Fly ash glass microbead content detection method, device, equipment and storage medium
By processing fly ash solution images using an image recognition model, the problems of low efficiency and low accuracy in fly ash glass microsphere content detection in existing technologies are solved, achieving efficient and accurate measurement of fly ash glass microsphere content.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGDONG TRANSPORTATION CONSTR ENG QUALITY INSPECTION CENT
- Filing Date
- 2026-03-09
- Publication Date
- 2026-07-14
AI Technical Summary
Existing methods for detecting the content of glass microspheres in fly ash are inefficient and have low accuracy, making it difficult to meet the requirements for high-efficiency detection.
An image recognition model was used to process fly ash solution images. Data augmentation was performed by acquiring a training sample set. Image preprocessing was carried out using multi-scale morphological operations and Gaussian mixture models. Combined with edge detection and contour fitting, the diameter of glass microspheres was identified and calculated.
This improved the accuracy and efficiency of glass microsphere content detection, enabling accurate measurement of glass microsphere content in fly ash with different particle sizes.
Smart Images
Figure CN122391064A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of fly ash technology, and in particular to a method, apparatus, computer equipment, and computer-readable storage medium for detecting the content of glass microspheres in fly ash. Background Technology
[0002] Fly ash is a byproduct of thermal power plants and is widely used as a key admixture in concrete engineering. Glass microspheres, due to their excellent properties such as "morphological effect" (improving concrete workability), "micro-aggregate effect" (increasing density), and "activity effect" (enhancing later-stage strength), are the most valuable component of fly ash. Their content directly affects the quality of fly ash and directly determines the workability, strength, and durability of concrete.
[0003] In related technologies, the detection methods for fly ash glass microspheres mainly rely on manual microscopy counting, chemical analysis, image analysis, and electron microscopy scanning, which generally suffer from low detection efficiency and low detection accuracy. Summary of the Invention
[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, and computer-readable storage medium for detecting the content of glass microspheres in fly ash, which can improve the detection efficiency and accuracy of glass microspheres.
[0005] Firstly, this application provides a method for detecting the content of glass microspheres in fly ash, the method comprising:
[0006] Acquire images of fly ash solutions; fly ash solution images are obtained by photographing fly ash solutions, which are prepared based on fly ash samples;
[0007] The fly ash solution image is input into a pre-trained image recognition model to obtain multiple glass microsphere images in the fly ash solution image;
[0008] Contour extraction is performed on the images of each glass microsphere to obtain the set of spherical contour points of each glass microsphere;
[0009] Curve fitting is performed on the pixel coordinates of each contour point in the set of contour points of each sphere to obtain the contour curve corresponding to the contour of each glass microsphere.
[0010] Based on the profile curves, the diameter of each glass microsphere is obtained, and based on the diameter of each glass microsphere, the content of fly ash glass microspheres with different particle sizes is obtained.
[0011] In one embodiment, the image recognition model includes an encoder and a decoder. Before inputting an image of fly ash solution into the pre-trained image recognition model to obtain multiple images of glass microspheres, the method further includes:
[0012] Obtain the training sample set; the training sample set includes fly ash solution sample images and corresponding label data for the fly ash solution sample images; the label data includes the location regions of glass microspheres and non-glass microspheres in the fly ash solution sample images;
[0013] Based on the training sample set, the encoder is frozen and the decoder is trained.
[0014] If the number of training iterations of the decoder exceeds the preset number, the encoder is unfrozen, and the encoder and decoder are trained simultaneously until the preset training termination condition is met, thus obtaining the pre-trained image recognition model.
[0015] In one embodiment, the training sample set is subjected to data augmentation processing; the data augmentation processing includes at least one of optical distortion processing, geometric transformation, and adding virtual reflective spots to the fly ash solution sample image.
[0016] In one embodiment, before inputting the fly ash solution image into a pre-trained image recognition model to obtain multiple glass microsphere images, the method further includes:
[0017] Image preprocessing is performed on fly ash solution images; image preprocessing includes at least one of the following: separating glass beads from background interference using multi-scale morphological operations and performing shadow correction using a Gaussian mixture model.
[0018] In one embodiment, acquiring an image of the fly ash solution includes:
[0019] The microscope stage is moved to take pictures of the slide in the stage and obtain images of the slide; fly ash solution is placed on the slide.
