Material microstructure quantitative characterization method coupled with image enhancement and machine learning

Abstract

The invention belongs to the technical field of material microstructure state test characterization, and particularly provides a material microstructure quantitative characterization method coupling image enhancement and machine learning, which comprises the following steps: obtaining an observation sample microstructure image set through a microscopic observation method; carrying out image enhancement, multiple noise reduction and binarization processing on the microstructure image set; extracting features of the binarized image to a high-dimensional hidden space by using a deep convolutional neural network algorithm, and then performing dimension reduction and quantitative characterization on the high-dimensional features by using a consistent manifold algorithm; and carrying out physical interpretability analysis on the dimension reduction result of the microstructure characteristics of the material, and evaluating the characterization effect of the dimension reduction quantitative index on the microstructure characteristics of the material. According to the method, unified description and quantitative characterization of various complex microstructure morphology, size and evolution of the material are achieved, the cost and errors of a manual statistical method are greatly reduced, and the method has high invention innovation and good engineering application prospects.

Description

Technical Field

[0001] The present invention belongs to the technical field of material microstructure state testing and characterization, and specifically provides a material microstructure quantitative characterization method that couples image enhancement and machine learning. Background Art

[0002] As carriers of a material's intrinsic properties and states, material microstructure images contain a rich microscopic landscape and direct physical information related to its deformation, failure, and damage mechanisms. The complex microstructural characteristics of a material directly determine its macroscopic mechanical properties. Therefore, identifying and quantifying microstructural features from microstructure images is crucial for material failure analysis and multiscale mechanical modeling.

[0003] With the further development of modern industry, materials undergo microstructural evolution during actual service, primarily manifesting as time-varying characteristics in microstructural morphology and size, which in turn influence the alloy's mechanical properties. Nickel-based superalloys, widely used in the manufacture of high-temperature components for advanced aeroengines and gas turbines, often exhibit a coherent two-phase structure: a reinforcement phase and a matrix phase. Therefore, it is necessary to develop methods for extracting and quantifying the microstructural characteristics of such two-phase alloys.

[0004] The traditional method for quantitatively characterizing the two-phase microstructure of alloys is to import alloy microscopic images into image processing software such as Image J and use the software's built-in threshold segmentation and statistical functions to calculate the two-phase structural parameters. However, the results of software characterization rely heavily on the operator's experience and understanding, which is highly subjective and requires high labor and material resources. Therefore, establishing a scientific and efficient method for quantitatively characterizing the two-phase microstructure of materials is a key issue that urgently needs to be addressed in the fields of alloy design, preparation, and operation and maintenance.

[0005] In recent years, machine learning methods within a data-driven paradigm have attracted widespread attention in fields such as microscopic image characterization, material optimization and design, and damage identification. The two-point correlation algorithm is one of the commonly used algorithms for microstructure feature extraction. Correlation maps can be used to directly obtain a statistical description of the microstructure feature size, or indirectly obtain a reduced-dimensional description of the microstructure by combining it with dimensionality reduction methods. Furthermore, deep learning algorithms, represented by convolutional neural networks, have become one of the core technologies in the fields of image processing and computer vision. Convolutional neural networks can identify high-dimensional structural information and low-dimensional pixel information from images and are widely used in image recognition, segmentation, and dimensionality reduction. They have broad application prospects in the field of material microstructure characterization.

[0006] Chinese patent CN111738130B discloses a quantitative characterization method for rafting in nickel-based single crystal alloys based on a two-point correlation algorithm and principal component analysis (PCA). This method extracts alloy microstructural dimensional parameters by analyzing the spatial variation of the cross-correlation function. PCA is then used to quantitatively identify the evolution of the rafting morphology and dimensional parameters, enabling the characterization of the material's microstructure in batches. However, the developed algorithm fails to uniformly consider the extraction and quantitative characterization of complex microstructural features, such as alloy orientation characteristics, isotropic coarsening, and anisotropic rafting. Furthermore, the effective enhancement and processing of raw microscopic SEM images using advanced image algorithms remains a challenge. Summary of the Invention

