CN115359003A - Two-step tunnel gray image crack identification method, system, medium and equipment - Google Patents
Two-step tunnel gray image crack identification method, system, medium and equipment Download PDFInfo
- Publication number
- CN115359003A CN115359003A CN202211013049.1A CN202211013049A CN115359003A CN 115359003 A CN115359003 A CN 115359003A CN 202211013049 A CN202211013049 A CN 202211013049A CN 115359003 A CN115359003 A CN 115359003A
- Authority
- CN
- China
- Prior art keywords
- tunnel
- crack
- image
- images
- gray
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the technical field of crack identification of rock-soil structures, and provides a two-step type tunnel gray image crack identification method, a system, a medium and equipment. The two-step tunnel gray level image crack identification method comprises the steps of respectively carrying out edge extraction on a batch of tunnel gray level images to obtain corresponding tunnel edge images, and screening out all suspected images with cracks possibly based on comparison between average gray levels of the edge images and a gray level threshold value; obtaining a crack segmentation image of the batch tunnel gray level image based on the suspected image and the fine segmentation network model, and obtaining a crack identification result by counting the crack segmentation image; the crack identification result comprises the number of images of actually existing cracks, the number of cracks and characteristic parameters of each crack.
Description
Technical Field
The invention belongs to the technical field of crack identification of rock-soil structures, and particularly relates to a two-step type tunnel gray image crack identification method, a two-step type tunnel gray image crack identification system, a two-step type tunnel gray image crack identification medium and two-step type tunnel gray image crack identification equipment.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Cracks are common diseases in tunnel structures and are key factors influencing the safety of the tunnel structures. The traditional detection method depends on manual judgment and auxiliary tool judgment; the detection efficiency is low, the working intensity is high, the requirement on professional knowledge of personnel is high, therefore, a rapid detection means is needed, and the current common method is to identify diseases by means of an image processing technology.
Common methods for crack identification are mainly classified into two categories: a conventional image processing method and a crack recognition method based on machine learning. The traditional image processing method comprises the steps of image segmentation, framework extension, a crack width transformation algorithm, automatic detection of pixel level cracks by multi-scale neighborhood information, state evaluation of the structure by three-dimensional scene reconstruction and the like; the crack identification method based on machine learning comprises a space tuning robust multi-feature classifier, automatic detection of road cracks is achieved by using a random forest, and classification and region extraction segmentation are conducted on the cracks by using a deep convolutional neural network. The traditional image processing method has poor effect on the crack images with complex background, fracture and rough wall surface. The machine learning method has high requirements on the number and the types of the crack samples and has high requirements on the training samples.
The inventor finds that for tunnel apparent gray image data collected in a large batch in a centralized and automatic mode, the data volume is large, the cracks are small, the data volume containing the cracks is small, the calculation amount of a neural network algorithm with high precision is high, and a large amount of time is consumed for processing invalid data. Therefore, the existing machine learning method is not suitable for tunnel structure crack identification in a large-batch data scene.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a two-step tunnel gray image crack identification method, a two-step tunnel gray image crack identification system, a two-step tunnel gray image crack identification medium and two-step tunnel gray image crack identification equipment, which are suitable for tunnel structure crack identification in a large-batch data scene, can greatly reduce data measurement, and greatly improve identification and detection efficiency.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a two-step tunnel gray image crack identification method, which comprises the following steps:
respectively extracting edges of the tunnel gray level images in batches to obtain corresponding tunnel edge images, and screening all suspected images which may have cracks based on comparison between the average gray level of the edge images and a gray level threshold;
obtaining a crack segmentation image of the tunnel gray level images in batches based on the suspected image and the fine segmentation network model, and obtaining a crack identification result by counting the crack segmentation image; the crack identification result comprises the number of images of actually existing cracks, the number of cracks and characteristic parameters of each crack.
