Concrete quality detection method and system for hydraulic engineering

Abstract

The invention provides a concrete quality detection method and system for hydraulic engineering, and relates to the technical field of image processing, and the method comprises the steps: obtaining a plurality of training samples; each training sample is preprocessed; based on a deep neural network algorithm, a concrete quality detection model is constructed in combination with a deep completion network module, a deep adaptive region proposing module and a multi-resolution detection network module; inputting each preprocessed training sample into a concrete quality detection model, and outputting a concrete quality detection result; according to the label of the concrete and the quality detection result, a concrete quality detection model is trained in a combined mode; obtaining an RGB-D image of to-be-detected concrete; and inputting the RGB-D image of the to-be-detected concrete into the trained concrete quality detection model to complete the quality detection of the to-be-detected concrete.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a concrete quality detection method and system for water conservancy projects. BACKGROUND

[0002] Water conservancy projects refer to engineering projects related to the development, utilization, protection and prevention of water resources, such as reservoirs, dams, river regulation, hydropower stations, etc. Concrete is a building material made of cement, aggregate (such as sand, gravel) and water, and is widely used in building structures, while concrete quality detection is to automatically and intelligently detect possible defects (such as cracks, peeling, etc.) in concrete through image processing, deep learning, etc.

[0003] For water conservancy projects, the quality of concrete is related to the safety and durability of the project, especially in infrastructure such as dams and water channels, cracks and peeling of concrete can cause serious safety problems, so detecting the quality of concrete in water conservancy projects helps to find problems in time and repair in advance to ensure the safe operation of the project, which is of great significance to improve the safety and quality of water conservancy projects.

[0004] However, traditional detection methods rely on manual inspection, which is easily affected by human factors, leading to missed detection or misjudgment. Secondly, manual detection is low in efficiency, especially in large-scale projects, the detection process is time-consuming and laborious, which leads to the inability to assess internal defects in depth, and usually lacks real-time performance, which cannot timely discover potential problems, affecting the post-maintenance and safety protection of the project. SUMMARY

[0005] In view of the above deficiencies of the prior art, the purpose of the embodiments of the present application is to provide a concrete quality detection method for water conservancy projects, which can solve the technical problems that traditional detection methods rely on manual inspection, are easily affected by human factors, leading to missed detection or misjudgment. Secondly, manual detection is low in efficiency, especially in large-scale projects, the detection process is time-consuming and laborious, which leads to the inability to assess internal defects in depth, and usually lacks real-time performance, which cannot timely discover potential problems, affecting the post-maintenance and safety protection of the project.

[0006] The first aspect of the embodiments of the present application proposes a concrete quality detection method for water conservancy projects, comprising: S1: obtaining a plurality of training samples, wherein each training sample comprises a set of RGB-D images of a concrete sample and a corresponding label set; S2: preprocessing each training sample; S3: based on a deep neural network algorithm, combining a depth completion network module, a depth adaptive region proposal module and a multi-resolution detection network module, a concrete quality detection model is constructed; S4: inputting each preprocessed training sample into the concrete quality detection model to output a quality detection result of the concrete; S5: training the concrete quality detection model in a joint manner according to the label of the concrete and the quality detection result; S6: obtaining an RGB-D image of the concrete to be detected; S7: inputting the RGB-D image of the concrete to be detected into the trained concrete quality detection model to complete quality detection of the concrete to be detected.

[0007] In a second aspect of the embodiment of the present application, a concrete quality detection system for hydraulic engineering is provided, which comprises a processor and a memory. The memory stores programs or instructions executable on the processor, and the programs or instructions are executed by the processor to implement the steps of the concrete quality detection method for hydraulic engineering according to the first aspect.

[0008] In a third aspect of the embodiment of the present application, a readable storage medium is provided, and the readable storage medium stores programs or instructions, and the programs or instructions are executed by the processor to implement the steps of the concrete quality detection method for hydraulic engineering according to the first aspect.

