CN115423995A - Lightweight curtain wall crack target detection method and system and safety early warning system - Google Patents
Lightweight curtain wall crack target detection method and system and safety early warning system Download PDFInfo
- Publication number
- CN115423995A CN115423995A CN202210959902.2A CN202210959902A CN115423995A CN 115423995 A CN115423995 A CN 115423995A CN 202210959902 A CN202210959902 A CN 202210959902A CN 115423995 A CN115423995 A CN 115423995A
- Authority
- CN
- China
- Prior art keywords
- crack
- target detection
- curtain wall
- network
- convolution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a lightweight curtain wall crack target detection method and system and a safety early warning system, wherein the target detection method comprises the steps of image collection, image preprocessing, model construction, model training and testing and crack detection, a large number of wall images are collected through the image collection step and preprocessed, then a lightweight curtain wall crack target detection model is constructed, the Ghost convolution is used for replacing the standard convolution in the original network, and the convolution in the C3 structure is replaced by the Ghost convolution, so that the calculated amount can be reduced, and the calculation efficiency is improved; a CA attention mechanism is added at the position where the number of main features extraction network channels is the largest, so that the network pays more attention to valuable features, and the target detection precision of the whole network is improved. The trained model is obtained through the training and testing of the model, and finally the trained model is used for crack detection.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a lightweight curtain wall crack target detection method and system and a safety early warning system.
Background
The target detection is a challenging field in computer vision, the achievement of the target detection is widely applied to many fields, and the target detection has many practical applications in the aspects of wall crack detection or wall damage detection and the like. The detection of cracks on walls is a constant problem, which is more obvious for the present higher and higher buildings, and the cracks of the high-rise buildings have great hidden damage to the buildings and are not easily observed by naked eyes. However, such cracks are widespread in nature and are not intended to be detected using conventional methods with significant limitations and costs.
The YOLO series algorithm is updated and iterated for many years, the generation of a candidate frame is omitted in the target detection algorithm of the primary target type, and the method has the advantage of certain detection speed compared with the secondary target type algorithm. However, due to the updating of the algorithm, the detection precision of the YOLO series target detection algorithm is more and more emphasized, so that the network model is more and more complex, and the original YOLO model is not suitable to be carried on a common computing unit any more.
Therefore, the method in the prior art has the technical problem of low detection precision.
Disclosure of Invention
The invention provides a light curtain wall crack target detection method and system and a safety early warning system, which are used for solving or at least partially solving the technical problem of low detection precision in the prior art.
In order to solve the technical problem, a first aspect of the present invention provides a method for detecting a crack target of a lightweight curtain wall, including:
collecting a large number of wall images;
preprocessing the collected wall body image;
constructing a lightweight curtain wall crack target detection model, wherein the model adopts an improved target detection network, the improved target detection network comprises a trunk feature extraction network, an enhanced feature extraction network and a detection head, the trunk feature extraction network uses a Ghost convolution to replace a standard convolution in an original YOLOv5 network, the trunk feature extraction network and the enhanced feature network replace a convolution in a C3 structure of the original YOLOv5 network with a Ghost convolution, and a CA attention module is added at the position with the largest number of main feature extraction network channels to pay attention to important features related to wall cracks;
training and testing the constructed lightweight curtain wall crack target detection model to obtain a trained lightweight curtain wall crack target detection model;
and carrying out crack detection on the image to be detected by using the trained lightweight curtain wall crack target detection model.
In one embodiment, pre-processing the collected wall images includes:
cleaning and sorting the collected wall body images, and uniformly converting the images into PNG or JPG formats;
and numbering and sequencing according to the physical positions of the images on the wall.
In one embodiment, the processing of the CA attention module includes:
performing average pooling by using pooling kernels along two vertical and horizontal directions to obtain a pair of one-dimensional feature codes along a horizontal coordinate and a vertical coordinate, wherein the pair of feature codes are feature representations with global receptive field and direction perception;
performing feature fusion on a pair of one-dimensional feature codes, then performing convolution transformation and nonlinear activation function processing, then changing the fused features into two independent tensor sums along the space direction again, and performing convolution again on the two tensor sums;
and carrying out normalization processing on the tensor obtained by the convolution again and the sigmoid activation function, and outputting the tensor to the next module in the trunk feature extraction network to be used as a weight for guiding the network to learn more important features in the channel.
