Large model-fractal combined auxiliary mineral image intelligent segmentation and identification method

Abstract

The invention relates to the technical field of computer vision and image processing. The invention provides a large model-fractal combined assisted mineral image intelligent segmentation and identification method, which comprises the following steps: acquiring mineral image data, and preprocessing to obtain a mineral image data set; constructing an initial joint model based on the SAM and the fractal model, and training the initial joint model to obtain a joint model; performing optimization processing on the joint model through the mineral image data set to obtain an optimized joint model; segmenting the to-be-processed mineral image, and marking boundaries of mineral particles in the to-be-processed mineral image; and fractal analysis is conducted on the labeled mineral particles through the optimized joint model, and segmentation masks and fractal quantitative information of the mineral particles are output. The problems that a traditional mineral image analysis method depends on manual observation and manual labeling, time and labor are consumed, and analysis results are inconsistent and uncertain due to the fact that the traditional mineral image analysis method is prone to being affected by subjective factors are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of computer vision and image processing, and in particular to a large model-fractal combined assisted mineral image intelligent segmentation and recognition method. Background Art

[0002] In geology, mineralogy and related engineering fields, the segmentation and recognition of mineral images is a crucial task. It not only provides basic data support for the exploration, mining and utilization of mineral resources, but also plays a key role in the research of multiple disciplines such as petrology and geochemistry.

[0003] Traditional mineral image analysis methods often rely on manual observation and annotation, which is not only time-consuming and labor-intensive, but also easily affected by subjective factors, resulting in inconsistency and uncertainty in the analysis results. Summary of the invention

[0004] The purpose of the present invention is to provide a large model-fractal joint assisted mineral image intelligent segmentation and recognition method, aiming to solve the problem that traditional mineral image analysis methods rely on manual observation and manual annotation, which is not only time-consuming and labor-intensive, but also easily affected by subjective factors, resulting in inconsistency and uncertainty in analysis results.

[0005] The present invention is achieved through the following technical solutions:

[0006] A large model-fractal joint assisted mineral image intelligent segmentation and recognition method, comprising the steps of:

[0007] Acquire mineral image data, and preprocess the mineral image data to obtain a mineral image data set;

[0008] An initial joint model is constructed based on the SAM and the fractal model, and the initial joint model is trained to obtain a joint model;

[0009] The joint model is optimized through the mineral image data set to obtain an optimized joint model;

[0010] The mineral image to be processed is segmented by optimizing the joint model, and the boundaries of the mineral particles in the mineral image to be processed are marked;

[0011] The fractal analysis of the annotated mineral particles is performed by optimizing the joint model, and the segmentation mask and fractal quantitative information of the mineral particles are output.

[0012] Optionally, the specific process of acquiring the mineral image data and preprocessing the mineral image data to obtain the image data set is:

[0013] The mineral image data including several mineral samples are obtained; the obtained mineral image data are denoised; the denoised mineral image data are enhanced to obtain a mineral image data set.

[0014] Optionally, the mineral image data is obtained by performing microscopic imaging of the mineral sample using a high-resolution microscope or a microscopic imaging system under an orthogonal polarization system and a single polarization system.

[0015] Optionally, the specific process of constructing the initial joint model based on SAM and fractal model is:

[0016] The SAM model is used as the basic segmentation model and combined with the analysis model to construct an initial joint model. The initial joint model includes an image encoding module, a hint encoding module, a mask decoding module and a fractal analysis module. The outputs of the image encoding module and the hint encoding module are connected to the input of the mask decoding module, and the output of the mask decoding module is connected to the input of the fractal analysis module. Among them, the image encoding module is used to extract the characteristic information of the mineral image; the hint encoding module is used to extract the hint information of the mineral image; the mask decoding module is used to decode the characteristic information and hint information of the mineral image into a segmentation mask image; and the fractal analysis module is used to perform fractal analysis on the segmentation mask image.

[0017] Optionally, the specific process of training the initial joint model to obtain the joint model is:

[0018] Divide the mineral image dataset into training set, validation set and test set;

[0019] The boundaries of the mineral particles in the training set are marked, and the point prompt information and bounding box prompt information of the mineral image are extracted through the prompt encoding module, and the marked mineral particles are segmented;

[0020] The fractal analysis module is used to calculate the fractal dimension of each segmented mineral particle and quantify the morphological complexity and spatial distribution characteristics of the mineral particles.

