Concrete joint surface defect detection method, device and equipment

Abstract

The invention relates to the technical field of image detection, in particular to a concrete joint surface defect detection method, device and equipment, and the method comprises the following steps: acquiring image data of a concrete joint surface; performing data preprocessing on the image data; performing feature extraction on the image data after data preprocessing to obtain feature data of the concrete joint surface; training a defect feature recognition model, and inputting the obtained concrete joint surface feature data into the trained model to recognize defect features and defect types; and marking the identified defect features and defect types in the image data, and storing an identification result in a computer or a server. The image processing technology and the machine learning technology are adopted to intelligently recognize the defect type of the concrete joint surface, automatic, efficient and accurate concrete joint surface defect detection is achieved, and the problem that a traditional concrete joint surface defect detection method is low in efficiency is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image detection, and in particular to a method, device and equipment for detecting defects on a concrete joint surface. Background Art

[0002] In construction projects, whether there are defects on the concrete interface directly affects the safety and durability of the overall structure of the building. Traditional methods for detecting defects on concrete interfaces usually rely on manual visual inspection or the use of tools such as probes. These methods are not only time-consuming and labor-intensive, but also easily affected by human factors, resulting in inaccurate test results and inefficient testing processes. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the present invention provides a method, device and equipment for detecting defects on a concrete joint surface.

[0004] The present invention is achieved through the following technical solutions:

[0005] A method for detecting defects on a concrete bonding surface is provided, the method comprising the following steps:

[0006] Step S10: Acquire image data of the concrete bonding surface;

[0007] Step S20: performing data preprocessing on the image data;

[0008] Step S30: extracting features from the image data after data preprocessing to obtain image feature data of the concrete bonding surface;

[0009] Step S40: training a defect feature recognition model, and using the trained model to identify defect features and defect types according to the acquired concrete bonding surface image feature data;

[0010] Preferably, the image data of the concrete bonding surface is obtained in step S10, including photographing the concrete bonding surface at different angles and lighting conditions using a camera or drone with a resolution of 2k or above, and storing the photographed images on a computer or server in a common image format such as JPEG, PNG, and PDF, etc., for subsequent processing.

[0011] Preferably, the data preprocessing step in step S20 includes:

[0012] Image conversion: convert the acquired concrete bonding surface image into grayscale image;

[0013] Image denoising: After converting to grayscale, use Gaussian filtering to remove the noise in the grayscale image and improve the image quality, as shown in formula (1):

[0014]

[0015] Among them, x is the horizontal coordinate of each pixel point in the grayscale image of the concrete bonding surface in the image coordinate system, y is the vertical coordinate of each pixel point in the grayscale image of the concrete bonding surface in the image coordinate system, and σ is the standard deviation of the Gaussian distribution;

[0016] Image segmentation: Sobel edge detection technology is used to separate the concrete interface area from the background in the grayscale image after noise removal. The Sobel operator formula is shown in formula (2), and the edge strength formula is shown in formula (3):

[0017]

[0018] Among them, A is the original concrete bonding surface image, G x is the image grayscale value of horizontal edge detection, G y is the image grayscale value of the vertical edge detection, G is the grayscale size of the point (x, y);

[0019] Image enhancement: After image segmentation, the contrast stretching technique is used to enhance the image features of the concrete interface area separated from the background. The calculation formula is shown in formula (4):

[0020]

[0021] Among them, s(k) is the enhanced grayscale, k is the grayscale of the concrete interface image, and k min is the minimum gray level of the concrete interface image, k max It is the maximum gray level of the concrete interface image.

[0022] Preferably, the step S30 of extracting features from the image data after data preprocessing comprises:

[0023] Texture feature extraction: The texture features are described by calculating the co-occurrence frequency between different gray levels in the concrete interface image through the gray level co-occurrence matrix. The formula is shown in formula (5):

[0024]

[0025] Where P(d, θ) is the gray level co-occurrence matrix, g(i, j) is the pixel value of the concrete interface image, i, j represent the gray level, d is the distance between adjacent pixels, θ is the direction angle, N is the total number of pixels of the concrete interface image, and δ is the Kroneckerdelta function;

[0026] Grayscale feature extraction: The distribution of different grayscale levels in the concrete interface image is calculated in the form of a grayscale histogram. The calculation formula is shown in formula (6):

[0027]

[0028] Among them, H(k) is the grayscale histogram.