[0007] In order to solve the above technical problems, the present invention provides a quantitative characterization method for material microstructure by coupling image enhancement and machine learning. Based on the developed Gaussian distribution adaptive threshold segmentation method, deep convolutional neural network algorithm (deep CNN) and uniform manifold dimensionality reduction algorithm (UMAP algorithm), the complex microstructure characteristics of the two phases of the alloy are quantitatively characterized by performing high-quality binarization processing on the microstructure image, extracting high-dimensional features of the image, and performing dimensionality reduction quantitative characterization.

[0008] The present invention is implemented by providing a method for quantitative characterization of material microstructure by coupling image enhancement and machine learning, comprising the following steps: S1: Prepare an observation sample of the target alloy material and obtain a microstructure image set of the observation sample by microscopic observation method; S2: Image enhancement, multiple noise reduction and binarization processing are performed on the microscopic tissue image set based on Gaussian distribution adaptive threshold segmentation, mask processing and adaptive histogram equalization; S3: Using a deep CNN algorithm to extract the features of the microstructure image set processed in step S2 into a high-dimensional latent space, obtaining high-dimensional latent space features, and then using the Unified Mapping (UMAP) algorithm to reduce the dimensionality and quantitatively characterize the high-dimensional latent space features, thereby obtaining a dimensionality-reduced quantitative characterization of the target alloy material microstructure; S4: Conduct physical interpretability analysis of the dimensionality reduction quantitative characterization of the target alloy material microstructure, and evaluate the effect of dimensionality reduction quantitative indicators on the quantitative characterization of the target alloy material microstructure.

[0009] Preferably, in step S1, sample blocks are cut from target alloy materials with different service histories or pre-treated materials, and are prepared into microscopic observation samples after mechanical sampling or fine slicing; and images of different microstructural states of the observation samples are captured using a microscopic observation method to form a microstructural image set including various microstructural features of the target alloy materials.

[0010] More preferably, in step S1, different service histories or pre-treatments refer to target alloy materials that have undergone service histories or pre-tests at different temperatures, times and stresses.

[0011] Further preferably, in step S1, the microstructure image set includes microstructure images of different microstructure morphologies and sizes of the target alloy material.

[0012] Preferably, the target alloy material is a two-phase metal alloy material, and step S2 takes a microstructure image set as input and specifically comprises the following steps: S2.1: Gaussian noise reduction and Otsu method binarization: A Gaussian filter is used to remove noise from the microstructure image set. The Otsu method is used to perform preliminary binarization on the denoised microstructure image set. In the black-and-white binary image, a pixel value of 1 represents Phase I of the target alloy material, and a pixel value of 0 represents Phase II of the target alloy material. S2.2: Mask enhancement processing: The microstructure image set after binarization by Otsu method is used as a mask for mask processing, which is applied to the microstructure image set after Gaussian noise reduction to make the phase boundary areas in the material microstructure more prominent, thereby obtaining a mask image set; S2.3: Adaptive Histogram Equalization: The mask image set is processed using an adaptive histogram equalization method so that the grayscale histogram is nonlinearly mapped from the foreground range to the entire grayscale range, thereby enhancing the contrast between different phases in the target alloy material and obtaining a balanced mask image set. S2.4: Histogram Figure 2 Second-order Gaussian distribution fitting: The grayscale histogram of the mask image set after equalization is a bimodal distribution. The second-order Gaussian distribution is used to fit the grayscale histogram, and the grayscale corresponding to the left peak is obtained, which mainly corresponds to phase II. S2.5: Image adaptive threshold binarization: Use the grayscale corresponding to the left peak of the second-order Gaussian distribution in step S2.4 as the threshold, and use a fixed threshold function to set pixels in the image below the threshold to 0 and pixels above the threshold to 1, thereby achieving the final binarization of the equalized mask image set; S2.6: Opening operation denoising and connected domain denoising: Based on the opening operation denoising algorithm, the discrete points near phase I are expanded and eroded by pixels, while the boundary of phase I after binarization is smoothed. The connected domain denoising method is used to eliminate discrete foreground points scattered in phase II and discrete pixel blocks that affect image quality. Finally, a set of microstructure images is obtained after image enhancement, multiple denoising and binarization processing.