[0009] The technical scheme provided by the embodiment of the present application has at least the following beneficial effects: In the embodiment of the present application, a diversified RGB-D image set and corresponding labels are obtained, and a concrete quality detection model is constructed based on a deep neural network algorithm, combined with a depth completion network module, a depth adaptive region proposal module and a multi-resolution detection network module. Then, each preprocessed training sample is input into the concrete quality detection model to output a quality detection result of the concrete. The concrete quality detection model is trained in a joint manner according to the label of the concrete and the quality detection result, ensuring efficient defect recognition. Finally, an RGB-D image of the concrete to be detected is obtained, and the RGB-D image of the concrete to be detected is input into the trained concrete quality detection model, realizing comprehensive evaluation of the quality of the concrete, greatly improving the detection precision and efficiency, and ensuring engineering safety by timely discovering potential defects. BRIEF DESCRIPTION OF DRAWINGS

[0010] The accompanying drawings are only for the purpose of illustrating specific embodiments and are not considered to be limiting of the present application. Throughout the drawings, the same reference symbols are used to denote the same components. Obviously, the accompanying drawings described below are only some embodiments described in the embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0011] Figure 1is a flowchart of a concrete quality detection method for water conservancy projects provided by an embodiment of the present application. Figure 2 is a structural diagram of a concrete quality detection system for water conservancy projects provided by an embodiment of the present application. DETAILED DESCRIPTION

[0012] In order for those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions of the present application will be described clearly and completely below in conjunction with the drawings. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor should fall within the scope of protection of the present application.

[0013] The concrete quality detection method for water conservancy projects provided by the embodiments of the present application will be described in detail below in conjunction with the drawings, specific embodiments and application scenarios.

[0014] Reference is made to the accompanying drawings Figure 1 , which shows a flowchart of a concrete quality detection method for water conservancy projects provided by an embodiment of the present application.

[0015] The embodiments of the present application provide a concrete quality detection method for water conservancy projects, which can include the following steps: S1: Obtain a plurality of training samples, wherein each training sample includes a set of RGB-D images of concrete and a corresponding set of labels.

[0016] Among them, the training sample is a data set used for training the model. Each sample includes input data RGB-D image and corresponding label concrete defect type or position, and the RGB-D image set is an image data set composed of RGB (color) image and D (depth) image. The RGB image provides the color information of the scene, and the D image provides the depth information of the objects in the scene, so that the system can obtain the three-dimensional structure of the space.

[0017] Among them, the label set is the correct answer corresponding to the input image in the training sample. For concrete quality detection, the label set usually includes information such as the type and position of defects.

[0018] Specifically, the RGB-D image is obtained by a UAV equipped with an RGB-D camera. The RGB-D camera can provide color images and depth images at the same time.

[0019] It should be noted that by using RGB-D image sets and label sets, comprehensive training data can be provided, helping the model accurately identify concrete defects in various complex scenarios. The combination of RGB-D images provides the model with rich spatial information, and depth images are particularly helpful in identifying surface and internal defects in concrete, such as cracks or spalling, thus achieving greater accuracy than traditional color images. Drones equipped with RGB-D cameras can efficiently acquire images at challenging hydraulic engineering sites, avoiding the tediousness and limitations of manual inspection.

[0020] S2: Preprocess each training sample.

[0021] Preprocessing refers to various data cleaning and optimization operations performed before the data is input into the model.

[0022] It should be noted that preprocessing is used to improve image quality, especially in complex environments where RGB-D images can be affected by noise and other interference. Subtraction preprocessing and median filtering effectively remove noise from the image, ensuring the clarity and accuracy of the image data.

[0023] In one possible implementation, the preprocessing is specifically: subtraction preprocessing.

[0024] Subtraction preprocessing specifically includes: S201: The RGB-D image is smoothed using a median filter to obtain a corrected image.