[0029] Preferably, the training step of training the defect feature recognition model in step S40 includes:

[0030] Data preparation and preprocessing: Prepare image data of defects existing on the concrete bonding surface, and perform data cleaning and data segmentation. Data cleaning includes removing noise and filling missing values. Data segmentation is used to divide the data set into three sub-datasets: training set, validation set, and test set.

[0031] Normalization: The data in the data set is normalized using Softmax. The calculation formula is shown in formula (7):

[0032]

[0033] Among them, x a is the ath element in the data set, and n is the total number of elements in the data set;

[0034] Construct defect feature recognition model: This is achieved through convolutional neural network. The model includes convolution layer, pooling layer and fully connected layer. Sigmoid is used as activation function, multi-class cross entropy loss function is used as loss function, Adam optimizer is used to provide different learning rates for the model, and Dropout is used to prevent overfitting in model training.

[0035] Model training: Use the training set divided in data preparation and preprocessing to train the model, so that the model can learn the image features of concrete bonding surface defects and the corresponding defect types;

[0036] Model validation: Use the validation set divided in data preparation and preprocessing to validate the trained model to adjust hyperparameters and prevent overfitting data;

[0037] Model testing: Use the test set divided in data preparation and preprocessing to test the trained model. When the model successfully identifies the image features and defect types of concrete bonding surface defects, the model test is successful and the concrete bonding surface defect feature recognition model training is completed.

[0038] In addition, in order to achieve the above-mentioned purpose, the present invention also proposes a concrete joint surface defect detection device, the concrete joint surface defect detection device comprising:

[0039] Image acquisition module: used to obtain image data of concrete bonding surface;

[0040] Image preprocessing module: used to preprocess image data;

[0041] Feature extraction module: used to extract features from the image data after data preprocessing, and obtain image feature data of the concrete bonding surface;

[0042] Defect feature recognition module: training defect feature recognition model, and using the trained model to identify defect features and defect types based on the acquired concrete bonding surface image feature data;

[0043] In addition, to achieve the above-mentioned purpose, the present invention also proposes a concrete bonding surface defect detection device, which includes: a memory, a processor, and programs such as data preprocessing, data feature extraction, and convolutional neural network algorithm stored in the memory and run on the processor. The programs such as data preprocessing, data feature extraction, and convolutional neural network algorithm are steps for implementing the concrete bonding surface defect detection method and system as described above.

[0044] Preferably, to achieve the above-mentioned purpose, the present invention also provides a computer program product, which includes programs such as data preprocessing, data feature extraction and convolutional neural network algorithm, and when the programs such as data preprocessing, data feature extraction and convolutional neural network algorithm are executed by a processor, the concrete bonding surface defect detection method and system as described above are implemented.

[0045] The advantages and effects of the present invention are:

[0046] The present invention proposes a method, device and equipment for detecting defects on concrete surfaces, which use image processing technology and machine learning technology to intelligently identify the defect types of concrete surfaces, thereby realizing automated, efficient and accurate detection of defects on concrete surfaces, and effectively solving the problem of low efficiency of traditional methods for detecting defects on concrete surfaces. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 The present invention is a flow chart of a method for detecting defects on a concrete bonding surface.

[0049] Figure 2 The present invention is a schematic structural diagram of a concrete bonding surface defect detection device.

[0050] Figure 3 The present invention is a schematic block diagram of the structure of an electronic device for detecting defects on a concrete joint surface. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The present invention provides a method for detecting defects on a concrete bonding surface. Figure 1 As shown, the following steps are included:

[0053] Step S10: Acquire image data of concrete joint surfaces, mainly image data of concrete joint surfaces that need to be inspected for defects, or image data of concrete joint surfaces that have been inspected, to ensure the safety of building structures and construction through re-inspection.