[0013] Further preferably, in step S2.3, the mask processing can mark most of phase II as 0, highlighting phase I and the boundary area between the two phases, so as to facilitate further processing of the image and observation of image details in the area of ​​interest.

[0014] Preferably, step S3 takes the microscopic tissue image set after image enhancement, multiple noise reduction and binarization processing in step S2 as input, and specifically includes the following steps: S3.1: Deep CNN algorithm selection and structure design: Select a deep CNN algorithm, remove the classification layer, retain the convolutional layer and pooling layer, and add a linear fully connected layer to average pool and flatten the output of the deep CNN algorithm; S3.2: Image preprocessing and cropping: Based on the microscopic tissue image set obtained in step S2, which has undergone image enhancement, multiple noise reduction, and binarization, select an image cropping algorithm to crop the microscopic tissue image set to the input dimension of the deep CNN algorithm selected in step S3.1. The cropped microscopic tissue image set is then expanded to three channels to form the deep CNN algorithm input image set. S3.3: Image feature extraction and mapping: The deep CNN algorithm input image set in step S3.2 is input into the deep CNN algorithm one by one to extract the microscopic tissue image features and map them to a high-dimensional latent space to obtain the high-dimensional latent space features of the microscopic tissue image set; S3.4: UMAP algorithm parameter setting: set the upper limit and lower limit parameters of the number of local connections of the UMAP algorithm data points; S3.5: Learning the manifold structure of the high-dimensional latent space: Input the high-dimensional latent space features of the microscopic tissue image set obtained in step S3.3 into the UMAP algorithm model to construct a weighted neighborhood graph and a connected neighborhood graph, thereby learning the manifold structure of the microscopic tissue image set in the high-dimensional latent space; S3.6: Optimal low-dimensional manifold representation and mapping: Project the approximate manifold structure of the high-dimensional latent space into the low-dimensional space, convert the variable distance of the approximate manifold structure into the standard Euclidean distance of the low-dimensional global coordinate system, take distance invariance as the optimization goal, and use cross-entropy as the loss function. By adjusting the upper limit and lower limit of the number of local connections of data points in the UMAP algorithm, the optimal low-dimensional manifold representation is found, and finally a dimensionality-reduced quantitative representation of the microscopic tissue image set is obtained.

[0015] Further preferably, in step S3.2, a single original image in the microscopic tissue image set is cropped into multiple sub-images, each sub-image containing two-phase features of several cycle widths to retain the basic features of the original image.

[0016] Preferably, in step S4, the UMAP components after dimensionality reduction by the UMAP algorithm in step S3 are used as quantitative characterization parameters of the microstructure characteristics of the target alloy material, and physical interpretability analysis of the dimensionality reduction results is carried out; global UMAP topological space mapping and local UMAP topological space mapping are performed on the microstructure image set to evaluate the characterization effect of the dimensionality reduction quantitative indicators, thereby achieving a comprehensive quantitative characterization of the microstructure of the target alloy material.

[0017] Preferably, in step S1 , an optical microscope, a scanning electron microscope, or a transmission electron microscope is used to perform microscopic observation of the sample.

[0018] Preferably, steps S2-S4 are implemented by programming in Python language or Java language or commercial computing software Matlab.

