein one set of signals includes a left repair status sensor signal and a right repair status sensor signal;
the signal prediction model comprises seven layers; the first layer is an input layer, the first layer is original signal data, the picture size of the first layer is 48 × 1, and the channel is 1; the second layer is a convolution layer, the input of the second layer is the output of the output layer, the filter size of the second layer is 5 x 5, the depth is 32, the second layer is filled with all 0 s, and the output matrix size of the second layer is 48 x 32; the third layer is a pooled layer, the input to the third layer is a 48 x 32 node matrix for the second layer output, the filter size of the third layer is 2 x 2, the step sizes for length and width are both 2, and the output matrix size of the third layer is 24 x 32; the fourth layer is a convolution layer, the input of the fourth layer is the output of the third layer, the filter size of the fourth layer is 5 x 5, the depth is 64, the fourth layer is filled with all 0 s, and the output matrix size of the fourth layer is 24 x 32; the fifth layer is a pooling layer, the input of the fifth layer is a 24 × 64 node matrix of the fourth layer output, the filter size of the fifth layer is 2 × 2, the length and width step lengths are both 2, and the output matrix size of the fifth layer is 12 × 64; the sixth layer is a full connection layer, the input of the sixth layer is the output of the fifth layer, the number of input nodes of the sixth layer is 12 × 64, the number of input nodes of the sixth layer is 9216, and the number of output nodes of the sixth layer is 512; the seventh layer is a full connection layer, the number of input nodes of the seventh layer is 512, and the number of output nodes of the seventh layer is 4.
A detection method of a detection device for repairing an ancient bridge comprises the following steps:
injecting an ancient bridge repairing raw material into a storage bin, lifting the ancient bridge repairing raw material to the position of the ancient bridge through a lifting pipe, collecting an image of the ancient bridge through a camera, and sending the image to a controller for processing;
secondly, detecting the crack information of the ancient bridge through a crack detection module by the controller; evaluating the health of the ancient bridge through a health evaluation module; a wireless communication module is used for connecting a wireless network by using a wireless signal to carry out remote control; diagnosing the device fault through a fault diagnosis module;
and step three, the controller starts the ejector according to the crack information to spray repair raw materials for repairing the ancient bridge cracks.
Further, the crack detection module detection method comprises the following steps:
(1) performing crack segmentation on a group of ancient bridge images acquired by a camera through a segmentation program;
(2) and detecting and classifying the ancient-built bridge cracks by adopting a pre-constructed ancient-built bridge crack classification model according to the crack segmentation result.
Further, the method for performing crack segmentation on a group of ancient bridge images acquired by the camera through the segmentation program comprises the following steps:
dividing cracks in each ancient bridge image as initial contours by adopting a spectral clustering algorithm based on Nystrom approximation theoretical generalization; sequentially mapping the divided cracks to the next ancient bridge image along the upward direction to serve as the initial contour of the crack in the ancient bridge image, and finishing the division of each crack by adopting an improved GAC model until the division of all the ancient bridge images is finished;
and sequentially mapping the cracks segmented along the upward direction to the next ancient bridge image along the downward direction, taking the cracks as initial contours of the cracks in the ancient bridge image, and completing the segmentation of each crack by adopting an improved GAC (generic object model) model until all the ancient bridge images are segmented.
Further, the method for completing the segmentation of each fracture by adopting the improved GAC model comprises the following steps: calculating the gray level mean value and the gray level standard deviation of each divided crack area, and taking the gray level mean value and the gray level standard deviation as the gray level similarity information of each crack area; constructing a gray level similarity information item according to the gray level similarity information; and adding the gray level similarity information item as an external energy item to an energy functional of the GAC model, thereby improving the GAC model.
Further, the construction method of the ancient bridge crack classification model comprises the following steps: collecting original ancient bridge images containing various cracks; carrying out crack marking and crack segmentation on the collected original ancient bridge image to construct a sample data set of the ancient bridge cracks; initially constructing 8 layers of deep convolutional neural networks by combining the global characteristics of the ancient bridge, wherein each deep convolutional neural network comprises a first input layer, a third convolutional layer, a third pooling layer, a second full-connection layer and an output layer, and a softmax classifier is adopted; and training and testing the constructed deep convolution neural network by adopting a sample data set so as to determine the structure and parameters of the ancient bridge crack classification model.