In one embodiment, training a constructed lightweight curtain wall crack target detection model comprises:
dividing the preprocessed wall body image into a training set, a verification set and a test set according to a preset proportion;
normalizing the divided training set into images with consistent resolution;
inputting the training set image after the normalization processing into a lightweight curtain wall crack target detection model, generating an additional feature by Ghost convolution, paying attention to the important feature by a CA attention mechanism, and obtaining an output image feature according to the additional feature and the important feature concerned by the CA attention mechanism;
performing regression prediction and positioning on the verification set according to the output image characteristics, verifying training and convergence conditions, and obtaining a detection result, a crack degree, a recall rate and target detection precision of cracks in the image in the training process;
and testing the trained lightweight curtain wall crack target detection model on a test set, obtaining the position of a crack of the image with the crack found in the test process according to the serial number of the wall body image, and taking the model with the final target detection precision larger than the threshold value as the trained lightweight curtain wall crack target detection model.
In one embodiment, the method for detecting the cracks of the image to be detected by using the trained lightweight curtain wall crack target detection model comprises the following steps:
preprocessing an image to be detected, and inputting the preprocessed image into a trained lightweight curtain wall crack target detection model to obtain a crack detection result, wherein the crack detection result uses anchor frame visual crack marking;
and distinguishing the crack grade and the crack shape according to the crack degree, and marking the crack grade and the crack shape beside the anchor frame.
In one embodiment, the method further comprises: and alarming according to the crack grade.
Based on the same inventive concept, the second aspect of the present invention provides a light curtain wall crack target detection system, comprising:
the image collection module is used for collecting a large number of wall images;
the preprocessing module is used for preprocessing the collected wall body images;
the model construction module is used for constructing a lightweight curtain wall crack target detection model, the model adopts an improved target detection network, the improved target detection network comprises a trunk feature extraction network, a reinforced feature extraction network and a detection head, wherein the trunk feature extraction network uses a Ghost convolution to replace a standard convolution in an original YOLOv5 network, the trunk feature extraction network and the reinforced feature network replace a convolution in a C3 structure of the original YOLOv5 network with the Ghost convolution, and a CA attention module is added at the position with the largest number of main feature extraction network channels and used for paying attention to important features related to wall cracks;
the training and testing module is used for training and testing the constructed lightweight curtain wall crack target detection model to obtain a trained lightweight curtain wall crack target detection model;
and the crack detection module is used for carrying out crack detection on the image to be detected by utilizing the trained lightweight curtain wall crack target detection model.
Based on the same inventive concept, the third aspect of the invention provides a safety alarm system, which comprises the lightweight curtain wall crack target detection system and an alarm module, wherein the alarm module is used for giving an alarm according to the crack grade.
Based on the same inventive concept, a fourth aspect of the present invention provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of the first aspect when executing the program.
Compared with the prior art, the invention has the advantages and beneficial technical effects as follows:
the invention provides a lightweight curtain wall crack target detection method which comprises the steps of image collection, image preprocessing, model construction, model training and testing and crack detection, wherein the constructed lightweight curtain wall crack target detection model improves a YOLOv5 network, ghost convolution is used for replacing standard convolution in an original network, convolution in a C3 structure is replaced by Ghost convolution, and C3 is replaced by Ghost C3, so that the calculated amount can be reduced, and the calculation efficiency is improved; a CA attention mechanism is added at the position where the number of main features extraction network channels is the largest, so that the network pays more attention to valuable features, and the target detection precision of the whole network is improved. In other words, the invention uses the lightweight target detection model, does not need a large number of calculation units, can detect the wall cracks in real time and find the positions of the cracks, and improves the detection precision while ensuring the efficiency.