[0019] Compared with the prior art, the advantages of the present invention are: 1. The present invention develops an image binarization method based on Gaussian distribution adaptive threshold segmentation, which realizes high-quality and high-fidelity enhancement and binarization processing of the original microscopic image of the material; 2. The present invention uses an efficient deep CNN algorithm to extract and map features from microstructure images, scientifically and efficiently achieving rapid characterization of the alloy's two-phase microstructure, greatly reducing the cost and error of manual statistical methods; 3. The present invention uses the advanced UMAP algorithm to reduce the dimensionality of high-dimensional image features in the latent space, successfully obtaining quantitative indicators of the alloy microstructure characteristics after dimensionality reduction, and realizing a unified description and quantitative characterization of various complex microstructure characteristics of the alloy and their evolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flow chart of the method provided by the present invention on two-phase materials; Figure 2 The original SEM images of a nickel-based single crystal superalloy, (a) is a positive cubic morphology, and (b) is a positive bamboo raft morphology; Figure 3 The SEM mask images after preliminary binarization using Gaussian noise reduction and Otsu's method, (a) is the positive cubic morphology, and (b) is the positive bamboo raft morphology; Figure 4 is the grayscale histogram after adaptive histogram equalization and its second-order Gaussian distribution fitting result; Figure 5 The final SEM mask image after segmentation based on Gaussian distribution adaptive thresholding, (a) is the positive cube morphology, (b) is the positive bamboo raft morphology; Figure 6 It is the operation process of image pre-processing and deep CNN algorithm; Figure 7 The distribution of various microstructures of nickel-based single crystal superalloys in the global UMAP topological space; Figure 8 The specific microstructural morphology of nickel-based single crystal high-temperature alloy is distributed in the local UMAP topological space. DETAILED DESCRIPTION

[0021] To make the objectives, technical solutions, and advantages of the present invention more clearly understood, the following, in conjunction with the accompanying drawings, uses the quantitative characterization of complex microstructural features of a nickel-based single crystal superalloy as an example to further illustrate specific embodiments of the present invention. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0022] The present invention provides a method for quantitative characterization of material microstructure by coupling image enhancement and machine learning, see Figure 1 , which is a simplified flow chart for quantitatively characterizing complex microstructural features of a nickel-based high-temperature alloy provided by an embodiment of the present invention. The specific steps are: S1: Prepare observation samples of target alloy materials and obtain their microstructure image sets through microscopic observation methods; S2: Based on Gaussian distribution adaptive threshold segmentation, mask processing, and adaptive histogram equalization, the original microscopic tissue image set is subjected to image enhancement, multiple noise reduction, and binarization processing; S3: Use the deep CNN algorithm to extract the features of the binary image into a high-dimensional latent space, and then use the UMAP algorithm to reduce the dimensionality and quantitatively characterize the high-dimensional latent space features; S4: Conduct physical interpretability analysis of the dimensionality reduction results of material microstructure characteristics, and evaluate the effectiveness of dimensionality reduction quantitative indicators on the characterization of material microstructure characteristics.

[0023] In a specific embodiment, the target alloy material in step S1 is selected as a nickel-based single crystal high-temperature alloy.

[0024] In a specific embodiment, the microscopic observation means in step S1 is a scanning electron microscope.

[0025] In a specific embodiment, the deep CNN algorithm described in step S3 is selected as the DenseNet121 network.

[0026] In a specific embodiment, the quantitative analysis described in step S4 uses the first two order UMAP components as dimensionality reduction characterization parameters.

[0027] In a specific embodiment, steps S2-S4 are implemented by programming in Python.

[0028] To more clearly demonstrate the application of the technical approach of the present invention, the present invention has applied a collection of microscopic SEM images of materials. Taking a selected nickel-based single crystal superalloy as an example, the above-mentioned technical approach was used to quantitatively characterize its complex microstructural features.