[0054] Among them, obtaining image data of the concrete bonding surface includes taking multiple images of the concrete bonding surface at different angles and lighting conditions to ensure that the captured images cover the entire concrete bonding surface area; using a high-resolution camera or drone with a resolution of 2k or more to photograph the concrete bonding surface to ensure that the image clarity and resolution are high enough to capture small defect features; storing the captured images on a computer or server in common image formats such as JPEG, PNG, and PDF for subsequent processing.

[0055] Step S20: performing image conversion, image denoising, image segmentation and image enhancement data preprocessing operations on the image data, specifically including:

[0056] Image conversion: convert the acquired concrete bonding surface image into grayscale image;

[0057] Image denoising: After converting to grayscale, use Gaussian filtering to remove the noise in the grayscale image and improve the image quality, as shown in formula (1):

[0058]

[0059] Among them, x is the horizontal coordinate of each pixel point in the grayscale image of the concrete bonding surface in the image coordinate system, y is the vertical coordinate of each pixel point in the grayscale image of the concrete bonding surface in the image coordinate system, and σ is the standard deviation of the Gaussian distribution;

[0060] Image segmentation: Sobel edge detection technology is used to separate the concrete interface area from the background in the grayscale image after noise removal. The Sobel operator formula is shown in formula (2), and the edge strength formula is shown in formula (3):

[0061]

[0062] Among them, A is the original concrete bonding surface image, G x is the image grayscale value of horizontal edge detection, G y is the image grayscale value of the vertical edge detection, G is the grayscale size of the point (x, y);

[0063] Image enhancement: After image segmentation, the contrast stretching technique is used to enhance the image features of the concrete interface area separated from the background. The calculation formula is shown in formula (4):

[0064]

[0065] Among them, s(k) is the enhanced grayscale, k is the grayscale of the concrete interface image, and k min is the minimum gray level of the concrete interface image, k max It is the maximum gray level of the concrete interface image.

[0066] Step S30: extracting features from the image data after data preprocessing to obtain image feature data of the concrete bonding surface.

[0067] Among them, feature extraction of image data after data preprocessing includes:

[0068] Texture feature extraction: The texture features are described by calculating the co-occurrence frequency between different gray levels in the concrete interface image through the gray level co-occurrence matrix. The formula is shown in formula (5):

[0069]

[0070] Where P(d, θ) is the gray level co-occurrence matrix, g(i, j) is the pixel value of the concrete interface image, i, j represent the gray level, d is the distance between adjacent pixels, θ is the direction angle, N is the total number of pixels of the concrete interface image, and δ is the Kroneckerdelta function;

[0071] Grayscale feature extraction: The distribution of different grayscale levels in the concrete interface image is calculated in the form of a grayscale histogram. The calculation formula is shown in formula (6):

[0072]

[0073] Among them, H(k) is the grayscale histogram.

[0074] Step S40: training a concrete joint surface defect feature recognition model, and using the trained model to recognize defect features and defect types based on the acquired concrete joint surface image feature data.

[0075] Among them, the training steps of training the concrete interface defect feature recognition model include:

[0076] Data preparation and preprocessing: Prepare image data of defects existing on the concrete bonding surface, and perform data cleaning and data segmentation. Data cleaning includes removing noise and filling missing values. Data segmentation is used to divide the data set into three sub-datasets: training set, validation set, and test set.

[0077] Normalization: The data in the data set is normalized using Softmax. The calculation formula is shown in formula (7):

[0078]

[0079] Among them, x a is the ath element in the data set, and n is the total number of elements in the data set;

[0080] Construct defect feature recognition model: This is achieved through convolutional neural network. The model includes convolution layer, pooling layer and fully connected layer. Sigmoid is used as activation function, multi-class cross entropy loss function is used as loss function, Adam optimizer is used to provide different learning rates for the model, and Dropout is used to prevent overfitting in model training.