Further, the health assessment module assessment method comprises the following steps:
1) acquiring ancient bridge health monitoring data based on a time sequence through a detection device, and generating a training sample set based on the pre-divided ancient bridge health state levels;
2) training by using a training sample set to obtain a neural network model; and inputting newly acquired ancient bridge health monitoring data into the neural network model, and outputting the predicted ancient bridge health state level.
Further, the training sample set comprises a plurality of training samples based on time sequence, and each training sample comprises a group of the ancient bridge health monitoring data and the corresponding ancient bridge health status level.
Further, the training process of the neural network model comprises the following steps: and taking the ancient building bridge health monitoring data in the training sample as input, taking the corresponding ancient building bridge health state level as output, training the initialized neural network model, calculating a loss function of the neural network model and updating a parameter matrix until a prediction error meets a termination condition.
Further, the training sample set is
Wherein the content of the first and second substances,
denotes tk-the training samples of the time of day,
is a feature vector of n +1 dimensions,
n is more than or equal to 1 for the ancient bridge health monitoring data,
is that
The corresponding health state levels of the ancient bridges, and the set of the health state levels of the ancient bridges is y epsilon { a1,a2,...,aN}，a1,a2,...,aNAll are different values, N is more than or equal to 2.
The invention has the advantages and positive effects that: according to the method, the crack detection module is used for detecting and classifying the cracks of the ancient-constructed bridge by adopting the bridge crack classification model, so that the crack detection precision and efficiency of the ancient-constructed bridge can be obviously improved; meanwhile, newly acquired health monitoring data of the ancient bridge is input into the neural network model through the health evaluation module, and the predicted health state grade of the ancient bridge is output, so that the health state of the ancient bridge is accurately and effectively evaluated and predicted.
Detecting the repair state of the ancient bridge repair equipment through a repair state sensor to obtain first signal data comprising the repair state value of the ancient bridge repair equipment;
collecting first signal data through a 4-by-6 24-point repair state sensor, wherein the first signal data comprise a left repair state sensor signal of the left repair state of the ancient bridge repair equipment and a right repair state sensor signal of the right repair state of the ancient bridge repair equipment, and the first signal data are classified according to a normal state and a fault state; includes preprocessing the first signal data, combining a plurality of 4 x 6 signals, and combining the combined signals into a 48 x 48 signal matrix using 48 sets of signals, wherein one set of signals includes a left repair status sensor signal and a right repair status sensor signal.
Drawings
Fig. 1 is a structural block diagram of a detection device for repairing an ancient bridge according to an embodiment of the invention.
Fig. 2 is a block diagram of a controller according to an embodiment of the present invention.
Fig. 3 is a flowchart of a detection method of the detection device for repairing the ancient bridge according to the embodiment of the invention.
Fig. 4 is a flowchart of a crack detection module detection method according to an embodiment of the present invention.
FIG. 5 is a flowchart of a health assessment module assessment method according to an embodiment of the present invention.
In fig. 1 and 2: 1. a chassis; 2. a storage bin; 3. a lifting pipe; 4. an ejector; 5. a camera; 6. a controller; 7. a crack detection module; 8. a health assessment module; 9. wireless communication module, 10, failure diagnosis module.
Detailed Description
For further understanding of the contents, features and effects of the invention, the following examples are given in conjunction with the accompanying drawings.