[0029] S1. Material preparation and microstructure image acquisition, including the following three steps: (1) Based on pretreatment tests at different temperatures, stresses, and times, nickel-based single crystal high-temperature alloy parent materials with different microstructures were obtained. 1 cm × 1 cm × 1 cm sample blocks were cut from the original samples using a wire cutting method. Metallographic specimens were prepared after mechanical grinding, polishing, chemical etching, cleaning, and drying. The chemical etching solution ratio was: 50 mL HCl + 100 mL H2O + 5 mL H2SO4 + 20 g CuSO4; (2) SEM grayscale images of alloys with different microstructural states were taken using a scanning electron microscope (SEM), with a fixed magnification of 10,000 times, the alloy crystallographic orientation being

[001] or

[011] , and the original image size being 3072 × 2304. The scanning electron microscope was a ZEISS SUPRA55 field emission scanning electron microscope. (3) A microscopic SEM image set containing various microstructural features of the alloy is formed. The micromorphology includes positive cubic morphology, bamboo raft morphology, needle morphology and oblique cubic morphology, bamboo raft morphology. The microscopic size includes different degrees of evolution from cubic morphology to bamboo raft and needle morphology, totaling 40 microstructural state characteristics. Figure 2 (a) (b) show the microscopic SEM images of the standard cubic morphology and the bamboo raft-like morphology. It can be seen that the microstructure contains a dark reinforcement phase and a light matrix phase.

[0030] S2. Microscopic image binarization and enhancement based on Gaussian distribution adaptive threshold segmentation. The original microscopic tissue image is used as input and includes the following six steps: S2.1: Gaussian noise reduction and Otsu method binarization: Use Gaussian filter to process the noise in the microstructure image to make the image clearer. Use Otsu method to perform preliminary binarization on the denoised image. In the black and white binary image, the black pixel value of 1 represents the strengthening phase of the alloy, and the white pixel value of 0 represents the matrix phase of the alloy. The preliminary binarized mask image is as follows Figure 3 As shown in (a) and (b), labels 1 to 5 show the labeling errors and noise on the two-phase boundary of the image after preliminary binarization; S2.2: Mask Enhancement: Using the Otsu binarized image as a mask, this is applied to the Gaussian noise-reduced image to highlight the phase boundaries within the microstructure. Masking can mark most of the matrix phase as zero, highlighting the enhanced phase and the boundary between two phases, facilitating further image processing and observing image details in areas of interest. S2.3: Adaptive histogram equalization: The mask image is processed using the adaptive histogram equalization method so that the grayscale histogram is nonlinearly mapped from the foreground interval to the entire grayscale range, thereby enhancing the contrast between different phases in the alloy. The grayscale histogram after mapping is as follows: Figure 4 As shown; S2.4: Histogram Figure 2 Second-order Gaussian distribution fitting: The grayscale histogram of the mask image after equalization can be regarded as a bimodal distribution. The second-order Gaussian distribution is used to fit the grayscale histogram to obtain the grayscale corresponding to the left peak, which mainly corresponds to the white matrix phase, such as Figure 4 The figure shows the fitting result of the second-order Gaussian distribution, where the left and right peaks correspond to the two-phase tissues respectively; S2.5: Image adaptive threshold binarization: Use the grayscale corresponding to the left peak of the second-order Gaussian distribution in step S2.4 as the threshold, and use a fixed threshold function to set pixels in the image below the threshold to 0 and pixels above the threshold to 1, thus achieving the final binarization of the image; S2.6: Opening operation denoising and connected domain denoising: Based on the opening operation denoising algorithm, the expansion and erosion of pixels in the image are used to eliminate discrete points near the black phase I, and at the same time, the boundary of the phase I after binarization is smoothed; the connected domain denoising method is used to eliminate discrete foreground points scattered in phase II and discrete pixel blocks that affect image quality. The final processed mask image is as follows Figure 5 As shown in (a) and (b), after image enhancement and multiple noise reduction, the image at markers 1' to 5' has been effectively removed and suppressed after the final binarization. The final output of step S2 is the microscopic tissue image after image enhancement, multiple noise reduction, and binarization.