Further, the invention also provides a safety alarm system, and different levels of alarms are carried out according to the grade of the cracks through the alarm module.
Drawings
In order to more clearly illustrate the embodiments or technical solutions of the present invention, the drawings used in the embodiments or technical solutions in the prior art are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for detecting a crack target of a lightweight curtain wall, which is disclosed by the embodiment of the invention;
FIG. 2 is a schematic diagram illustrating a structure diagram of a crack target detection model (GC-YOLOv 5 network) of a lightweight curtain wall in an embodiment of the invention in comparison with a structure diagram of a basic YOLOv5s network;
FIG. 3 is a schematic diagram of a Ghost convolution operation mode in the embodiment of the present invention;
FIG. 4 is a diagram illustrating a CA attention mechanism according to an embodiment of the present invention;
fig. 5 is a flow chart of implementing a safety alarm by using a lightweight curtain wall crack target detection model in the embodiment of the invention.
Detailed Description
The invention discloses a method for detecting a crack target of a lightweight curtain wall, which comprises the steps of image collection, image preprocessing, model construction, model training and testing and crack detection. The method comprises the steps of image collection, image preprocessing, image number obtaining, data cleaning and format conversion, light curtain wall crack target detection model construction, model deployment, model training and testing, and detection and analysis of targets, wherein the steps of image collection include that wall body pictures are shot through tools such as an unmanned aerial vehicle, image preprocessing is carried out to obtain picture numbers and carry out data cleaning and format conversion, the model construction step constructs a light curtain wall crack target detection model, the model can be conveniently deployed on equipment such as the unmanned aerial vehicle, the detection and analysis of the targets are realized by combining computing units (CPU, GPU and the like) with limited computing capacity, which are carried on the tools such as the unmanned aerial vehicle, the model training and testing step trains and tests the model to obtain the model with higher detection precision, the step of crack classification detection is to detect the wall body to be detected by using the trained model, and classification is carried out according to the shape and the grade of cracks, and an anchor frame is marked. The method is based on the target detection technology to identify the wall cracks, and has the advantages that a lightweight target detection model is used, a large number of calculation units are not needed, the wall cracks can be detected in real time, the positions of the cracks can be found, and the detection precision is improved while the detection efficiency is ensured.
Furthermore, the invention also discloses a safety alarm system which sends out alarms of different levels according to the levels of the anchor frames marked in the crack classification detection step so as to ensure that the cracks are processed.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
The embodiment of the invention provides a lightweight curtain wall crack target detection method, which comprises the following steps:
collecting a large number of wall images;
preprocessing the collected wall body image;
constructing a lightweight curtain wall crack target detection model, wherein the model adopts an improved target detection network, the improved target detection network comprises a trunk feature extraction network, an enhanced feature extraction network and a detection head, the trunk feature extraction network uses a Ghost convolution to replace a standard convolution in an original YOLOv5 network, the trunk feature extraction network and the enhanced feature network replace a convolution in a C3 structure of the original YOLOv5 network with a Ghost convolution, and a CA attention module is added at the position with the largest number of main feature extraction network channels to pay attention to important features related to wall cracks;
training and testing the constructed lightweight curtain wall crack target detection model to obtain a trained lightweight curtain wall crack target detection model;
and carrying out crack detection on the image to be detected by using the trained lightweight curtain wall crack target detection model.
Fig. 1 is a schematic flow chart of a method for detecting a crack target of a lightweight curtain wall according to an embodiment of the present invention.
In the specific implementation process, the high-definition images of the wall body can be acquired by using equipment such as an unmanned aerial vehicle, the subsequent training process is divided into images for training and detection images, the images for training aim at training the network model to acquire better detection precision, and the positions of cracks of the images for training are marked to help a target detection network model to learn. The detection image is used for actual detection and is directly processed by a target detection network.
The preprocessing of the wall body image mainly comprises the steps of cleaning, format conversion and numbering of the image so as to facilitate subsequent processing, and determining the position of a crack according to the image number.