A second aspect of the present invention provides a two-step tunnel grayscale image crack identification system, which includes:
the rough screening module is used for respectively carrying out edge extraction on the batch of tunnel gray level images to obtain corresponding tunnel edge images, and screening out all suspected images which possibly have cracks based on comparison between the average gray level of the edge images and a gray level threshold value;
the fine identification module is used for obtaining a crack segmentation image of the tunnel gray level images in batches based on the suspected image and the fine segmentation network model and then obtaining a crack identification result by counting the crack segmentation image; the crack identification result comprises the number of images of actually existing cracks, the number of cracks and characteristic parameters of each crack.
A third aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for crack identification of a two-step tunnel grayscale image as described above.
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 steps of the method for crack identification of a two-step tunnel grayscale image as described above when executing the program.
Compared with the prior art, the invention has the beneficial effects that:
according to the method, the characteristics that the proportion of the cracks to the image size is large and the size of the cracks is small are adopted, the edges of the tunnel gray level images in batches are extracted respectively to obtain corresponding tunnel edge images, and all suspected images with the cracks are screened out based on the comparison between the average gray level of the edge images and the gray level threshold value; and then based on the suspected image and the fine segmentation network model, obtaining a crack segmentation image of the batch tunnel gray level image, and then obtaining a crack identification result by counting the crack segmentation image, so that the method is suitable for identifying the tunnel structure crack under the scene of mass data, greatly reduces the data measurement, and improves the accuracy and efficiency of identification and detection.
Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a schematic diagram of crack identification of a two-step tunnel grayscale image according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for identifying a crack in a two-step tunnel grayscale image according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a fine-segmented network model structure according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a two-step tunnel grayscale image crack identification system according to an embodiment of the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
As shown in fig. 1 and fig. 2, the present embodiment provides a two-step tunnel grayscale image crack identification method, which specifically includes the following steps:
s101: and respectively carrying out edge extraction on the tunnel gray level images in batches to obtain corresponding tunnel edge images, and solving the average gray level of the data. And screening all suspected images which may have cracks based on comparison between the average gray level of the edge image and a gray level threshold value.
Alternatively, the grayscale threshold may be determined in two ways in general. One is a preset gray level threshold, which can be set to 0.001 through simulation experiments. The other method is to count the average gray scale of all the edge extraction results, take the average gray scale as a gray scale threshold, compare the average gray scale with the edge extraction result of each image respectively, and judge whether the image is a suspected crack image.
In step S101, the process of acquiring the batch tunnel grayscale images includes:
and calculating an upper threshold value and a lower threshold value of the original tunnel gray image in batch based on a gradient detection method, and adjusting the upper threshold value and the lower threshold value of the original tunnel gray image to a set gray range. This results in a substantially uniform tunnel grayscale image appearance.
And respectively carrying out edge extraction on the batch tunnel gray level images by using a Canny edge detection algorithm.
In a specific implementation, the original tunnel grayscale image is acquired using a tunnel automated acquisition device.
S102: obtaining a crack segmentation image of the tunnel gray level images in batches based on the suspected image and the fine segmentation network model, and obtaining a crack identification result by counting the crack segmentation image; the crack identification result comprises the number of images of actually existing cracks, the number of cracks and characteristic parameters of each crack.
In a specific implementation, before the suspected image is input to the fine segmentation network model, the method further includes:
and carrying out normalization processing on all the suspected images.
The normalization processing method comprises the following steps:
the input image size is scaled to 640 x 640 pixels and the image data is converted to an RGB color pattern. The image mean value mean is set to [0.485,0.456,0.406]Setting the image variance std to [0.229,0.224,0.225]According to the formula P img ＝(P img Mean)/std calculation of normalized image, where P img The color value of each pixel point in the image. The image data is loaded into the Dataloader for data training.
As shown in fig. 3, the fine segmentation network model is a U-shaped convolutional neural network including residual network modules. The network model is divided into two parts, namely an encoding module and a decoding module, wherein the encoding module comprises a 4-layer down-sampling module, the decoding module comprises a 4-layer up-sampling module, and a residual error module is used for connecting mutually corresponding layers in the encoding module and the decoding module.
For the training data set of the fine segmentation network model and the tunnel crack image samples in the verification data set, the crack areas are marked as white, and the non-crack areas are marked as black.