The structure of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in fig. 1 and fig. 2, a detection apparatus for repairing an ancient bridge according to an embodiment of the present invention includes: the system comprises a case 1, a storage bin 2, a lifting pipe 3, an ejector 4, a camera 5, a controller 6, a crack detection module 7, a health evaluation module 8, a wireless communication module 9 and a fault diagnosis module 10;
the storage bin 2 is fixed on the left side in the case 1 through screws; the top end of the storage bin 2 is sleeved with a lifting pipe 3; the top end of the lifting pipe 3 is provided with an ejector 4; the top end of the ejector 4 is fixed with a camera 5 through a screw; the right side of the front surface of the case 1 is embedded with a controller 6; a crack detection module 7, a health evaluation module 8, a wireless communication module 9 and a fault diagnosis module 10 are arranged in the controller 6;
the crack detection module 7 is used for detecting crack information of the ancient bridge;
the health evaluation module 8 is used for evaluating the health of the ancient bridge;
the wireless communication module 9 is used for connecting a wireless network through a wireless signal to carry out remote control;
and the fault diagnosis module 10 is used for diagnosing the device fault. Detecting the repair state of the ancient bridge repair equipment through a repair state sensor to obtain first signal data comprising the repair state value of the ancient bridge repair equipment;
collecting first signal data through a 4-by-6 24-point repair state sensor, wherein the first signal data comprise a left repair state sensor signal of the left repair state of the ancient bridge repair equipment and a right repair state sensor signal of the right repair state of the ancient bridge repair equipment, and the first signal data are classified according to a normal state and a fault state;
includes preprocessing the first signal data, combining a plurality of 4 x 6 signals, and synthesizing a 48 x 48 signal matrix by using 48 sets of signals, wherein one set of signals includes a left repair status sensor signal and a right repair status sensor signal;
the signal prediction model comprises seven layers; the first layer is an input layer, the first layer is original signal data, the picture size of the first layer is 48 × 1, and the channel is 1; the second layer is a convolutional layer, the input of the second layer is the output of the output layer, the filter size of the second layer is 5 x 5, the depth is 32, the second layer is filled with all 0 s, and the output matrix size of the second layer is 48 x 32; the third layer is a pooled layer, the input of the third layer is a 48 x 32 node matrix of the second layer output, the filter size of the third layer is 2 x 2, the step size of the length and width is 2, and the output matrix size of the third layer is 24 x 32; the fourth layer is a convolution layer, the input of the fourth layer is the output of the third layer, the filter size of the fourth layer is 5 x 5, the depth is 64, the fourth layer is filled with all 0 s, and the output matrix size of the fourth layer is 24 x 32; the fifth layer is a pooling layer, the input of the fifth layer is a 24 × 64 node matrix of the fourth layer output, the filter size of the fifth layer is 2 × 2, the length and width step lengths are both 2, and the output matrix size of the fifth layer is 12 × 64; the sixth layer is a full connection layer, the input of the sixth layer is the output of the fifth layer, the number of input nodes of the sixth layer is 12 × 64, the number of input nodes of the sixth layer is 9216, and the number of output nodes of the sixth layer is 512; the seventh layer is a full connection layer, the number of input nodes of the seventh layer is 512, and the number of output nodes of the seventh layer is 4.
As shown in fig. 3, the detection method of the detection device for repairing the ancient bridge provided by the invention comprises the following steps:
s101, injecting an ancient bridge repairing raw material into a storage bin, lifting to the position of the ancient bridge through a lifting pipe, collecting an image of the ancient bridge through a camera, and sending the image to a controller for processing;
s102, detecting crack information of the ancient-constructed bridge through a crack detection module by a controller; evaluating the health of the ancient bridge through a health evaluation module; a wireless communication module is used for connecting a wireless network by using a wireless signal to carry out remote control; diagnosing the device fault through a fault diagnosis module;
s103, the controller starts the ejector to spray the repair raw materials to repair the ancient bridge cracks according to the crack information.
As shown in fig. 4, the detection method of the crack detection module 7 provided by the present invention is as follows:
s201, performing crack segmentation on a group of ancient bridge images acquired by a camera through a segmentation program;
s202, detecting and classifying the ancient-built bridge cracks by adopting a pre-constructed ancient-built bridge crack classification model according to the crack segmentation result.