[0025] S202: Based on the corrected image, remove noise from the RGB-D image through subtraction preprocessing:

[0026] in, I s ( x i ) represents the _th ... i The value of each pixel, max indicates the maximum value. Represents the RGB-D image of the th j The brightness value of each pixel. Represents the RGB-D image of the th i The brightness value of each pixel. R i Indicates the first i The set of pixel values ​​in the region surrounding each pixel. median This indicates the median operation.

[0027] It is worth noting that through the subtraction preprocessing, the corrected image can retain important structural information and remove irrelevant interference, which provides more accurate input data for the subsequent quality detection model. Compared with the original image without processing, the preprocessed image is easier for the model to understand, thereby improving the detection accuracy and efficiency. In general, by eliminating noise and improving image quality, more reliable data support is provided for concrete quality detection, ensuring that the model can accurately identify various defects in the concrete.

[0028] S3: Based on the deep neural network algorithm, combined with the deep completion network module, the deep adaptive region proposal module and the multi-resolution detection network module, a concrete quality detection model is constructed.

[0029] Among them, the deep neural network algorithm is a kind of artificial neural network, which processes input data through multiple layers of nonlinear transformation, and is widely used in pattern recognition, image classification and other tasks.

[0030] Among them, the deep completion network module is a network module used to fill in the blank areas in the image. Especially in RGB-D images, when the depth information is missing or blocked, the completion network can infer and reconstruct the missing depth information to make the subsequent detection more accurate.

[0031] Among them, the deep adaptive region proposal module is a deep learning module that automatically extracts regions that may contain defects based on image depth information and texture information. In concrete quality detection, the deep adaptive region proposal module can accurately locate the defect area, avoiding the waste of calculation caused by full image processing.

[0032] Among them, the multi-resolution detection network module refers to the network module that processes images at different resolutions. By analyzing images at different scales, it can better detect defects of different sizes and scales, improving the robustness and accuracy of the model.

[0033] Among them, the concrete quality detection model is a model integrating deep completion network, deep adaptive region proposal module and multi-resolution detection network module.

[0034] It should be noted that by integrating the deep completion network, the deep adaptive region proposal module and the multi-resolution detection network module, a powerful concrete quality detection model is constructed. The deep completion network can fill in the blank areas in the RGB-D image, ensuring that all image information is effectively utilized. The deep adaptive region proposal module can accurately identify and locate possible defect areas, significantly improving the accuracy of detection and reducing the waste of computing resources. The use of the multi-resolution detection network module enables the model to detect defects at different scales, especially when facing defects of different sizes and shapes, which improves the flexibility and comprehensiveness of detection.

[0035] S4: input the pre-processed training samples into the concrete quality detection model, and output the quality detection result of the concrete.

[0036] Among them, the quality detection result is the conclusion drawn by the model after processing the input data, usually including the category of defects (such as cracks, peeling, etc.) and the location of defects, and the quality detection result is the output of the model, used to evaluate the quality of the concrete.

[0037] It should be noted that by inputting the pre-processed training samples into the trained concrete quality detection model, automatic quality detection can be achieved. The pre-processed samples have removed noise and enhanced the quality of the images, ensuring that the model can make accurate judgments based on high-quality data. After inputting these data into the trained model, the model can quickly process and give the quality detection result, not only improving the detection efficiency, but also reducing the error of manual operation.

[0038] In one possible implementation, the quality detection result includes: the defect category of the concrete and the defect location.

[0039] Among them, the defect category refers to different types of defects found in the concrete, and common concrete defect categories include: cracks, peeling, air holes and corrosion, and the defect location refers to the specific area or coordinate position of the defect in the concrete structure.

[0040] In one possible implementation, S4 specifically includes: S401: complete the blank areas of the RGB-D image through the deep completion network module.

[0041] S402: according to the completed RGB-D image, generate a deep adaptive region proposal through the deep adaptive region proposal module:

[0042] Among them, flag represents the classification label, n ( Trepresents the total number of pixels in the input image, T represents the number of pixels in the input image located within the defect region, f1ag = 0 represents that there is no any defect pixel in the input image and the image is classified as a background image, N ( T represents the total number of pixels in the input image, T represents the total number of pixels in the input image, k represents the empirical percentage threshold, f1ag = discard represents that the input image does not meet the condition of a defect image and should be discarded, f1ag = 1 represents that the defect type of the input image is labeled as a crack, f1ag = 2 represents that the defect type of the input image is labeled as spalling.