The improved target detection network improves the original target detection network yoolov 5, please refer to fig. 2, which is a schematic diagram of a comparison between a structure diagram of a lightweight curtain wall crack target detection model (GC-yoolov 5 network) adopted in the present invention and a structure diagram of a basic yoolov 5s network.
The improvement mainly comprises the following parts:
firstly, the method comprises the following steps: replacing standard convolution in an original network by using Ghostconvolution, wherein a Conv structure in YOLOv5 is a basic convolution module and consists of a common convolution layer, a batch normalization layer and an activation function, namely replacing the Conv structure with the GhostConv structure;
II, secondly: in a network model, a C3 structure comprises a residual error structure and three convolutions, which can play a role in reducing parameters, the convolution in the C3 structure is replaced by a Ghost convolution, namely C3 is replaced by Ghost C3;
thirdly, the method comprises the following steps: a CA attention mechanism is added at the position where the number of main features extraction network channels is the largest, so that the network pays more attention to valuable features, and the target detection precision of the whole network is improved.
Fig. 3 is a schematic diagram of the method of the Ghost convolution according to the embodiment of the present invention.
The Ghost convolution has the function of generating some redundant features by using linear operation with relatively low calculation cost, and replacing part of features generated by standard convolution with the redundant features, wherein the redundant features can greatly improve the detection accuracy although the consumed calculation amount is small, and the operation mode of the Ghost convolution is as shown in fig. 3. Where n is the number of convolutions used and s is a coefficient, such thatIs to use the proportion of the conventional convolution in the Ghost convolution, andis the ratio using linear operation, w 'and h' are the length and width of the output feature graph, c is the number of channels of the output, k and d are the kernel length and width of the conventional convolution and look-ahead operation, respectively, and k × k and d × d are used because the kernel length and width are generally the same, and here k and d are generally made equal due to the parallelism of the device operation.
According to the working mode of the Ghost convolution, compared with the corresponding standard convolution, the speed-up ratio is as follows:
generally, the invention detects wall cracks in real time through the improved YOLO network model GC-YOLOv5, can detect whether the wall cracks appear, and can obtain crack positions and visually mark the grade and the shape of the cracks simultaneously because the wall images are sequenced. The light GC-YOLOv5 target detection network can be carried on common tools such as an unmanned aerial vehicle and the like to operate, real-time detection can be guaranteed, meanwhile, the safety of detection personnel is guaranteed, and manpower is saved.
In one embodiment, preprocessing the collected wall images includes:
cleaning and sorting the collected wall body images, and uniformly converting the images into PNG or JPG formats;
and numbering and sequencing according to the physical positions of the images on the wall.
In the specific implementation process, the images which cannot be distinguished are retaken, and the images are also segmented. The images are numbered and ordered by column or row.
In one embodiment, the processing of the CA attention module includes:
performing average pooling by using pooling kernels along two vertical and horizontal directions to obtain a pair of one-dimensional feature codes along a horizontal coordinate and a vertical coordinate, wherein the pair of feature codes are feature representations with global receptive field and direction perception;
performing feature fusion on a pair of one-dimensional feature codes, then performing convolution transformation and nonlinear activation function processing, then changing the fused features into two independent tensor sums along the space direction again, and performing convolution again on the two tensor sums;
and carrying out normalization processing on the tensor obtained by the convolution again and the sigmoid activation function, and outputting the tensor to the next module in the trunk feature extraction network to be used as a weight for guiding the network to learn more important features in the channel.
Specifically, the CA attention module is used to acquire the importance of each feature channel from both the horizontal and vertical spatial directions, and in this process, a spatial relationship for the entire image can be obtained, and this spatial relationship can help the network to pay attention to valuable spatial information better. The overall steps of the operation of the CA attention module are shown in FIG. 4.
The convolution neural network model has convolution as main modules, and has an SPPF structure including serial convolutions with the same three convolution kernels except introduced Conv standard convolution and C3 residual convolution modules, so as to fuse more features with different resolutions and obtain more information.