[0031] S3. Use a deep convolutional neural network and the consistent manifold algorithm to extract features and quantitatively characterize the microscopic images. This method uses the enhanced microscopic tissue images from step S2 as input. The method includes the following six steps: S3.1: Deep CNN Selection and Structure Design: Select the DenseNet121 network as the deep CNN for this example. Remove its classification layers and retain only the convolutional and pooling layers. Add a linear fully connected layer to average pool and flatten the output of the DenseNet121 network. S3.2: Image Preprocessing and Cropping: Based on the binary images of the alloy microstructure obtained in step S2, select an appropriate image cropping algorithm to crop the original image (3072×2304) of each microstructural state to the DenseNet121 network input dimension (512×512), and expand the cropped image to 3 channels. This embodiment uses a random cropping method to randomly crop 25 512×512 subimages from each original image, resulting in a total of 40×25=1000 images, forming the CNN input image set. S3.3: Image feature extraction and mapping: Input the image set in step S3.2 into the deep CNN one by one, extract the microscopic image features based on the DenseNet121 network and map them to the latent space. The operation process of the deep CNN is as follows: Figure 6 As shown; S3.4: UMAP model parameter setting: Set the core parameters of the UMAP algorithm, that is, set the upper limit and lower limit of the number of local connections of data points. In this embodiment, the upper limit is set to 10 and the lower limit is set to 1; S3.5: Learning the manifold structure in high-dimensional space: Input the high-dimensional latent space features of the microstructure image set obtained in step S3.3 into the UMAP model to construct a weighted neighborhood graph and a connected neighborhood graph of the dataset, thereby learning the manifold structure of the dataset in high-dimensional space; S3.6: Optimal Low-Dimensional Manifold Representation and Mapping: Project the approximate manifold in the high-dimensional space into the low-dimensional space. Convert the variable distances in the manifold space to standard Euclidean distances in the low-dimensional global coordinate system. With distance invariance as the optimization objective and cross-entropy as the loss function, adjust the core algorithm parameters to find the optimal low-dimensional manifold representation. Given the powerful nonlinear dimensionality reduction capabilities of the UMAP algorithm, this embodiment selects a 2-dimensional dimension after dimensionality reduction.

[0032] The final output of step S3 is the dimensionality reduction mapping of the data set and the quantitative characterization of the microstructure.

[0033] S4. Physical interpretation and quantitative analysis of material microstructural characteristics: The first two UMAP components (UMAP-1 and UMAP-2) after the UMAP dimensionality reduction in step S3 are used as quantitative characterization parameters of the material's complex microstructural characteristics. The physical interpretability analysis of the dimensionality reduction results is performed.

[0034] A total of 1000 images of 40 kinds of alloy microstructures were characterized by global dimensionality reduction. Figure 7 The dimensionality reduction results for various microstructures of a nickel-based single-crystal superalloy in the global UMAP topological space are presented, with the horizontal and vertical coordinates representing the UMAP-1 and UMAP-2 components, respectively. Similar microstructural morphologies are found clustered in localized regions of the UMAP space: Region A exhibits positive needle-like morphology; Region B exhibits oblique bamboo-raft-like morphology; positive cubic morphology and smaller bamboo-raft-like strengthening phases, similar in morphology, are clustered in Region C; and positive bamboo-raft-like morphologies of varying sizes also cluster in the lower left portion of the UMAP space.

[0035] In terms of morphological evolution, Figure 7As shown, the positive cubic morphology is distributed in the upper right area of ​​region C. As the UMAP-1 and UMAP-2 components decrease, the vertical matrix phase gradually disappears and a slight bamboo raft-like morphology appears. When the UMAP-1 component decreases to around -7, as the size of the bamboo raft-like morphology further increases, the UMAP-1 component begins to increase, and this process is also accompanied by a slight increase in the UMAP-2 component. In addition, whether it is a positive bamboo raft-like morphology, a positive needle-like morphology, or an oblique bamboo raft-like morphology, as the UMAP-2 component decreases, the tissue size shows an increasing trend. In the microstructure state studied in the embodiment, the UMAP-1 component and the UMAP-2 component jointly control the evolution of the microstructure morphology and size, and the UMAP-2 component is closely related to the size evolution.