The method for segmenting the cracks of the group of ancient bridge images acquired by the camera through the segmentation program comprises the following steps:
dividing cracks in each ancient bridge image as initial contours by adopting a spectral clustering algorithm based on Nystrom approximation theoretical generalization; sequentially mapping the divided cracks to the next ancient bridge image along the upward direction to serve as the initial contour of the crack in the ancient bridge image, and finishing the division of each crack by adopting an improved GAC model until the division of all the ancient bridge images is finished;
and sequentially mapping the cracks segmented along the upward direction to the next ancient bridge image along the downward direction, taking the cracks as initial contours of the cracks in the ancient bridge image, and completing the segmentation of each crack by adopting an improved GAC model until all the ancient bridge images are segmented.
The method for completing the segmentation of each fracture by adopting the improved GAC model comprises the following steps: calculating the gray level mean value and the gray level standard deviation of each divided crack area, and taking the gray level mean value and the gray level standard deviation as the gray level similarity information of each crack area; constructing a gray level similarity information item according to the gray level similarity information; and adding the gray level similarity information item as an external energy item to an energy functional of the GAC model, thereby improving the GAC model.
The construction method of the ancient bridge crack classification model comprises the following steps: collecting original ancient bridge images containing various cracks; carrying out crack marking and crack segmentation on the collected original ancient bridge image to construct a sample data set of the ancient bridge cracks; initially constructing 8 layers of deep convolutional neural networks by combining the global characteristics of the ancient bridge, wherein each deep convolutional neural network comprises a first input layer, a third convolutional layer, a third pooling layer, a second full-connection layer and an output layer, and a softmax classifier is adopted; and training and testing the constructed deep convolution neural network by adopting the sample data set so as to determine the structure and parameters of the ancient bridge crack classification model.
As shown in fig. 5, the health assessment module 8 provided by the present invention has the following assessment method:
s301, acquiring ancient building bridge health monitoring data based on time sequence through a detection device, and generating a training sample set based on the pre-divided ancient building bridge health state levels;
s302, training by using a training sample set to obtain a neural network model; and inputting newly acquired ancient bridge health monitoring data into the neural network model, and outputting the predicted ancient bridge health state level.
The training sample set provided by the invention comprises a plurality of training samples based on time sequence, wherein each training sample comprises a group of ancient bridge health monitoring data and a corresponding ancient bridge health state grade.
The training process of the neural network model provided by the invention comprises the following steps: and taking the ancient building bridge health monitoring data in the training sample as input, taking the corresponding ancient building bridge health state level as output, training the initialized neural network model, calculating a loss function of the neural network model and updating a parameter matrix until a prediction error meets a termination condition.
The invention provides a training sample set of
Wherein the content of the first and second substances,
represents tkThe training samples at a time of day are,
is a feature vector of n +1 dimensions,
n is more than or equal to 1 for the ancient bridge health monitoring data,
is that
The corresponding health state level of the ancient bridge, and the set of the health state level of the ancient bridge is y e { a ∈1,a2,...,aN}，a1,a2,...,aNAll are different values, N is more than or equal to 2.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications, equivalent changes and modifications made to the above embodiment according to the technical spirit of the present invention are within the scope of the technical solution of the present invention.