Different from a general improved model in the prior art, the method not only uses the GhostConv and GhostC3 modules in the trunk feature extraction network, but also changes the corresponding modules in the subsequent reinforced feature extraction network into the GhostConv and GhostC3 modules. In addition, as more models replace GhostConv and GhostC3 modules, and partial characteristics are lost due to the calculation mode, so that the final precision is slightly reduced, the CA attention module is added, the module adopts an attention mechanism to enable the network to pay more attention to the original high-value characteristics, the weight of the redundant characteristics generated by the GhostConv and GhostC3 modules is reduced, and the combination of the modules enables the models to be highly light-weighted and simultaneously maintain high detection precision.
Specifically, the method uses the Ghost convolution to replace the standard convolution in the original network, the Ghost convolution has the effects that a plurality of redundant features are generated by using linear operation with low calculation cost, partial features generated by the standard convolution are replaced by the redundant features, the redundant features consume a small amount of calculation, but the detection precision can be greatly improved, so that when the method is used for identifying the wall cracks, a large number of calculation units are not needed, the wall cracks can be detected in real time, the positions of the cracks can be found, the serious grades of the cracks can be distinguished, and then alarms of different grades can be given.
Furthermore, the invention adds a CA attention mechanism at the position where the number of main feature extraction network channels is the largest, so that the network pays more attention to valuable features, and the target detection precision of the whole network is improved. This attention mechanism effectively improves the target detection accuracy which inevitably decreases because of the use of Ghost convolution. Under the combination of the two methods, the target detection network is simplified and lightened, and the detection precision is kept.
In one embodiment, training a constructed lightweight curtain wall crack target detection model comprises:
dividing the preprocessed wall body image into a training set, a verification set and a test set according to a preset proportion;
carrying out normalization processing on the divided training sets to obtain images with consistent resolution;
inputting the training set image after normalization processing into a lightweight curtain wall crack target detection model, generating additional features by Ghost convolution, paying attention to the important features by a CA attention mechanism, and obtaining output image features according to the additional features and the important features paid attention by the CA attention mechanism;
performing regression prediction and positioning on the verification set according to the output image characteristics, verifying training and convergence conditions, and obtaining a detection result, a crack degree, a recall rate and target detection precision of cracks in the image in the training process;
the trained lightweight curtain wall crack target detection model is tested on a test set, the positions of cracks of the images with the cracks found in the test process are obtained according to the serial numbers of the wall images, and the model with the final target detection precision larger than the threshold value is used as the trained lightweight curtain wall crack target detection model.
In a specific implementation process, the preset ratio can be set in an actual situation, for example, the ratio of 8. The threshold value may also be set according to the actual situation, e.g. 70%, 80%, etc.
In addition, when the target detection precision is less than 80%, the data set needs to be expanded, and the data set division step is returned to for training again so as to ensure the detection precision.
In one embodiment, the method for detecting the cracks of the image to be detected by using the trained lightweight curtain wall crack target detection model comprises the following steps:
preprocessing an image to be detected, and inputting the preprocessed image into a trained lightweight curtain wall crack target detection model to obtain a crack detection result, wherein the crack detection result is used for visually marking cracks by using an anchor frame;
and distinguishing the crack grade and the crack shape according to the crack degree, and marking the crack grade and the crack shape beside the anchor frame.
Specifically, cracks are classified into three grades according to the cracking program: mild, moderate, and severe. The visual result is used for visually finding the crack information during manual inspection.
In one embodiment, the method further comprises: and alarming according to the crack grade.
Specifically, when the crack is found in the crack detection step, different levels of alarms are issued according to the crack levels, mild and moderate cracks can indicate the crack levels and crack shapes in the submitted log, and the specific positions of the wall where the crack appears are indicated in the log according to the numbers of the wall when the crack is segmented in the image processing. And (4) directly giving an alarm for the severe cracks, and suspending subsequent work of the unmanned aerial vehicle carrying the light-weight target detection network and continuously detecting the wall until the alarm is eliminated.