[0043] S403: According to the deep adaptive region proposal, the defect category and location of the concrete are output by the multi-resolution detection network module.

[0044] It should be noted that the blank area in the RGB-D image is filled by the deep completion network module to ensure the completeness of the depth data of each pixel and improve the accuracy of the image. The deep adaptive region proposal module is used to automatically identify and locate possible defect regions based on the completed image, reducing the processing of irrelevant regions and improving the detection efficiency. In this process, the defect type and region of the image are accurately labeled, such as cracks and spalling, ensuring that subsequent processing is more accurate. Combined with the multi-resolution detection network module, the category and location of the defect are accurately identified, improving the flexibility and accuracy of the detection. The advantages of this series of operations are that the concrete defects can be automatically and efficiently identified, significantly improving the accuracy, processing speed, and robustness of the system, reducing manual intervention and errors.

[0045] In one possible implementation, S401 specifically includes: S4011: Calculate the gradient of the depth map in the RGB-D image by an independent U-Net framework:

[0046] where △( x ) represents the horizontal gradient of the depth map, and △( y ) represents the vertical gradient of the depth map.

[0047] It should be noted that U-Net is a deep learning network architecture widely used in image segmentation tasks, especially in medical image analysis. Its structure consists of an encoder and a decoder, which can effectively extract features from images and make accurate pixel-level predictions. In this scenario, U-Net is used to process the depth map in the RGB-D image to calculate the gradient of the image.

[0048] S4012: Determine the surface normal of each pixel in the depth map according to the gradient:

[0049] wherein, represents the surface normal of the depth map.

[0050] wherein, the surface normal refers to a vector perpendicular to the image surface, which is used to describe the direction of the surface. In 3D image processing, the surface normal helps determine the geometry and direction of the image surface, and in the depth map, the calculation of the normal is crucial for reconstructing the blank area of the image.

[0051] S4013: Complete the blank area of the RGB-D image according to the surface normal.

[0052] It should be noted that by combining the U-Net framework, gradient calculation and surface normal reconstruction, the completion accuracy of the blank area in the RGB-D image is effectively improved. The U-Net framework accurately processes the depth map through deep learning, calculates the gradient of the image, and effectively captures the subtle changes in the depth map, especially in complex environments. Through gradient calculation, areas with significant depth changes can be identified, which is crucial for determining the details of the concrete surface (such as cracks or irregularities). Then, based on the gradient information, the calculation of the surface normal helps accurately reconstruct the blank area in the depth map, ensuring that the depth information of all areas is complete and accurate, thereby avoiding false detection caused by missing data.

[0053] In one possible implementation, S403 specifically includes: S4031: Adjust the size of the depth adaptive region proposal to extract image features:

[0054] wherein, X 1 represents the feature map, CNN represents the convolutional neural network, w represents the convolution kernel, b represents the bias, T represents the input image, represents the image after the input image is reduced by 1 / 2, represents the image after the input image is reduced by 1 / 4, represents the feature map of the input image after being reduced by 1 / 2, represents the feature map of the input image after being reduced by 1 / 4.

[0055] wherein the depth adaptive region proposal refers to the selection of regions in the image that may contain defects by analyzing depth image data. This proposal automatically determines which regions require more attention based on the depth information of the image, thereby improving the accuracy of quality detection. Image features refer to the key information extracted from the input image by deep learning algorithms that can describe the content of the image. Feature maps contain the structure, texture and other visual information in the image, which helps subsequent classification and analysis.

[0056] S4032: Upsampling operation is performed on the image features.

[0057] wherein upsampling refers to enlarging the size of the image feature map to restore the spatial resolution of the image. It is usually used to generate higher resolution images for more accurate subsequent analysis.