The training process of the fine segmentation network model is as follows:
marking extracted data of a plurality of existing (for example, 300) tunnel apparent gray level images, marking a crack area as white and uniformly marking a non-crack area as black;
and constructing a training set and a verification set on the basis of the marking data, wherein the ratio of the training set to the verification set is preset, and is 0.85: and 0.15, training the U-shaped neural network by using the data set, training for a plurality of rounds, such as 30 rounds, and setting the learning rate to be 0.03 to complete the training of the network.
The crack identification result includes but is not limited to the number of the existing crack images, the proportion of the crack images in the collected images, the number of the cracks, characteristic parameters such as the length, the width and the shape of each crack, and a fine segmentation result of each crack.
Example two
As shown in fig. 4, the present embodiment provides a two-step tunnel grayscale image crack identification system, which specifically includes the following modules:
the rough screening module 201 is configured to perform edge extraction on the batch of tunnel grayscale images to obtain corresponding tunnel edge images, and then screen out all suspected images that may have cracks based on comparison between the average grayscale of the edge images and a grayscale threshold;
the fine identification module 202 is configured to obtain a crack segmentation image of the batch tunnel grayscale images based on the suspected image and the fine segmentation network model, and obtain a crack identification result by counting the crack segmentation image; the crack identification result comprises the number of images of actually existing cracks, the number of cracks and characteristic parameters of each crack.
It should be noted that, each module in the present embodiment corresponds to each step in the first embodiment one to one, and the specific implementation process is the same, which is not described herein again.
EXAMPLE III
The present embodiment provides a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements the steps in the method for crack identification of a two-step tunnel grayscale image as described above.
Example four
The embodiment provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of the two-step tunnel gray scale image crack identification method.
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.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A two-step tunnel gray image crack identification method is characterized by comprising the following steps:
respectively extracting edges of the tunnel gray level images in batches to obtain corresponding tunnel edge images, and screening out all suspected images with cracks based on comparison between the average gray level of the edge images and a gray level threshold;
obtaining a crack segmentation image of the batch tunnel gray level image based on the suspected image and the fine segmentation network model, and obtaining a crack identification result by counting the crack segmentation image; the crack identification result comprises the number of images of actually existing cracks, the number of cracks and characteristic parameters of each crack.
2. The method for crack recognition of a two-step tunnel gray scale image according to claim 1, wherein the obtaining process of the batch of tunnel gray scale images is as follows:
and calculating an upper threshold and a lower threshold of the original tunnel gray image in batch based on a gradient detection method, and adjusting the upper threshold and the lower threshold of the original tunnel gray image to a set gray range.
3. The method for crack recognition of a two-step tunnel gray image as claimed in claim 1, wherein the Canny edge detection algorithm is used to perform edge extraction on the batch of tunnel gray images respectively.
4. The method for crack recognition of a two-step tunnel gray scale image as claimed in claim 1, wherein before the suspected image is inputted to the fine segmentation network model, the method further comprises:
and carrying out normalization processing on all the suspected images.
5. The method for crack recognition of a two-step tunnel grayscale image of claim 1, wherein the fine segmentation network model is a U-shaped convolutional neural network comprising residual network modules.
6. The method for identifying cracks in a two-step tunnel grayscale image as claimed in claim 1, wherein for the training dataset and the tunnel crack image samples in the verification dataset of the fine segmentation network model, the crack regions are marked as white, and the non-crack regions are marked as black.
7. A two-step tunnel gray scale image crack identification system, comprising:
the rough screening module is used for respectively carrying out edge extraction on the batch of tunnel gray level images to obtain corresponding tunnel edge images, and screening out all suspected images which possibly have cracks based on comparison between the average gray level of the edge images and a gray level threshold value;
the fine identification module is used for obtaining a crack segmentation image of the batch tunnel gray level images based on the suspected image and the fine segmentation network model, and then obtaining a crack identification result by counting the crack segmentation image; the crack identification result comprises the number of images of actually existing cracks, the number of cracks and characteristic parameters of each crack.