Claims (10)
1. The utility model provides a detection device for ancient building bridge restoration which characterized in that, ancient building bridge restoration includes with detection device:
the system comprises a case, a storage bin, a lifting pipe, an ejector, a camera, a controller, a crack detection module, a health evaluation module, a wireless communication module and a fault diagnosis module;
a storage bin is fixed on the left side in the case through screws; the top end of the storage bin is sleeved with a lifting pipe; the top end of the lifting pipe is provided with an ejector; the top end of the ejector is fixed with a camera through a screw; the right side of the front surface of the case is embedded with a controller; the controller is internally provided with a crack detection module, a health evaluation module, a wireless communication module and a fault diagnosis module;
the crack detection module is used for detecting crack information of the ancient bridge;
the health evaluation module is used for evaluating the health of the ancient bridge;
the wireless communication module is used for connecting a wireless network through a wireless signal to carry out remote control;
the fault diagnosis module is used for diagnosing device faults; detecting the repair state of the ancient bridge repair equipment through a repair state sensor to obtain first signal data comprising the repair state value of the ancient bridge repair equipment;
collecting first signal data through a 4-to-6 24-point repair state sensor, wherein the first signal data comprise a left repair state sensor signal of the left repair state of the ancient bridge repair equipment and a right repair state sensor signal of the right repair state of the ancient bridge repair equipment, and the first signal data are classified according to a normal state and a fault state;
includes preprocessing the first signal data, combining a plurality of 4 x 6 signals, and synthesizing a 48 x 48 signal matrix by using 48 sets of signals, wherein one set of signals includes a left repair status sensor signal and a right repair status sensor signal;
the signal prediction model comprises seven layers; the first layer is an input layer, the first layer is original signal data, the picture size of the first layer is 48 × 1, and the channel is 1; the second layer is a convolutional layer, the input of the second layer is the output of the output layer, the filter size of the second layer is 5 x 5, the depth is 32, the second layer is filled with all 0 s, and the output matrix size of the second layer is 48 x 32; the third layer is a pooled layer, the input of the third layer is a 48 x 32 node matrix of the second layer output, the filter size of the third layer is 2 x 2, the step size of the length and width is 2, and the output matrix size of the third layer is 24 x 32; the fourth layer is a convolutional layer, the input to the fourth layer is the output of the third layer, the filter size of the fourth layer is 5 x 5, the depth is 64, the fourth layer is filled with all 0 s, and the output matrix size of the fourth layer is 24 x 32; the fifth layer is a pooling layer, the input of the fifth layer is a 24 x 64 node matrix of the fourth layer output, the filter size of the fifth layer is 2 x 2, the length and width steps are 2, and the output matrix size of the fifth layer is 12 x 64; the sixth layer is a full connection layer, the input of the sixth layer is the output of the fifth layer, the number of input nodes of the sixth layer is 12 × 64, the number of input nodes of the sixth layer is 9216, and the number of output nodes of the sixth layer is 512; the seventh layer is a full connection layer, the number of input nodes of the seventh layer is 512, and the number of output nodes of the seventh layer is 4.
2. The detection method for the detection device for the ancient bridge restoration according to claim 1, characterized by comprising the following steps of:
injecting an ancient bridge repairing raw material into a storage bin, lifting the ancient bridge repairing raw material to the position of the ancient bridge through a lifting pipe, collecting an image of the ancient bridge through a camera, and sending the image to a controller for processing;
secondly, detecting the crack information of the ancient bridge through a crack detection module by the controller; evaluating the health of the ancient bridge through a health evaluation module; a wireless communication module is used for connecting a wireless network by using a wireless signal to carry out remote control; diagnosing the device fault through a fault diagnosis module;
and step three, the controller starts the ejector according to the crack information to spray repair raw materials for repairing the ancient bridge cracks.
3. The detection method for repairing the ancient bridge according to claim 2, wherein the detection method of the crack detection module is as follows:
(1) performing crack segmentation on a group of ancient bridge images acquired by a camera through a segmentation program;
(2) and detecting and classifying the ancient-built bridge cracks by adopting a pre-constructed ancient-built bridge crack classification model according to the crack segmentation result.
4. The detection method for repairing the ancient bridge according to claim 3, wherein the method for performing crack segmentation on the set of ancient bridge images acquired by the camera through the segmentation program comprises the following steps:
dividing cracks in each ancient bridge image as initial contours by adopting a spectral clustering algorithm based on Nystrom approximation theoretical generalization; sequentially mapping the divided cracks to the next ancient bridge image along the upward direction to serve as the initial contour of the crack in the ancient bridge image, and finishing the division of each crack by adopting an improved GAC model until the division of all the ancient bridge images is finished;
and sequentially mapping the cracks segmented along the upward direction to the next ancient bridge image along the downward direction, taking the cracks as initial contours of the cracks in the ancient bridge image, and completing the segmentation of each crack by adopting an improved GAC (generic object model) model until all the ancient bridge images are segmented.