[0058] S4033: Based on the image features after upsampling operation, combined with convolution operation and average pooling operation, output feature values:

[0059] wherein, f q represents the feature value of the first q layer feature map obtained by average pooling operation, C q represents the feature map of the first q layer, size represents the number of pixels, represents the summation symbol.

[0060] wherein, convolution operation is to filter and extract features of the image with convolution kernel, and average pooling operation reduces the resolution of the image by calculating the average value of pixels in a small area of the image, thereby reducing the amount of calculation and retaining important features.

[0061] S4034: According to the feature values, through the full connection layer, output the defect category and position of the concrete.

[0062] It should be noted that through the multi-level image processing and feature extraction method, efficient and accurate concrete defect classification and positioning are realized. First, the size of the depth adaptive region proposal is adjusted to extract image features, which can accurately focus on potential defect areas in the image, thereby improving the effect of subsequent detection. Through upsampling operation, the resolution of the image feature map is improved, which helps to retain more image details and enhance the recognition ability of defects. Combined with convolution operation and average pooling operation, further extraction and compression of image features are performed to ensure efficient feature expression and computing power. Finally, through the full connection layer, the defect category and position of the concrete are output, realizing accurate defect classification and positioning.

[0063] S5: According to the label of the concrete and the quality detection result, the concrete quality detection model is trained by a joint manner.

[0064] The joint manner refers to the process of optimizing multiple objectives or modules simultaneously during model training. Through simultaneously optimizing the loss function, the multiple parts of the concrete quality detection model are adjusted to work more coordinately and efficiently.

[0065] It should be noted that the overall performance of the concrete quality detection model is effectively improved through joint training. By considering the label of the concrete and the quality detection result of the model, joint training enables the model to optimize multiple objectives simultaneously, thereby improving the comprehensive performance of the model. Compared with optimizing each module separately, joint training can more accurately balance the influence of different modules, avoiding excessive adjustment of certain modules, and improving the stability and accuracy of the model. In this way, the model can not only accurately identify the defect category, but also accurately locate the defect position, thereby providing more comprehensive and reliable quality detection results.

[0066] In one possible implementation, S5 specifically includes: S501: According to the error between the label of the concrete and the quality detection result, the first loss function and the second loss function of the concrete quality detection model are determined.

[0067] The first loss function is specifically:

[0068] Wherein, Loss 1 represents the loss function of the deep completion network module, i.e. the first loss function, and arg min represents minimization, L 1 represents the L1 loss, represents the surface normal of the depth map, represents the true normal of the depth map, represents the predicted depth of the input image T represents the true depth of the input image T

[0069] The second loss function is specifically:

[0070] Wherein, Loss 2 represents the loss function of the multi-resolution detection network module, i.e. the second loss function, y r represents the true label of the r th training sample, log represents the logarithmic function, x r ​​denotes r one training sample, W AdaNet denotes AdaNet weights of the model, f denotes the probability of predicting the output class.

[0071] S502: If both the first loss function and the second loss function are less than the preset loss function value, the training of the concrete quality detection model is completed.

[0072] wherein the loss function is a function used in deep learning to quantify the gap between the model's prediction and the actual label. The first loss function and the second loss function correspond to different model modules or tasks, and the model is trained and adjusted by optimizing the loss function.

[0073] wherein the size of the preset loss function value can be set by a person skilled in the art according to the actual situation, and the present application does not make any limitation.

[0074] It should be noted that the first loss function focuses on the optimization of the deep completion network module, while the second loss function focuses on the accuracy of the multi-resolution detection module, ensuring that the model's performance at different levels is optimized. In addition, through the weight adjustment of the AdaNet model, the model can dynamically optimize the weight distribution during the training process, improving the adaptability and performance of the model. Ultimately, through joint training, the model can better capture the complex features of concrete defects, improving the overall prediction accuracy and stability. In summary, through the fine-grained loss function design and joint training strategy, the performance of the model in concrete quality detection is improved, providing a solid foundation for efficient and accurate detection.