8. The system for crack recognition of a two-step tunnel grayscale image of claim 7, wherein in the coarse screening module, the acquisition of the batch of tunnel grayscale images is performed by:
and calculating an upper threshold and a lower threshold of the original tunnel gray image in batch based on a gradient detection method, and adjusting the upper threshold and the lower threshold of the original tunnel gray image to a set gray range.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method for crack identification of a two-step tunnel grayscale image according to any one of claims 1 to 6.
10. 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 steps in the method for crack recognition of a two-step tunnel grayscale image according to any one of claims 1 to 6 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211013049.1A CN115359003A (en) | 2022-08-23 | 2022-08-23 | Two-step tunnel gray image crack identification method, system, medium and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211013049.1A CN115359003A (en) | 2022-08-23 | 2022-08-23 | Two-step tunnel gray image crack identification method, system, medium and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115359003A true CN115359003A (en) | 2022-11-18 |
Family
ID=84002307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211013049.1A Pending CN115359003A (en) | 2022-08-23 | 2022-08-23 | Two-step tunnel gray image crack identification method, system, medium and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115359003A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117409263A (en) * | 2023-12-15 | 2024-01-16 | 成都时代星光科技有限公司 | Unmanned aerial vehicle automatic image correction guiding landing method and computer storage medium |
-
2022
- 2022-08-23 CN CN202211013049.1A patent/CN115359003A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117409263A (en) * | 2023-12-15 | 2024-01-16 | 成都时代星光科技有限公司 | Unmanned aerial vehicle automatic image correction guiding landing method and computer storage medium |
| CN117409263B (en) * | 2023-12-15 | 2024-04-05 | 成都时代星光科技有限公司 | Unmanned aerial vehicle automatic image correction guiding landing method and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115861135B (en) | Image enhancement and recognition method applied to panoramic detection of box body | |
| CN111833306B (en) | Defect detection method and model training method for defect detection | |
| CN111310558A (en) | Pavement disease intelligent extraction method based on deep learning and image processing method | |
| CN109858539A (en) | A kind of ROI region extracting method based on deep learning image, semantic parted pattern | |
| CN103488974A (en) | Facial expression recognition method and system based on simulated biological vision neural network | |
| CN109948625A (en) | Definition of text images appraisal procedure and system, computer readable storage medium | |
| CN109241867B (en) | Method and device for recognizing digital rock core image by adopting artificial intelligence algorithm | |
| CN109598681B (en) | No-reference quality evaluation method for image after repairing of symmetrical Thangka | |
| CN110969620A (en) | Method and device for detecting magnetic shoe ripple defects | |
| CN117974601B (en) | Method and system for detecting surface defects of silicon wafer based on template matching | |
| CN117094975A (en) | Method and device for detecting surface defects of steel and electronic equipment | |
| CN109522768A (en) | The recognition methods of QR code and device | |
| CN115359003A (en) | Two-step tunnel gray image crack identification method, system, medium and equipment | |
| CN115170548A (en) | Leather defect automatic detection method and device based on unsupervised learning | |
| CN112085721A (en) | Damage assessment method, device and equipment for flooded vehicle based on artificial intelligence and storage medium | |
| CN115797314A (en) | Part surface defect detection method, system, equipment and storage medium | |
| CN113223098B (en) | Preprocessing optimization method for image color classification | |
| CN112508860B (en) | Artificial intelligence interpretation method and system for positive check of immunohistochemical image | |
| CN116402822B (en) | Concrete structure image detection method and device, electronic equipment and storage medium | |
| CN116596428B (en) | Rural logistics intelligent distribution system based on unmanned aerial vehicle | |
| CN112784922A (en) | Extraction and classification method of intelligent cloud medical images | |
| CN109829511B (en) | Texture classification-based method for detecting cloud layer area in downward-looking infrared image | |
| CN116934729A (en) | Bridge crack repair trace and secondary crack classification detection method based on image preprocessing and YOLOv7 | |
| CN116152162A (en) | Digitizing method and device for appearance quality residual injury index of cured tobacco leaves | |
| CN115546799A (en) | Backlight-free water meter liquid crystal display screen display number identification method under poor lighting condition |
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 |