[0036] Further, all microscopic images of the alloy in a specific positive bamboo raft morphology were characterized by local dimensionality reduction. Figure 8 The dimensionality reduction results of the specific microstructural morphology of a nickel-based single crystal superalloy in the local UMAP topological space are presented, where the horizontal and vertical coordinates of the schematic diagram represent the UMAP-1 and UMAP-2 components, respectively. Compared with the global UMAP space, the morphology distribution in the local space is more uniform. The size of the positive bamboo-raft morphology is related to both components of the local UMAP space, and is more closely related to the UMAP-1 component. As the UMAP-1 component increases, the slightly bamboo-raft-like vertical matrix phase completely disappears, forming a completely horizontally connected bamboo-raft morphology. As the UMAP-2 component decreases, the width of the matrix phase increases significantly, and the severely bamboo-raft-like morphology forms a cluster distributed in the lower right corner of the local space.

[0037] Using the above process, professionals in this field can determine the microstructural morphology, size, and evolution of nickel-based single-crystal superalloys based on the values ​​of the UMAP-1 and UMAP-2 components. In summary, the coupled image enhancement and machine learning techniques successfully achieve comprehensive quantitative characterization of the material's complex microstructural features.

[0038] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Various modifications are possible. In other words, any simple, equivalent changes and modifications made in accordance with the claims and description of the present application are within the scope of protection of the patent claims.

Claims

1. A quantitative characterization method for material microstructure by coupling image enhancement and machine learning, characterized in that: The following steps are involved: S1: preparing an observation sample of the target alloy material, and obtaining a microstructure image set of the observation sample by a microscopic observation method; S2: Image enhancement, multiple noise reduction and binarization processing of microscopic tissue image set based on Gaussian distribution adaptive threshold segmentation, mask processing and adaptive histogram equalization; S3: using a deep convolutional neural network algorithm to extract the features of the microstructure image set processed in step S2 to a high-dimensional latent space, obtaining high-dimensional latent space features, and then using a consistent manifold dimensionality reduction algorithm to reduce the dimension and quantitatively characterize the high-dimensional latent space features, that is, to obtain a reduced-dimensional quantitative characterization of the microstructure of the target alloy material; S4: Conduct physical interpretability analysis of dimensionality reduction quantitative characterization of the target alloy material microstructure, and evaluate the effect of dimensionality reduction quantitative indicators on the quantitative characterization of the target alloy material microstructure.

2. The material microstructure quantitative characterization method coupled with image enhancement and machine learning according to claim 1, characterized in that: In the step S1, sample blocks are cut from target alloy materials with different service histories or pre-treated, and are prepared into microscopic observation samples after mechanical sample preparation or fine slicing; images of different microstructural states of the observation samples are captured using a microscopic observation method to form a microstructural image set including various microstructural features of the target alloy materials.

3. The material microstructure quantitative characterization method coupled with image enhancement and machine learning according to claim 1, characterized in that: In the step S1, the microstructure image set includes microstructure images of different microstructure morphologies and sizes of the target alloy material.