[0075] S6: Obtain the RGB-D image of the concrete to be detected.

[0076] S7: Input the RGB-D image of the concrete to be detected into the trained concrete quality detection model to complete the quality detection of the concrete to be detected.

[0077] wherein the trained concrete quality detection model refers to the deep learning model optimized through the previous training process, which has learned how to identify and classify defects in concrete based on RGB-D images, and has been trained to accurately determine the defect type and location in new images. Quality detection refers to the inspection and analysis of concrete to determine whether there are defects or problems. In this context, quality detection mainly refers to analyzing the concrete image to be detected by a deep learning model to identify defects such as cracks and peeling.

[0078] It should be noted that by inputting the RGB-D image of the concrete to be inspected into the trained quality inspection model, quality inspection can be completed quickly and accurately. The trained model has learned the characteristics of concrete defects, thus enabling it to efficiently identify potential defects when faced with new images, reducing manual intervention. The depth information of the RGB-D image enhances the spatial understanding of quality inspection, allowing the model to not only identify surface defects but also analyze the deep structure within the image, showing a significant advantage, especially in detecting complex defects such as cracks and spalling. Furthermore, the automated inspection process significantly improves inspection efficiency, shortens inspection time, provides engineers with more timely feedback, and effectively improves the level of concrete quality control. In summary, the intelligent and automated inspection process ensures both accuracy and efficiency in the inspection process.

[0079] In one possible implementation, the process after S7 includes: S8: When the defect type is crack, determine the width of the crack.

[0080] Cracks are a common defect in concrete, usually caused by external forces, temperature changes, or material problems. The presence of cracks can affect the structural strength and durability of concrete, therefore accurate measurement of their width and other characteristics is necessary. Width is a key parameter of a crack, representing the distance from one side to the other. In concrete quality inspection, determining the width of cracks helps assess their impact on structural safety and, consequently, whether repair is required.

[0081] It's important to note that accurately measuring crack width provides a more detailed analysis of concrete defects, especially since width is a critical indicator in crack assessment. Crack width directly impacts the structural strength and safety of concrete; therefore, automated and accurate crack width measurement enables the timely detection of potential risk areas. This method not only improves detection accuracy but also reduces the bias of manual measurements, allowing engineers to obtain reliable crack data and make informed decisions.

[0082] In one possible implementation, S8 specifically includes: S801: Determine the coordinates of the crack's centerline:

[0083] in, Indicates the first l The pixel at the th point f The coordinates on the central axis of the crack. Indicates the first f The center of the crack, h l Indicates the first la scale factor for each pixel point, represents the edge point of the f th crack.

[0084] where the center axis coordinate refers to the position of each point on the center line of the crack. The center axis of the crack is usually the center line of symmetry of the crack, and by calculating the position of each point, the shape of the crack can be described more accurately.

[0085] S802: Calculate the brightness value of each point on the center axis of the crack according to the center axis coordinate:

[0086] where, represents the brightness value of the f th pixel point of the l th crack, g 0 represents the brightness of the image.

[0087] where the brightness value refers to the brightness of each pixel in the image, usually related to the color or depth of the pixel. In crack detection, the brightness value can help distinguish the difference between the crack and the surrounding area.

[0088] S803: Smooth the edge brightness change of the crack by using the Cardan spline interpolation method: where, represents the smooth brightness value of the f th pixel point of the l th crack under the parameter t , represents the brightness value of the f th pixel point in the l th crack, j + 1, represents the basis function of the Cardan spline interpolation method for the j th pixel point under the parameter t .

[0089] where the Cardan spline interpolation method is a mathematical method used to smooth and approximate the relationship between data points, commonly used in image processing to smooth brightness changes, remove noise, and preserve the overall shape of the data.

[0090] S804: Derive the edge brightness change of the crack after smoothing to determine the edge position of the crack.