[0020] Edge extraction and contour detection are performed on the slide image to determine the location area of the fly ash solution on the slide;
[0021] The microscope is controlled to photograph different sub-regions within the location area of the fly ash solution to obtain images of the fly ash solution.
[0022] In one embodiment, the diameter of each glass microsphere is obtained based on each profile curve, including:
[0023] Calculate the curvature of each contour curve;
[0024] Calculate the radius of curvature of each glass microsphere based on its curvature.
[0025] The diameter of each glass microsphere is obtained based on the radius of curvature.
[0026] In one embodiment, the content of fly ash glass microspheres of different particle sizes is obtained according to the diameter of each glass microsphere, including:
[0027] The diameter of each glass microsphere is compared with the preset particle size distribution range, and the number of glass microspheres in each particle size distribution range is counted.
[0028] The number of glass microspheres within each particle size distribution range is taken as the fly ash glass microsphere content within each particle size distribution range. Secondly, this application also provides a fly ash glass microsphere content detection device, the device comprising:
[0029] The fly ash solution image acquisition module is used to acquire images of fly ash solutions; the fly ash solution images are obtained by photographing fly ash solutions, which are prepared based on fly ash samples;
[0030] The glass microsphere image acquisition module is used to input the fly ash solution image into a pre-trained image recognition model to obtain multiple glass microsphere images in the fly ash solution image;
[0031] The module for obtaining the sphere contour point set is used to extract the contour of each glass microbead image and obtain the sphere contour point set of each glass microbead.
[0032] The contour curve acquisition module is used to perform curve fitting on the pixel coordinates of each contour point in the set of contour points of each sphere to obtain the contour curve corresponding to the contour of each glass microsphere.
[0033] The glass microsphere content acquisition module is used to obtain the diameter of each glass microsphere based on each profile curve, and to obtain the content of fly ash glass microspheres with different particle sizes based on the diameter of each glass microsphere.
[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method steps of the first aspect.
[0035] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method steps of the first aspect.
[0036] The aforementioned method, apparatus, computer equipment, and computer-readable storage medium for detecting the content of glass microspheres in fly ash involve acquiring images of fly ash solutions (images obtained by photographing fly ash solutions prepared from fly ash samples); inputting these images into a pre-trained image recognition model to obtain multiple glass microsphere images within the fly ash solution images; extracting contours from each glass microsphere image to obtain a set of spherical contour points for each glass microsphere; performing curve fitting on the pixel coordinates of each contour point in each spherical contour point set to obtain a contour curve corresponding to the spherical contour of each glass microsphere; obtaining the diameter of each glass microsphere based on the contour curves; and determining the content of fly ash glass microspheres of different particle sizes based on the diameter of each glass microsphere. As can be seen from the above, this application uses an image recognition model to intelligently identify glass microspheres in fly ash solution images, obtaining multiple glass microsphere images. By extracting the contours of glass microsphere images and fitting curves to the contour points of the spheres, the detection accuracy of the glass microsphere diameter was improved, thereby improving the detection accuracy of the content of fly ash glass microspheres of different particle sizes. Attached Figure Description
[0037] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0038] Figure 1 This is a diagram illustrating the application environment of a fly ash glass microsphere content detection method in one embodiment.
[0039] Figure 2 This is a flowchart illustrating a method for detecting the content of glass microspheres in fly ash in one embodiment;
[0040] Figure 3 This is a diagram illustrating the effect of glass microsphere image recognition in one embodiment;
[0041] Figure 4 This is a schematic diagram of the training process of an image recognition model in one embodiment;
[0042] Figure 5 This is a structural block diagram of a fly ash glass microsphere content detection device in one embodiment;
[0043] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation
[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0045] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.
[0046] The fly ash glass microsphere content detection method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Terminal 102 acquires images of fly ash solution; these images are taken from fly ash solution, which is prepared based on fly ash samples. The fly ash solution images are input into a pre-trained image recognition model to obtain multiple glass microsphere images within the fly ash solution images. Contour extraction is performed on each glass microsphere image to obtain a set of spherical contour points for each glass microsphere. Curve fitting is performed on the pixel coordinates of each contour point in the set of spherical contour points to obtain the contour curve corresponding to the spherical contour of each glass microsphere. Based on each contour curve, the diameter of each glass microsphere is obtained. Based on the diameter of each glass microsphere, the content of fly ash glass microspheres with different particle sizes is obtained. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection equipment. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.