[0081] Model training: Use the training set divided in data preparation and preprocessing to train the model, so that the model can learn the image features of concrete bonding surface defects and the corresponding defect types;

[0082] Model validation: Use the validation set divided in data preparation and preprocessing to validate the trained model to adjust hyperparameters and prevent overfitting data;

[0083] Model testing: Use the test set divided in data preparation and preprocessing to test the trained model. When the model successfully identifies the image features and defect types of concrete bonding surface defects, the model test is successful and the concrete bonding surface defect feature recognition model training is completed.

[0084] In addition, the present invention also proposes a concrete joint surface defect detection device, please refer to Figure 2 The concrete interface defect detection device comprises:

[0085] Image acquisition module: used to obtain image data of concrete bonding surface;

[0086] Image preprocessing module: used to preprocess image data;

[0087] Feature extraction module: used to extract features from the image data after data preprocessing, and obtain image feature data of the concrete bonding surface;

[0088] Defect feature recognition module: training defect feature recognition model, and using the trained model to identify defect features and defect types based on the acquired concrete bonding surface image feature data;

[0089] The concrete interface defect detection device provided by the present application adopts the concrete interface defect detection method in the above embodiment, which can solve the technical problem that complex public data is difficult to process. Compared with the prior art, the beneficial effects of the concrete interface defect detection device provided by the present application are the same as the beneficial effects of the concrete interface defect detection method provided by the above embodiment, and other technical features in the concrete interface defect detection device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0090] The present application provides a concrete bonding surface defect detection device, which includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the concrete bonding surface defect detection method in the above-mentioned embodiment 1.

[0091] Reference below Figure 3 , which shows a schematic diagram of the structure of a concrete interface defect detection device suitable for implementing the embodiment of the present application. The concrete interface defect detection device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The concrete interface defect detection device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0092] Figure 3The concrete interface defect detection device shown may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM) 1004. In RAM 1004, various programs and data required for the operation of the concrete interface defect detection device are also stored. The processing device 1001, ROM 1002, and RAM 1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the concrete interface defect detection device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a concrete interface defect detection device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0093] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0094] The concrete interface defect detection device provided by the present application adopts the concrete interface defect detection method in the above embodiment, which can solve the technical problem that complex public data is difficult to process. Compared with the prior art, the beneficial effects of the concrete interface defect detection device provided by the present application are the same as the beneficial effects of the concrete interface defect detection method provided by the above embodiment, and the other technical features of the concrete interface defect detection device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0095] The various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0096] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned concrete bonding surface defect detection method when executed by a processor.

[0097] The computer program product provided by this application can solve the technical problem that complex public data is difficult to process. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the concrete interface defect detection method provided by the above embodiment, which will not be repeated here.

[0098] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations 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 include these modifications and variations.

Claims

1. A method for detecting defects on a concrete bonding surface, characterized in that: The method comprises the following steps: Step S10: Acquire image data of the concrete bonding surface; Step S20: performing data preprocessing on the image data; Step S30: extracting features from the image data after data preprocessing to obtain image feature data of the concrete bonding surface; Step S40: training a concrete joint surface defect feature recognition model, inputting the acquired concrete joint surface feature data into the trained model to recognize defect features and defect types; The step of obtaining the image data of the concrete joint surface in step S10 includes photographing the concrete joint surface using a camera or drone with a resolution of 2k or above under different angles and lighting conditions; and storing the photographed images on a computer or server.