[0091] where the edge point of the crack refers to the extreme position at both ends of the crack, usually used to define the start and end positions of the crack.

[0092] S805: Determine the width of the crack according to the edge position:

[0093] in, Indicates the first f The width of the crack, Indicates the first f The edge of the crack, Indicates the first f The edge point at the other end of the crack.

[0094] It is important to note that by determining the coordinates of the crack's central axis and its brightness value, the center location and shape of the crack can be accurately identified, providing crucial geometric references for subsequent analysis. Smoothing the crack edge brightness using Cardan spline interpolation not only eliminates errors caused by noise but also ensures the accuracy of the crack edge shape. Differentiating the smoothed data allows for more precise positioning of the crack edge, leading to a more accurate calculation of the crack width. Crack width is a crucial indicator for assessing the degree of damage to concrete structures; precise width measurement helps engineers determine the impact of cracks on structural safety. The advantage of this process lies in its ability to acquire accurate crack geometric information through a highly automated method, reducing manual intervention and improving the accuracy and efficiency of detection, particularly in large-scale concrete inspections, where it significantly saves time and costs. Furthermore, the use of mathematical interpolation and derivative methods better addresses the analysis of complex cracks, providing more accurate data support for concrete quality control.

[0095] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this embodiment of the invention, a concrete quality inspection model is constructed by acquiring diverse RGB-D image sets and corresponding labels, and based on a deep neural network algorithm, combining a deep completion network module, a deep adaptive region proposal module, and a multi-resolution detection network module. Then, the preprocessed training samples are input into the concrete quality inspection model, which outputs the concrete quality inspection results. Based on the concrete labels and the quality inspection results, the concrete quality inspection model is trained in a joint manner, ensuring efficient defect identification. Finally, the RGB-D image of the concrete to be inspected is acquired and input into the trained concrete quality inspection model, achieving a comprehensive assessment of concrete quality, significantly improving detection accuracy and efficiency, and ensuring engineering safety by timely detection of potential defects.

[0096] Reference manual attached Figure 2 The diagram shows a structural schematic of a concrete quality testing system for water conservancy projects provided by an embodiment of the present invention.

[0097] The embodiment of the present application provides a concrete quality detection system 20 for hydraulic engineering, comprising a processor 201 and a memory 202; The memory 202 stores programs or instructions which can run on the processor 201, the programs or instructions are executed by the processor 201 to realize the steps of the above-mentioned concrete quality detection method for hydraulic engineering, and the same technical effects can be achieved, for the sake of avoiding repetition, the present application will not be described again.

[0098] It should be understood that the processor 201 in the embodiment of the present application can be a central processing unit (CPU), and the processor can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0099] It is also to be understood that the memory 202 in embodiments of the present application can be volatile or nonvolatile memory, or can include both volatile and nonvolatile memory. Nonvolatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrically EPROM (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), used as external cache. By way of example, and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0100] The above-described embodiments can be implemented in part or in whole through software, hardware (e.g., circuitry), firmware, or any combination thereof. When implemented in software, the above-described embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When loaded and executed by a computer, the computer instructions or computer programs can generate the flow or function according to the embodiments of the present application in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, such as from one website site, computer, server, or data center to another website site, computer, server, or data center through a wired (e.g., infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. containing a set of one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0101] It should be understood that the size of the sequence number of each process described above does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0102] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be realized in electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those of ordinary skill in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0103] Those of ordinary skill in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, apparatuses, and units can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0104] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other manners. For example, the embodiments of the apparatus described above are merely schematic. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation. For example, a plurality of units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0105] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0106] In addition, each functional unit in the various embodiments of the present application can be integrated in one processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit.

[0107] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts of the technical solutions that make contributions to the prior art, or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0108] The embodiments of the present application provide a readable storage medium, which includes: a program or instructions stored on the readable storage medium, the program or instructions are executed by a processor to implement the steps of the concrete quality detection method for concrete of hydraulic engineering, and the same technical effects can be achieved. To avoid repetition, the present application will not be described again.