[0047] In one embodiment, such as Figure 2 As shown, a method for detecting the glass microsphere content in fly ash is provided. This embodiment applies this method to... Figure 1 Taking terminal 102 as an example, the method includes the following steps:
[0048] Step S210: Obtain an image of the fly ash solution; the fly ash solution image is obtained by taking a picture of the fly ash solution, which is prepared based on a fly ash sample.
[0049] In this embodiment, the fly ash sample is sieved, and a certain amount of fly ash sample is weighed to prepare a fly ash solution. The fly ash solution is placed in an ultrasonic disperser and shaken until the fly ash sample is homogeneous. The fly ash suspension is drawn up with a pipette and dropped onto a glass slide, a coverslip is placed on top, and the slide is gently moved until no air bubbles appear, thus uniformly dispersing the fly ash particles.
[0050] The slide is fed into the microscope's automatic stage. The magnification, horizontal and vertical movement steps are set, and the microscope automatically focuses to make the particles clear. Images are acquired in different test areas to obtain images of fly ash solution.
[0051] Step S220: Input the fly ash solution image into the pre-trained image recognition model to obtain multiple glass microsphere images in the fly ash solution image.
[0052] Among them, the image recognition model is an artificial intelligence (AI) model. An AI model refers to a system that can simulate human intelligent behavior, obtained through computer algorithms and data training. It utilizes technologies such as machine learning and deep learning, inputting a large amount of known data into the computer for training, enabling the model to automatically learn and recognize patterns and rules in the data, thereby possessing the ability to complete specific tasks.
[0053] In the embodiments of this application, such as Figure 3 As shown, a pre-trained image recognition model is used to identify glass microbeads in fly ash solution images, obtain the location regions of glass microbeads in fly ash solution images, and obtain multiple glass microbead images.
[0054] Step S230: Extract the contours of each glass microbead image to obtain a set of spherical contour points for each glass microbead.
[0055] In this embodiment, each glass microsphere image is converted to grayscale to obtain a corresponding grayscale image. Edge detection algorithms such as Canny and Sobel are used to perform edge detection on the corresponding grayscale images of each glass microsphere image, obtaining edge detection results. Based on the edge detection results, the cv2.findContours function is used to extract multiple contour points. Based on geometric features (such as roundness and area), spherical contour points are selected from the multiple contour points to obtain a set of spherical contour points for each glass microsphere.
[0056] Step S240: Perform curve fitting on the pixel coordinates of each contour point in the set of contour points of each sphere to obtain the contour curve corresponding to the contour of each glass microsphere.
[0057] In this embodiment, a polynomial fitting method is used to perform curve fitting on the pixel coordinates of each contour point in the set of contour points of each sphere to obtain the contour curve corresponding to the contour of each glass microsphere. The polynomial fitting method includes, but is not limited to, least squares method, Lagrange interpolation method, and Newton interpolation method.
[0058] Step S250: Based on each profile curve, obtain the diameter of each glass microsphere, and based on the diameter of each glass microsphere, obtain the content of fly ash glass microspheres with different particle sizes.
[0059] In this embodiment, the curvature of each profile curve is calculated using a curvature calculation formula based on the profile curve equation corresponding to each profile curve. The radius of curvature of each profile curve is then calculated based on the radius of curvature. Finally, the diameter of each glass microsphere is calculated based on the radius of curvature.
[0060] After obtaining the diameter of each glass microsphere, the number of fly ash glass microspheres of different particle sizes can be counted, and the content of fly ash glass microspheres of different particle sizes can be obtained based on the number of fly ash glass microspheres of different particle sizes.