2. A method for detecting defects on a concrete bonding surface according to claim 1, characterized in that: The data preprocessing step in step S20 includes: Image conversion: convert the acquired concrete bonding surface image into grayscale image; Image denoising: After converting to grayscale, use Gaussian filtering to remove the noise in the grayscale image, as shown in formula (1): Among them, x is the horizontal coordinate of each pixel point in the grayscale image of the concrete bonding surface in the image coordinate system, y is the vertical coordinate of each pixel point in the grayscale image of the concrete bonding surface in the image coordinate system, and σ is the standard deviation of the Gaussian distribution; Image segmentation: Sobel edge detection technology is used to separate the concrete interface area from the background in the grayscale image after noise removal. The Sobel operator formula is shown in formula (2), and the edge strength formula is shown in formula (3): Among them, A is the original concrete bonding surface image, G x is the image grayscale value of horizontal edge detection, G y is the image grayscale value of the vertical edge detection, G is the grayscale size of the point (x, y); Image enhancement: After image segmentation, the contrast stretching technique is used to enhance the image features of the concrete interface area separated from the background. The calculation formula is shown in formula (4): Among them, s(k) is the enhanced grayscale, k is the grayscale of the concrete interface image, and k min is the minimum gray level of the concrete interface image, k max It is the maximum gray level of the concrete interface image.

3. A method for detecting defects on a concrete bonding surface according to claim 1, characterized in that: The step S30 of extracting features from the image data after data preprocessing includes: Texture feature extraction: The texture features are described by calculating the co-occurrence frequency between different gray levels in the concrete interface image through the gray level co-occurrence matrix. The formula is shown in formula (5): Where P(d, θ) is the gray level co-occurrence matrix, g(i, j) is the pixel value of the concrete interface image, i, j represent the gray level, d is the distance between adjacent pixels, θ is the direction angle, N is the total number of pixels of the concrete interface image, and δ is the Kroneckerdelta function; Grayscale feature extraction: The distribution of different grayscale levels in the concrete interface image is calculated in the form of a grayscale histogram. The calculation formula is shown in formula (6): Among them, H(k) is the grayscale histogram.

4. A method for detecting defects on a concrete bonding surface according to claim 1, characterized in that: The training step of training the concrete interface defect feature recognition model in step S40 includes: Data preparation and preprocessing: Prepare image data of defects existing on the concrete bonding surface, and perform data cleaning and data segmentation. Data cleaning includes removing noise and filling missing values. Data segmentation is used to divide the data set into three sub-datasets: training set, validation set, and test set. Normalization: The data in the data set is normalized using Softmax. The calculation formula is shown in formula (7): Among them, x a is the ath element in the data set, and n is the total number of elements in the data set; Construct a concrete interface defect feature recognition model: This is implemented through a convolutional neural network. The model includes a convolutional layer, a pooling layer, and a fully connected layer. Sigmoid is used as the activation function, a multi-class cross entropy loss function is used as the loss function, and the Adam optimizer is used to provide different learning rates for the model. Dropout is used to prevent overfitting in model training. Model training: Use the training set divided in data preparation and preprocessing to train the model, so that the model can learn the image features of concrete bonding surface defects and the corresponding defect types; Model validation: Use the validation set divided in data preparation and preprocessing to validate the trained model to adjust hyperparameters and prevent overfitting data; Model testing: Use the test set divided in data preparation and preprocessing to test the trained model. When the model successfully identifies the image features and defect types of concrete bonding surface defects, the model test is successful and the concrete bonding surface defect feature recognition model training is completed.

5. A concrete joint surface defect detection device, characterized in that: The concrete joint surface defect detection device comprises: Image acquisition module: used to obtain image data of concrete bonding surface; Image preprocessing module: used to preprocess image data; Feature extraction module: used to extract features from the image data after data preprocessing, and obtain image feature data of the concrete bonding surface; Defect feature recognition module: train the defect feature recognition model, and use the trained model to identify defect features and defect types based on the acquired concrete bonding surface image feature data.

6. A concrete joint surface defect detection device, characterized in that: The concrete interface defect detection equipment comprises: A memory, a processor, and a concrete bonding surface defect detection program stored in the memory and executable on the processor, wherein the concrete bonding surface defect detection program, when executed by the processor, implements the concrete bonding surface defect detection method according to any one of claims 1 to 4.

7. A computer program product, characterized in that The computer program product comprises a concrete bonding surface defect detection program, and when the concrete bonding surface defect detection program is executed by a processor, the concrete bonding surface defect detection method according to any one of claims 1 to 4 is implemented.

Images