4. The material microstructure quantitative characterization method coupled with image enhancement and machine learning according to claim 1, characterized in that: The target alloy material is a two-phase metal alloy material, and the step S2 specifically comprises the following steps: S2.1: A Gaussian filter is used to process the noise in the microstructure image set, and the Otsu method is used to perform preliminary binarization processing on the denoised microstructure image set. The black pixel value of 1 in the black-and-white binary image represents the I phase of the target alloy material, and the white pixel value of 0 represents the II phase of the target alloy material; S2.2: The microstructure image set binarized by the Otsu method is used as a mask for mask processing, and is applied to the microstructure image set after Gaussian noise reduction, so that the phase boundary area in the material microstructure is more prominent, and a mask image set is obtained; S2.3: using an adaptive histogram equalization method to process the mask image set, so that the grayscale histogram is nonlinearly mapped from the foreground interval to the entire grayscale range, thereby enhancing the contrast between different phases in the target alloy material, and obtaining a balanced mask image set; S2.4: The grayscale histogram of the mask image set after equalization is a bimodal distribution. The second-order Gaussian distribution is used to fit the grayscale histogram, and the grayscale corresponding to the left peak value is mainly corresponding to phase II. S2.5: The grayscale corresponding to the left peak of the second-order Gaussian distribution in step S2.4 is used as the threshold, and the pixels in the image that are less than the threshold are set to 0 and the pixels that are greater than the threshold are set to 1 using a fixed threshold function, thereby achieving the final binarization processing of the equalized mask image set; S2.6: Based on the opening operation denoising algorithm, the discrete points near phase I are expanded and eroded by pixels, and the boundary of phase I after binarization is smoothed. The connected domain denoising method is used to eliminate the discrete foreground points scattered in phase II and the discrete pixel blocks that affect the image quality. Finally, a set of microstructure images is obtained after image enhancement, multiple denoising and binarization processing.

5. The material microstructure quantitative characterization method coupled with image enhancement and machine learning according to claim 1, characterized in that: The step S3 specifically comprises the following steps: S3.1: Select a deep convolutional neural network algorithm, remove the classification layer, retain the convolution layer and pooling layer, and add a linear fully connected layer to average pool and flatten the output of the deep convolutional neural network algorithm; S3.2: Based on the microscopic tissue image set obtained in step S2 after image enhancement, multiple noise reduction and binarization processing, select an image cropping algorithm, crop the microscopic tissue image set to the input dimension of the deep convolutional neural network algorithm selected in step S3.1, and expand the cropped microscopic tissue image set to 3 channels to form the deep convolutional neural network algorithm input image set; S3.3: Input the deep convolutional neural network algorithm input image set in step S3.2 into the deep convolutional neural network algorithm one by one, extract the microscopic tissue image features and map them to the high-dimensional latent space, and obtain the high-dimensional latent space features of the microscopic tissue image set; S3.4: Setting the upper limit and selected lower limit parameters for the number of local connections of data points in the consistent manifold dimensionality reduction algorithm; S3.5: inputting the high-dimensional latent space features of the microscopic tissue image set obtained in step S3.3 into the consistent manifold dimensionality reduction algorithm model, constructing a weighted neighborhood graph and a connected neighborhood graph, thereby learning the manifold structure of the microscopic tissue image set in the high-dimensional latent space; S3.6: Project the approximate manifold structure of the high-dimensional latent space to the low-dimensional space, convert the variable distance of the approximate manifold structure into the standard Euclidean distance of the low-dimensional global coordinate system, take distance invariance as the optimization goal, and use cross entropy as the loss function. By adjusting the upper limit and lower limit of the number of local connections of data points in the consistent manifold dimensionality reduction algorithm, find the optimal low-dimensional popular representation, and finally obtain the dimensionality reduction quantitative representation of the microscopic tissue image set.

6. The material microstructure quantitative characterization method coupled with image enhancement and machine learning according to claim 1, characterized in that: In the step S4, the consistent manifold component after dimensionality reduction by the consistent manifold dimensionality reduction algorithm in step S3 is used as a quantitative characterization parameter of the microstructure characteristics of the target alloy material, and a physical interpretability analysis of the dimensionality reduction results is carried out; global consistent manifold topological space mapping and local consistent manifold topological space mapping are performed on the microstructure image set to evaluate the characterization effect of the dimensionality reduction quantitative index, thereby achieving a comprehensive quantitative characterization of the microstructure of the target alloy material.

7. The material microstructure quantitative characterization method coupled with image enhancement and machine learning according to claim 1, characterized in that: The step S1 uses an optical microscope, a scanning electron microscope or a transmission electron microscope to perform microscopic observation of the sample.

Images