[0061] The aforementioned method for detecting the content of glass microspheres in fly ash involves acquiring an image of a fly ash solution. This image is obtained by photographing the fly ash solution, which is prepared from a fly ash sample. The fly ash solution image is input into a pre-trained image recognition model to obtain multiple glass microsphere images within the fly ash solution image. Edge detection and contour extraction are performed on each glass microsphere image to obtain a set of spherical contour points for each glass microsphere. Curve fitting is then performed on the pixel coordinates of each contour point in the set of spherical contour points to obtain the contour curve corresponding to the spherical contour of each glass microsphere. Based on each contour curve, the diameter of each glass microsphere is obtained, and based on the diameter of each glass microsphere, the content of fly ash glass microspheres of different particle sizes is obtained. As can be seen from the above, this application uses an image recognition model to intelligently identify glass microspheres in fly ash solution images, obtaining multiple glass microsphere images. By performing edge detection and contour extraction on the glass microsphere images and curve fitting on the spherical contour points, the detection accuracy of the glass microsphere diameter is improved, thereby improving the detection accuracy of the content of fly ash glass microspheres of different particle sizes.
[0062] In one embodiment, acquiring an image of a fly ash solution includes:
[0063] Step S211: Control the movement of the microscope stage to take a picture of the slide in the stage and obtain an image of the slide; fly ash solution is placed on the slide.
[0064] In this embodiment, the stage of the microscope is controlled to move according to a preset step size and a preset direction. During the movement, the eyepiece of the microscope is controlled to take pictures of the slide in the stage to obtain an image of the slide.
[0065] Step S212: Perform contour detection on the slide image to determine the location area of the fly ash solution on the slide.
[0066] In this embodiment, the slide image is first subjected to Gaussian filtering for noise reduction, and then the Canny algorithm is used to extract edges. Candidate regions matching the area and shape criteria are filtered through contour search, and sliding windows are generated near the candidate regions to extract shape or color features. A lightweight classifier is used to quickly determine whether it is a fly ash sample, reducing the computational cost of full-image search and obtaining the location region of the fly ash solution on the slide.
[0067] Step S213: Control the microscope to take pictures of different sub-regions in the location area of the fly ash solution to obtain fly ash solution images.
[0068] In this embodiment, the stage can be controlled to rotate according to a preset rotation step size. During the rotation, the microscope is controlled to capture images of different sub-regions within the location area of the fly ash solution, thereby obtaining multiple fly ash solution images. For example, the location area of the fly ash solution includes sub-region A, sub-region B, and sub-region C. Sub-region A, sub-region B, and sub-region C are captured separately to obtain three different fly ash solution images.
[0069] This application embodiment uses the sliding window method and edge contour detection method to locate the position area of the fly ash suspension on the glass slide, ensuring that subsequent imaging is always carried out within the fly ash sample area.
[0070] In one embodiment, before inputting the fly ash solution image into a pre-trained image recognition model to obtain multiple glass microsphere images, the method further includes:
[0071] Step S221: Perform image preprocessing on the fly ash solution image; image preprocessing includes at least one of the following: using multi-scale morphological operations to separate glass beads from background interference and using a Gaussian mixture model for shadow correction.
[0072] In this embodiment of the application, considering that glass beads may adhere to the background or have blurred boundaries in fly ash sample images, performing multi-scale morphological operations (such as expansion, erosion, opening, and closing operations of structural elements of different sizes) can analyze and process the image from multiple scales, highlight the shape features of the glass beads, reduce background interference, and make the glass microbeads clearer and more distinguishable in the image.
[0073] Considering that shadows may occur during image acquisition due to uneven lighting, and various noises (such as Gaussian noise and salt-and-pepper noise) may also exist in the image, a Gaussian mixture model can be used to model the pixel distribution of the image, dividing the pixels in the image into different Gaussian distribution components. By calculating the parameters of the Gaussian distribution components (such as mean and variance), noise can be effectively removed, and shadow areas can be corrected, making the image brightness more uniform.
[0074] In one embodiment, such as Figure 4 As shown, the image recognition model includes an encoder and a decoder. Before inputting the fly ash solution image into the pre-trained image recognition model to obtain multiple glass microsphere images, the method further includes:
[0075] Step S222: Obtain the training sample set; the training sample set includes fly ash solution sample images and corresponding label data for the fly ash solution sample images; the label data includes the location regions of glass microspheres and non-glass microspheres in the fly ash solution sample images.
[0076] The network structure of the image recognition model is an improved U-Net++ architecture, consisting of an encoder and a decoder connected by a series of nested dense convolutional blocks.