Specifically, please refer to fig. 5, which is a flowchart illustrating an implementation of a safety alarm implemented by using a lightweight curtain wall crack target detection model according to an embodiment of the present invention.
According to the method, a CG-YOLOv5 network which is lighter than original YOLOv5s is used, the original convolution is replaced by the Ghost convolution, so that the calculated amount is greatly reduced, partial redundancy features are generated by the Ghost convolution through cheaper linear operation, and the redundancy features generated by original standard convolution are filled up. The use of redundant features necessarily reduces some of the detection accuracy, so inserting a CA attention mechanism makes the network more concerned about important features. The CG-YOLOv5 network using the two methods can greatly reduce the complexity of the network on the premise of keeping the detection precision, thereby reducing the requirement on hardware.
Example two
Based on the same inventive concept, the embodiment provides a light curtain wall crack target detection system, which comprises:
the image collection module is used for collecting a large number of wall images;
the preprocessing module is used for preprocessing the collected wall body images;
the model construction module is used for constructing a lightweight curtain wall crack target detection model, the model adopts an improved target detection network, the improved target detection network comprises a trunk feature extraction network, a reinforced feature extraction network and a detection head, wherein the trunk feature extraction network uses a Ghost convolution to replace a standard convolution in an original YOLOv5 network, the trunk feature extraction network and the reinforced feature network replace a convolution in a C3 structure of the original YOLOv5 network with the Ghost convolution, and a CA attention module is added at the position with the largest number of main feature extraction network channels and used for paying attention to important features related to wall cracks;
the training and testing module is used for training and testing the constructed lightweight curtain wall crack target detection model to obtain a trained lightweight curtain wall crack target detection model;
the crack detection module is used for carrying out crack detection on an image to be detected by utilizing the trained lightweight curtain wall crack target detection model.
Since the system described in the second embodiment of the present invention is a system adopted for implementing the method for detecting the crack target of the lightweight curtain wall in the first embodiment of the present invention, based on the method described in the first embodiment of the present invention, a person skilled in the art can understand the specific structure and deformation of the system, and thus, details are not described herein. All systems adopted by the method in the first embodiment of the invention belong to the protection scope of the invention.
EXAMPLE III
Based on the same conception, the invention also provides a safety alarm system, which comprises the lightweight curtain wall crack target detection system and an alarm module, wherein the alarm module is used for giving an alarm according to the crack grade.
Since the system described in the third embodiment of the present invention is implemented based on the system described in the second embodiment of the present invention, the specific structure and the variations of the safety alarm system can be understood by those skilled in the art based on the system described in the second embodiment of the present invention, and thus, the details thereof are not described herein. Example four
Based on the same inventive concept, the present application further provides a computer device, which includes a storage, a processor, and a computer program stored in the storage and executable on the processor, and when the processor executes the computer program, the method in the first embodiment is implemented.
Since the computer device introduced in the fourth embodiment of the present invention is a computer device used for implementing the method for detecting a crack target of a lightweight curtain wall in the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, a person skilled in the art can understand the specific structure and deformation of the computer device, and thus, no further description is given here. All the computer devices used in the method in the first embodiment of the present invention are within the scope of the present invention.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made in the embodiments of the present invention without departing from the spirit or scope of the embodiments of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to encompass such modifications and variations.
Claims (9)
1. A lightweight curtain wall crack target detection method is characterized by comprising the following steps:
collecting a large number of wall images;
preprocessing the collected wall body image;
constructing a lightweight curtain wall crack target detection model, wherein the model adopts an improved target detection network, the improved target detection network comprises a trunk feature extraction network, an enhanced feature extraction network and a detection head, the trunk feature extraction network uses Ghost convolution to replace standard convolution in the original YOLOv5 network, the trunk feature extraction network and the enhanced feature network replace convolution in the C3 structure of the original YOLOv5 network with Ghost convolution, and a CA attention module is added at the position with the maximum trunk feature extraction network channel number to pay attention to important features related to wall cracks;
training and testing the constructed lightweight curtain wall crack target detection model to obtain a trained lightweight curtain wall crack target detection model;
and carrying out crack detection on the image to be detected by using the trained lightweight curtain wall crack target detection model.