[0109] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present application, but not to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. Any changes or replacements that can be easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered within the protection scope of the present application.

Claims

1. A method for quality testing of concrete used in water conservancy projects, characterized in that, include: S1: Obtain multiple training samples, wherein each training sample includes an RGB-D image set of concrete samples and a corresponding label set; S2: Preprocess each of the training samples; S3: Based on deep neural network algorithms, a concrete quality detection model is constructed by combining a deep completion network module, a deep adaptive region proposal module, and a multi-resolution detection network module. S4: Input the preprocessed training samples into the concrete quality detection model and output the quality detection results of the concrete samples; S5: Based on the label of the concrete sample and the quality test results, the concrete quality test model is trained in a combined manner; S6: Acquire the RGB-D image of the concrete to be inspected; S7: Input the RGB-D image of the concrete to be tested into the trained concrete quality inspection model, output the quality inspection result of the concrete to be tested, and complete the quality inspection of the concrete to be tested.

2. The method for testing the quality of concrete used in water conservancy projects according to claim 1, characterized in that, The preprocessing specifically includes: subtraction preprocessing; The subtraction preprocessing specifically includes: S201: The RGB-D image is smoothed using a median filter to obtain a corrected image; S202: Based on the corrected image, noise in the RGB-D image is removed through the subtraction preprocessing.

3. The method for testing the quality of concrete used in water conservancy projects according to claim 1, characterized in that, The quality inspection results include: the type and location of defects in the concrete.

4. The method for testing the quality of concrete used in water conservancy projects according to claim 1, characterized in that, S4 specifically includes: S401: The blank areas of the RGB-D images in the RGB-D image set are filled in using the depth completion network module; S402: Based on the completed RGB-D image, a depth-adaptive region proposal is generated through the depth-adaptive region proposal module; S403: Based on the depth adaptive region proposal, the quality detection result of the concrete sample is output through the multi-resolution detection network module.

5. The method for testing the quality of concrete used in water conservancy projects according to claim 4, characterized in that, Specifically, S401 includes: S4011: Calculate the gradient of the depth map in the RGB-D images of the RGB-D image set using an independent U-Net framework; S4012: Determine the surface normal of each pixel in the depth map based on the gradient; S4013: Fill in the blank areas of the RGB-D images in the RGB-D image set according to the surface normal.

6. The method for testing the quality of concrete used in water conservancy projects according to claim 4, characterized in that, Specifically, S403 includes: S4031: Adjust the size of the depth-adaptive region proposal to extract image features; S4032: Perform an upsampling operation on the image features; S4033: Based on the image features after upsampling, combine convolution and average pooling operations to output feature values; S4034: Based on the aforementioned feature values, the quality test results of the concrete sample are output through the fully connected layer.

7. The method for testing the quality of concrete used in water conservancy projects according to claim 1, characterized in that, S5 specifically includes: S501: Based on the error between the label of the concrete sample and the quality test result, determine the first loss function and the second loss function of the concrete quality test model; S502: When the function values ​​of both the first loss function and the second loss function are less than the preset loss function value, the training of the concrete quality detection model is completed.

8. The method for testing the quality of concrete used in water conservancy projects according to claim 1, characterized in that, Following S7, the following is also included: S8: When the defect category of the concrete to be tested is crack, determine the width of the crack.

9. The method for testing the quality of concrete used in water conservancy projects according to claim 8, characterized in that, S8 specifically includes: S801: Determine the coordinates of the central axis of the crack; S802: Calculate the brightness value of each point on the central axis of the crack based on the central axis coordinates; S803: The edge brightness variation of the crack is smoothed by Cardan spline interpolation. S804: Differentiate the edge brightness change of the smoothed crack to determine the edge position of the crack; S805: Determine the width of the crack based on the edge position.

10. A concrete quality testing system for hydraulic engineering projects, characterized in that, include: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the method for testing the quality of concrete for hydraulic engineering as described in any one of claims 1 to 9.

Images