[0077] In this embodiment of the application, images of fly ash solution samples from different origins and batches are collected, and two types of targets, "glass microbeads" and "non-glass microbeads", are manually labeled to construct label data.
[0078] Step S223: Freeze the encoder and train the decoder based on the training sample set.
[0079] In this embodiment, the image recognition model is trained in two stages. In the first training stage, the encoder is frozen and the decoder is trained up to a preset number of times (e.g., 100 epochs).
[0080] Step S224: If the number of training iterations of the decoder is greater than the preset number, the encoder is unfrozen, and the encoder and decoder are trained simultaneously until the preset training termination condition is met, and a pre-trained image recognition model is obtained.
[0081] The preset training result conditions may include at least one of the following: the number of training iterations reaches a preset threshold, or the loss function value converges.
[0082] In this embodiment of the application, during the second training phase, the encoder is unfrozen, and the encoder and decoder are fine-tuned and trained to a preset number of times (e.g., 50 epochs) until the preset training termination condition is met, thereby obtaining a pre-trained image recognition model.
[0083] The embodiments of this application reduce the complexity of model training, improve the model convergence speed, avoid overfitting, and improve the robustness of the model by training the encoder and decoder in stages.
[0084] In one embodiment, the training sample set is subjected to data augmentation processing; the data augmentation processing includes at least one of optical distortion processing, geometric transformation, and adding virtual reflective spots to the fly ash solution sample image.
[0085] Among them, optical distortion processing includes at least one of color temperature deviation and blur processing, which can simulate the image acquisition environment under different lighting conditions.
[0086] Geometric transformations include at least one of rotation, translation, and scaling, which can change the spatial structure of an image.
[0087] Adding virtual reflective spots can simulate physical phenomena such as light spot interference in the real environment.
[0088] In this embodiment, by performing optical distortion processing on the fly ash solution sample image, the image recognition model can learn image features under various lighting conditions, thereby improving its adaptability to different lighting environments. By performing geometric transformations on the fly ash solution sample image, the model can learn the image's invariance characteristics under different spatial transformations, thereby improving its robustness to image spatial transformations. By adding virtual reflective spots to the fly ash solution sample image, the model's anti-interference ability in complex environments is improved, ensuring the model's stability and reliability.
[0089] In one embodiment, obtaining the diameter of each glass microsphere based on each profile curve includes:
[0090] Step S251: Calculate the curvature of each contour curve.
[0091] In this embodiment of the application, for the parametric curve equations x=x(t) and y=y(t) of the contour curve, the curvature calculation formula is as follows:
[0092]
[0093] For the curve equation y=f(x) of the contour curve, the formula for calculating the curvature is:
[0094]
[0095] Step S252: Calculate the radius of curvature of each glass microsphere based on the curvature.
[0096] In this embodiment, the formula for calculating the radius of curvature is:
[0097]
[0098] Step S253: Obtain the diameter of each glass microsphere based on the radius of curvature.
[0099] In this embodiment, the diameter of each glass microsphere is:
[0100]
[0101] In one embodiment, the content of fly ash glass microspheres is obtained based on the diameter of each glass microsphere, including:
[0102] Step S254: Compare the diameter of each glass microsphere with the preset particle size distribution range, and count the number of glass microspheres in each particle size distribution range.
[0103] Step S255: The number of glass microspheres within each particle size distribution range is taken as the fly ash glass microsphere content within each particle size distribution range.
[0104] The preset particle size distribution ranges can be set according to actual needs. For example, the preset particle size distribution ranges are 0~10um, 10um~20um, 20um~30um, ...
[0105] In this embodiment, the diameter of each glass microsphere is compared with a preset particle size distribution range to classify each glass microsphere into its corresponding particle size distribution range, and the number of glass microspheres within each particle size distribution range is counted. The number of glass microspheres within each particle size distribution range is taken as the fly ash glass microsphere content within each particle size distribution range.
[0106] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.
[0107] Based on the same inventive concept, this application also provides a fly ash glass microsphere content detection device for implementing the fly ash glass microsphere content detection method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more fly ash glass microsphere content detection device embodiments provided below can be found in the limitations of the large model inference request processing method above, and will not be repeated here.