2. The method for detecting the crack target of the light-weight curtain wall as claimed in claim 1, wherein the preprocessing of the collected wall body image comprises:
cleaning and sorting the collected wall body images, and uniformly converting the images into PNG or JPG formats;
and numbering and sequencing according to the physical positions of the images on the wall.
3. The method for detecting the crack target of the light-weight curtain wall as claimed in claim 1, wherein the CA attention module comprises:
performing average pooling by using pooling kernels along two vertical and horizontal directions to obtain a pair of one-dimensional feature codes along a horizontal coordinate and a vertical coordinate, wherein the pair of feature codes are feature representations with global receptive field and direction perception;
performing feature fusion on a pair of one-dimensional feature codes, then performing convolution transformation and nonlinear activation function processing, then changing the fused features into two independent tensor sums along the space direction again, and performing convolution again on the two tensor sums;
and carrying out normalization processing on the tensor obtained by the convolution again and the sigmoid activation function, and outputting the tensor to the next module in the trunk feature extraction network to be used as a weight for guiding the network to learn more important features in the channel.
4. The method for detecting the crack target of the lightweight curtain wall as claimed in claim 2, wherein training the constructed lightweight curtain wall crack target detection model comprises:
dividing the preprocessed wall body image into a training set, a verification set and a test set according to a preset proportion;
carrying out normalization processing on the divided training sets to obtain images with consistent resolution;
inputting the training set image after normalization processing into a lightweight curtain wall crack target detection model, generating additional features by Ghost convolution, paying attention to the important features by a CA attention mechanism, and obtaining output image features according to the additional features and the important features paid attention by the CA attention mechanism;
performing regression prediction and positioning on the verification set according to the output image characteristics, verifying training and convergence conditions, and obtaining a detection result, a crack degree, a recall rate and target detection precision of cracks in the image in the training process;
and testing the trained lightweight curtain wall crack target detection model on a test set, obtaining the position of a crack of the image with the crack found in the test process according to the serial number of the wall body image, and taking the model with the final target detection precision larger than the threshold value as the trained lightweight curtain wall crack target detection model.
5. The method for detecting the crack target of the light-weight curtain wall as claimed in claim 1, wherein the step of performing crack detection on the image to be detected by using the trained light-weight curtain wall crack target detection model comprises the following steps:
preprocessing an image to be detected, and inputting the preprocessed image into a trained lightweight curtain wall crack target detection model to obtain a crack detection result, wherein the crack detection result uses anchor frame visual crack marking;
and distinguishing the crack grade and the crack shape according to the crack degree, and marking the crack grade and the crack shape beside the anchor frame.
6. The method for detecting the crack target of the light-weight curtain wall as claimed in claim 5, wherein the method further comprises: and alarming according to the crack grade.
7. The utility model provides a light-weight curtain wall crack target detecting system which characterized in that includes:
the image collection module is used for collecting a large number of wall images;
the preprocessing module is used for preprocessing the collected wall body images;
the model construction module is used for constructing a lightweight curtain wall crack target detection model, the model adopts an improved target detection network, the improved target detection network comprises a trunk feature extraction network, a reinforced feature extraction network and a detection head, wherein the trunk feature extraction network uses a Ghost convolution to replace a standard convolution in an original YOLOv5 network, the trunk feature extraction network and the reinforced feature network replace a convolution in a C3 structure of the original YOLOv5 network with the Ghost convolution, and a CA attention module is added at the position with the largest number of main feature extraction network channels and used for paying attention to important features related to wall cracks;
the training and testing module is used for training and testing the constructed lightweight curtain wall crack target detection model to obtain a trained lightweight curtain wall crack target detection model;
and the crack detection module is used for carrying out crack detection on the image to be detected by utilizing the trained lightweight curtain wall crack target detection model.