5. The detection method for repairing the ancient bridge according to claim 4, wherein the method for completing the segmentation of each crack by adopting the improved GAC model comprises the following steps: calculating the gray level mean value and the gray level standard deviation of each divided crack area, and taking the gray level mean value and the gray level standard deviation as the gray level similarity information of each crack area; constructing a gray level similarity information item according to the gray level similarity information; and adding the gray level similarity information item as an external energy item to an energy functional of the GAC model, thereby improving the GAC model.
6. The detection method for repairing the ancient bridge according to claim 3, wherein the construction method of the ancient bridge crack classification model comprises the following steps: collecting original ancient bridge images containing various cracks; carrying out crack marking and crack segmentation on the collected original ancient bridge image to construct a sample data set of the ancient bridge cracks; initially constructing an 8-layer deep convolutional neural network by combining the global characteristics of the ancient bridge, wherein the deep convolutional neural network comprises a first input layer, three convolutional layers, three pooling layers, two full-connection layers and an output layer, and a softmax classifier is adopted; and training and testing the constructed deep convolution neural network by adopting the sample data set so as to determine the structure and parameters of the ancient bridge crack classification model.
7. The detection method for repairing the ancient bridge according to claim 1, wherein the health assessment module comprises the following assessment methods:
1) acquiring ancient building bridge health monitoring data based on a time sequence through a detection device, and generating a training sample set based on the health state grade of the ancient building bridge divided in advance;
2) training by using a training sample set to obtain a neural network model; and inputting newly acquired ancient bridge health monitoring data into the neural network model, and outputting the predicted ancient bridge health state level.
8. The method according to claim 7, wherein the training sample set comprises a plurality of time-series-based training samples, and each training sample comprises a set of the ancient bridge health monitoring data and the corresponding ancient bridge health status level.
9. The method for detecting the restoration of the ancient bridge according to claim 7, wherein the training process of the neural network model comprises the following steps: and taking the ancient building bridge health monitoring data in the training sample as input, taking the corresponding ancient building bridge health state level as output, training the initialized neural network model, calculating a loss function of the neural network model and updating a parameter matrix until a prediction error meets a termination condition.
10. The method for detecting the return port of an ancient bridge according to claim 7, wherein the method comprises the steps ofThe training sample set is
Wherein, the first and the second end of the pipe are connected with each other,
denotes tk-the training samples of the time of day,
is a feature vector with the dimension of n +1,
x0＝1，x1，...，xnn is more than or equal to 1 for the ancient bridge health monitoring data,
is that
The corresponding health state levels of the ancient bridges, and the set of the health state levels of the ancient bridges is y epsilon { a1,a2,...,aN}，a1,a2,...,aNAll are different values, N is more than or equal to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210256905.XA CN114775457A (en) | 2022-03-16 | 2022-03-16 | Detection device for repairing ancient bridge and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210256905.XA CN114775457A (en) | 2022-03-16 | 2022-03-16 | Detection device for repairing ancient bridge and detection method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114775457A true CN114775457A (en) | 2022-07-22 |
Family
ID=82425390
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210256905.XA Pending CN114775457A (en) | 2022-03-16 | 2022-03-16 | Detection device for repairing ancient bridge and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114775457A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108629144A (en) * | 2018-06-11 | 2018-10-09 | 湖北交投智能检测股份有限公司 | A kind of bridge health appraisal procedure |
| CN110161047A (en) * | 2019-06-14 | 2019-08-23 | 汕头大学 | A kind of road and bridge Crack Detection and repair integrated robot |