[0108] In one exemplary embodiment, such as Figure 5 As shown, a device for detecting the content of glass microspheres in fly ash is provided. The device includes:
[0109] The fly ash solution image acquisition module 510 is used to acquire fly ash solution images; the fly ash solution images are obtained by photographing the fly ash solution, which is prepared based on fly ash samples;
[0110] The glass microsphere image acquisition module 520 is used to input the fly ash solution image into a pre-trained image recognition model to obtain multiple glass microsphere images in the fly ash solution image;
[0111] The spherical contour point set acquisition module 530 is used to extract the contours of each glass microbead image and obtain the spherical contour point set of each glass microbead.
[0112] The contour curve acquisition module 540 is used to perform curve fitting on the pixel coordinates of each contour point in the set of contour points of each sphere to obtain the contour curve corresponding to the contour of each glass microsphere.
[0113] The glass microsphere content acquisition module 550 is used to obtain the diameter of each glass microsphere based on each profile curve; and to obtain the content of fly ash glass microspheres of different particle sizes based on the diameter of each glass microsphere.
[0114] In one embodiment, the image recognition model includes an encoder and a decoder. Before inputting an image of fly ash solution into the pre-trained image recognition model to obtain multiple images of glass microspheres, the method further includes:
[0115] Obtain the training sample set; the training sample set includes fly ash solution sample images and corresponding label data for the fly ash solution sample images; the label data includes the location regions of glass microspheres and non-glass microspheres in the fly ash solution sample images;
[0116] Based on the training sample set, the encoder is frozen and the decoder is trained.
[0117] If the number of training iterations of the decoder exceeds the preset number, the encoder is unfrozen, and the encoder and decoder are trained simultaneously until the preset training termination condition is met, thus obtaining the pre-trained image recognition model.
[0118] In one embodiment, the training sample set is subjected to data augmentation processing; the data augmentation processing includes at least one of optical distortion processing, geometric transformation, and adding virtual reflective spots to the fly ash solution sample image.
[0119] In one embodiment, before inputting the fly ash solution image into a pre-trained image recognition model to obtain multiple glass microsphere images, the method further includes:
[0120] Image preprocessing is performed on fly ash solution images; image preprocessing includes at least one of the following: separating glass beads from background interference using multi-scale morphological operations and performing shadow correction using a Gaussian mixture model.
[0121] In one embodiment, acquiring an image of the fly ash solution includes:
[0122] The microscope stage is moved to take pictures of the slide in the stage and obtain images of the slide; fly ash solution is placed on the slide.
[0123] Contour detection is performed on the slide image to determine the location area of the fly ash solution on the slide;
[0124] The microscope is controlled to photograph different sub-regions within the location area of the fly ash solution to obtain images of the fly ash solution.
[0125] In one embodiment, the diameter of each glass microsphere is obtained based on each profile curve, including:
[0126] Calculate the curvature of each contour curve;
[0127] Calculate the radius of curvature of each glass microsphere based on its curvature.
[0128] The diameter of each glass microsphere is obtained based on the radius of curvature.
[0129] In one embodiment, the content of fly ash glass microspheres of different particle sizes is obtained according to the diameter of each glass microsphere, including:
[0130] The diameter of each glass microsphere is compared with the preset particle size distribution range, and the number of glass microspheres in each particle size distribution range is counted.
[0131] The number of glass microspheres within each particle size distribution range is taken as the fly ash glass microsphere content within each particle size distribution range.
[0132] Each module in the aforementioned fly ash glass microsphere content detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.
[0133] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows. Figure 6 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores fly ash glass microsphere content detection data. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for detecting fly ash glass microsphere content.
[0134] Those skilled in the art will understand that Figure 6 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the above-described method for detecting the content of fly ash glass microspheres. The steps of the fly ash glass microsphere content detection method described here can be the steps of the fly ash glass microsphere content detection method of the various embodiments described above.
[0135] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, causes the processor to perform the steps of the fly ash glass microsphere content detection method described above. The steps of the fly ash glass microsphere content detection method described here can be the steps from the fly ash glass microsphere content detection methods of the various embodiments described above.
[0136] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, causes the processor to perform the steps of the fly ash glass microsphere content detection method described above. The steps of the fly ash glass microsphere content detection method described here can be the steps in the fly ash glass microsphere content detection method of the various embodiments described above.