8. A safety warning system, characterized by comprising the light-weight curtain wall crack target detection system as claimed in claim 7 and a warning module, wherein the warning module is used for giving a warning according to the crack grade.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 6 when executing the program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210959902.2A CN115423995A (en) | 2022-08-11 | 2022-08-11 | Lightweight curtain wall crack target detection method and system and safety early warning system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210959902.2A CN115423995A (en) | 2022-08-11 | 2022-08-11 | Lightweight curtain wall crack target detection method and system and safety early warning system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115423995A true CN115423995A (en) | 2022-12-02 |
Family
ID=84198210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210959902.2A Pending CN115423995A (en) | 2022-08-11 | 2022-08-11 | Lightweight curtain wall crack target detection method and system and safety early warning system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115423995A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116502810A (en) * | 2023-06-28 | 2023-07-28 | 威胜信息技术股份有限公司 | Standardized production monitoring method based on image recognition |
CN117113010A (en) * | 2023-10-24 | 2023-11-24 | 北京化工大学 | Power transmission channel safety monitoring method and system based on convolutional network lightweight |
-
2022
- 2022-08-11 CN CN202210959902.2A patent/CN115423995A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116502810A (en) * | 2023-06-28 | 2023-07-28 | 威胜信息技术股份有限公司 | Standardized production monitoring method based on image recognition |
CN116502810B (en) * | 2023-06-28 | 2023-11-03 | 威胜信息技术股份有限公司 | Standardized production monitoring method based on image recognition |
CN117113010A (en) * | 2023-10-24 | 2023-11-24 | 北京化工大学 | Power transmission channel safety monitoring method and system based on convolutional network lightweight |
CN117113010B (en) * | 2023-10-24 | 2024-02-09 | 北京化工大学 | Power transmission channel safety monitoring method and system based on convolutional network lightweight |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113705478B (en) | Mangrove single wood target detection method based on improved YOLOv5 | |
CN115423995A (en) | Lightweight curtain wall crack target detection method and system and safety early warning system | |
CN108038846A (en) | Transmission line equipment image defect detection method and system based on multilayer convolutional neural networks | |
CN114283469B (en) | Improved YOLOv4-tiny target detection method and system | |
CN110992349A (en) | Underground pipeline abnormity automatic positioning and identification method based on deep learning | |
CN113920107A (en) | Insulator damage detection method based on improved yolov5 algorithm | |
CN109506628A (en) | Object distance measuring method under a kind of truck environment based on deep learning | |
CN115457395A (en) | Lightweight remote sensing target detection method based on channel attention and multi-scale feature fusion | |
CN113971660A (en) | Computer vision method for bridge health diagnosis and intelligent camera system | |
CN116824335A (en) | YOLOv5 improved algorithm-based fire disaster early warning method and system | |
CN116665148A (en) | Marine ship detection method based on synthetic aperture radar data | |
CN116129135A (en) | Tower crane safety early warning method based on small target visual identification and virtual entity mapping | |
CN117726991B (en) | High-altitude hanging basket safety belt detection method and terminal | |
CN115512387A (en) | Construction site safety helmet wearing detection method based on improved YOLOV5 model | |
CN114708518A (en) | Bolt defect detection method based on semi-supervised learning and priori knowledge embedding strategy | |
CN116524356A (en) | Ore image small sample target detection method and system | |
CN111339967A (en) | Pedestrian detection method based on multi-view graph convolution network | |
CN115546223A (en) | Method and system for detecting loss of fastening bolt of equipment under train | |
CN116309536A (en) | Pavement crack detection method and storage medium | |
CN114881998A (en) | Workpiece surface defect detection method and system based on deep learning | |
CN114662605A (en) | Flame detection method based on improved YOLOv5 model | |
CN111179278B (en) | Image detection method, device, equipment and storage medium | |
CN116863271A (en) | Lightweight infrared flame detection method based on improved YOLO V5 | |
CN115797970A (en) | Dense pedestrian target detection method and system based on YOLOv5 model | |
CN115410102A (en) | SAR image airplane target detection method based on combined attention mechanism |
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 |