| CN111056395A (en) * | 2019-12-25 | 2020-04-24 | 武汉科技大学 | Band-type brake fault diagnosis method based on multipoint pressure sensor |
| CN111222189A (en) * | 2019-12-31 | 2020-06-02 | 湖南工程学院 | Efficient bridge structure health early warning control system and method |
| CN111402227A (en) * | 2020-03-13 | 2020-07-10 | 河海大学常州校区 | Bridge crack detection method |
| CN112627023A (en) * | 2020-11-23 | 2021-04-09 | 山东奥邦交通设施工程有限公司 | Intelligent bridge detection method and system and intelligent bridge detection robot |
| CN112726432A (en) * | 2020-12-29 | 2021-04-30 | 安徽建筑大学 | Bridge operation and maintenance method, device, system, computer equipment and storage medium |
-
2022
- 2022-03-16 CN CN202210256905.XA patent/CN114775457A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108629144A (en) * | 2018-06-11 | 2018-10-09 | 湖北交投智能检测股份有限公司 | A kind of bridge health appraisal procedure |
| CN110161047A (en) * | 2019-06-14 | 2019-08-23 | 汕头大学 | A kind of road and bridge Crack Detection and repair integrated robot |
| CN111056395A (en) * | 2019-12-25 | 2020-04-24 | 武汉科技大学 | Band-type brake fault diagnosis method based on multipoint pressure sensor |
| CN111222189A (en) * | 2019-12-31 | 2020-06-02 | 湖南工程学院 | Efficient bridge structure health early warning control system and method |
| CN111402227A (en) * | 2020-03-13 | 2020-07-10 | 河海大学常州校区 | Bridge crack detection method |
| CN112627023A (en) * | 2020-11-23 | 2021-04-09 | 山东奥邦交通设施工程有限公司 | Intelligent bridge detection method and system and intelligent bridge detection robot |
| CN112726432A (en) * | 2020-12-29 | 2021-04-30 | 安徽建筑大学 | Bridge operation and maintenance method, device, system, computer equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 邹超英, 哈尔滨工业大学出版社 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sony et al. | A systematic review of convolutional neural network-based structural condition assessment techniques | |
| CN109272123B (en) | Sucker-rod pump working condition early warning method based on convolution-circulation neural network | |
| Bai et al. | An optimized railway fastener detection method based on modified Faster R-CNN | |
| CN108257114A (en) | A kind of transmission facility defect inspection method based on deep learning | |
| Yang et al. | Deep learning‐based bolt loosening detection for wind turbine towers | |
| CN111241994B (en) | Deep learning remote sensing image rural highway sanded road section extraction method | |
| CN114548278A (en) | In-service tunnel lining structure defect identification method and system based on deep learning | |
| CN111368702A (en) | Composite insulator hydrophobicity grade identification method based on YOLOv3 network | |
| CN108154498A (en) | A kind of rift defect detecting system and its implementation | |
| Guldur et al. | Condition assessment of bridges using terrestrial laser scanners | |
| CN109635879B (en) | Coal mining machine fault diagnosis system with optimal parameters | |
| CN114705689A (en) | Unmanned aerial vehicle-based method and system for detecting cracks of outer vertical face of building | |
| CN114066808A (en) | Pavement defect detection method and system based on deep learning | |
| CN113379737A (en) | Intelligent pipeline defect detection method based on image processing and deep learning and application | |
| CN115841466A (en) | Automatic quantitative assessment method for defects of drainage pipe network | |
| CN114818826A (en) | Fault diagnosis method based on lightweight Vision Transformer module | |
| Sarkar et al. | Revolutionizing concrete analysis: An in-depth survey of AI-powered insights with image-centric approaches on comprehensive quality control, advanced crack detection and concrete property exploration | |
| Abedi et al. | Infrastructure damage assessment via machine learning approaches: a systematic review | |
| CN114775457A (en) | Detection device for repairing ancient bridge and detection method thereof | |
| CN116861361A (en) | Dam deformation evaluation method based on image-text multi-mode fusion | |
| CN110765900A (en) | DSSD-based automatic illegal building detection method and system | |
| CN115984286A (en) | Arch bridge point cloud segmentation method based on synthetic simulator and bridged neural network | |
| Li et al. | A real-time multi-defect automatic identification framework for concrete dams via improved YOLOv5 and knowledge distillation | |
| CN116030292A (en) | Concrete surface roughness detection method based on improved ResNext | |
| CN113486599B (en) | Method for calculating effective stroke of oil pumping unit |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220722 |
|
| RJ01 | Rejection of invention patent application after publication |