[0137] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.
[0138] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.
[0139] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0140] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method for detecting the content of glass microspheres in fly ash, characterized in that, The method includes: Acquire images of fly ash solutions; the fly ash solution images are obtained by photographing fly ash solutions, which are prepared based on fly ash samples; The fly ash solution image is input into a pre-trained image recognition model to obtain multiple glass microsphere images in the fly ash solution image; Contour extraction is performed on the images of each glass microsphere to obtain a set of spherical contour points for each glass microsphere; Curve fitting is performed on the pixel coordinates of each contour point in each set of sphere contour points to obtain the contour curve corresponding to each glass microsphere contour. Based on the aforementioned profile curves, the diameter of each glass microsphere is obtained, and based on the diameter of each glass microsphere, the content of fly ash glass microspheres with different particle sizes is obtained.
2. The method according to claim 1, characterized in that, The image recognition model includes an encoder and a decoder. Before inputting the fly ash solution image into the pre-trained image recognition model to obtain multiple glass microsphere images, the method further includes: Obtain a training sample set; the training sample set includes fly ash solution sample images and corresponding label data for the fly ash solution sample images; the label data includes the location regions of glass microspheres and non-glass microspheres in the fly ash solution sample images; Based on the training sample set, the encoder is frozen, and the decoder is trained. If the number of training iterations of the decoder exceeds a preset number, the encoder is unfrozen, and the encoder and the decoder are trained simultaneously until the preset training termination condition is met, thereby obtaining the pre-trained image recognition model.
3. The method according to claim 2, characterized in that, The method further includes: The training sample set is subjected to data augmentation processing; the data augmentation processing includes at least one of optical distortion processing, geometric transformation, and adding virtual reflective spots to the fly ash solution sample image.
4. The method according to claim 1, characterized in that, Before inputting the fly ash solution image into a pre-trained image recognition model to obtain multiple glass microsphere images, the method further includes: The image of the fly ash solution is preprocessed; the image preprocessing includes at least one of the following: using multi-scale morphological operations to separate glass beads from background interference and using a Gaussian mixture model for shadow correction.
5. The method according to claim 1, characterized in that, The acquisition of fly ash solution images includes: The microscope stage is moved to take pictures of the slide in the stage and obtain an image of the slide; the slide contains fly ash solution. Contour detection is performed on the slide image to determine the location region of the fly ash solution on the slide; The microscope is controlled to photograph different sub-regions in the location area of the fly ash solution to obtain images of the fly ash solution.
6. The method according to any one of claims 1 to 5, characterized in that, The step of obtaining the diameter of each glass microsphere based on each of the aforementioned contour curves includes: Calculate the curvature of each of the aforementioned contour curves; Calculate the radius of curvature of each glass microsphere based on the curvature. The diameter of each glass microsphere is obtained based on the radius of curvature.
7. The method according to any one of claims 1 to 5, characterized in that, The process of obtaining the content of fly ash glass microspheres with different particle sizes based on the diameter of each glass microsphere includes: The diameter of each glass microsphere is compared with the preset particle size distribution range, and the number of glass microspheres in each particle size distribution range is counted. The number of glass microspheres within each particle size distribution range is taken as the fly ash glass microsphere content within each particle size distribution range.
8. A device for detecting the content of glass microspheres in fly ash, characterized in that, The device includes: A fly ash solution image acquisition module is used to acquire images of fly ash solutions; the fly ash solution images are obtained by photographing fly ash solutions, and the fly ash solutions are prepared based on fly ash samples; A glass microsphere image acquisition module is used to input the fly ash solution image into a pre-trained image recognition model to obtain multiple glass microsphere images in the fly ash solution image; The sphere contour point set acquisition module is used to extract the contours of each glass microsphere image to obtain the sphere contour point set of each glass microsphere. The contour curve acquisition module is used to perform curve fitting on the pixel coordinates of each contour point in the set of contour points of each sphere to obtain the contour curve corresponding to the contour of each glass microsphere. The glass microsphere content obtaining module is used to obtain the diameter of each glass microsphere according to the respective contour curves, and to obtain the content of fly ash glass microspheres of different particle sizes according to the diameter of each glass microsphere.
9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.