[0021] Optionally, the specific process of calculating the fractal dimension of each segmented mineral particle through the fractal analysis module and quantifying the morphological complexity and spatial distribution characteristics of the mineral particles is:

[0022] The box dimension method is used to calculate the fractal dimension. The boundary area of ​​the segmented mineral particles is divided into grids of different step sizes, and the step size of the grid is defined as ∈. For each step size ∈, the minimum number of boxes N(∈) required to cover the mineral particles is calculated on the boundary of the mineral particles. The calculation expression of the fractal dimension is as follows (1):

[0023]

[0024] Where D represents the fractal dimension; the fractal dimension D is obtained by fitting with different step sizes ∈ and corresponding N(∈) values;

[0025] By calculating the fractal dimension distribution of all mineral particles, the average fractal dimension and standard deviation of different types of mineral particles are counted, and the distribution characteristics of the fractal dimension of mineral particles are extracted;

[0026] The threshold range is set according to the distribution characteristics of the fractal dimension of mineral particles, and mineral particles with matching morphological complexity are classified into the same category;

[0027] Different types of mineral particles are classified and processed by setting the threshold range.

[0028] Optionally, the specific process of optimizing the joint model through the image data set to obtain the optimized joint model is:

[0029] When fine-tuning the joint model, the parameters of the mask decoding module are set to fixed values; the parameters of the image encoding module and the prompt encoding module are updated through the mask image of the real mineral particles and the corresponding center point prompt information, and the optimized joint model is obtained.

[0030] Optionally, the optimization process also includes iterative optimization; acquiring the embedding vector of the mineral image data set through the image encoding module; generating prompt information of the mineral image data set through the prompt encoding module; inputting the embedding vector and the prompt information into the mask decoding module to generate a predicted mask image; setting the total loss function, and updating the parameters of the joint model through the gradient descent method until the loss function value is lower than the preset parameter threshold, and stopping the iteration.

[0031] Optionally, the specific process of segmenting the mineral image to be processed by optimizing the joint model and marking the boundaries of mineral particles in the mineral image to be processed is:

[0032] The mineral image to be processed is input into the image encoding module of the optimized joint model to extract the characteristic information of the mineral image to be processed; the prompt information of the mineral image to be processed is extracted through the prompt encoding module; the extracted characteristic information and prompt information are input into the mask decoding module to generate a segmentation mask image of the mineral image to be processed; according to the segmentation mask image, the boundary of each mineral particle in the mineral image to be processed is marked.

[0033] Optionally, the specific process of performing fractal analysis on the labeled mineral particles by optimizing the joint model and outputting the segmentation mask and fractal quantitative information of the mineral particles is:

[0034] The segmentation mask image of the labeled mineral particles is input into the fractal analysis module of the optimized joint model; the fractal analysis module calculates the fractal dimension of each mineral particle; based on the calculated fractal dimension, the morphological complexity and spatial distribution characteristics of each mineral particle are quantified; and quantitative information including the segmentation mask and fractal dimension of the mineral particles is output.

[0035] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0036] Improve the accuracy of mineral image segmentation: By combining the powerful feature extraction capability of the large model and the fine structure analysis capability of the fractal model, the mineral particles in the mineral image can be segmented more accurately, especially for those mineral particles with complex textures and boundaries, achieving a significant improvement in segmentation accuracy.

[0037] Enhanced mineral identification capabilities: Fractal analysis not only provides the morphological characteristics of mineral particles, but also reveals the complexity of their internal structure, which is crucial for the accurate identification of minerals. Combined with the classification capabilities of large models, it can more reliably identify different types of minerals, providing strong support for subsequent mineralogical research.

[0038] Optimize processing flow and improve efficiency: By building and optimizing the joint model, rapid processing of mineral image data is achieved, optimization reduces manual intervention, improves processing efficiency, and makes analysis of large-scale mineral image data possible.

[0039] Provide rich fractal quantitative information: It can output fractal quantitative information of mineral particles, such as fractal dimension, which is of great significance for understanding the formation process, physical properties and chemical composition of minerals, and provides a scientific basis for the development and utilization of mineral resources.

[0040] Strong adaptability and wide application: Through the combination of large models and fractal models, it has strong adaptability to different types of mineral images and can be widely used in geological exploration, mineral resource development, material science research and other fields, promoting the development of related disciplines. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The figure is a flow chart of the method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0043] Example 1

[0044] Reference Figure 1 , a large model-fractal joint assisted mineral image intelligent segmentation and recognition method, comprising the steps of:

[0045] Step 1: Acquire mineral image data and preprocess the mineral image data to obtain a mineral image dataset.

[0046] In this embodiment, mineral image data including several mineral samples are obtained; denoising is performed on the obtained mineral image data; image enhancement is performed on the denoising mineral image data to obtain a mineral image data set.

[0047] In this embodiment, a variety of mineral samples, such as quartz, mica, albite, etc., are selected from geological laboratories or mineral specimen libraries to ensure that the samples are representative and cover different mineral forms and structures. Use a high-resolution microscope or microscopic imaging system to perform microscopic imaging of mineral samples under an orthogonal polarization system and a single polarization system. The collected image should clearly show the boundaries, textures, and internal structures of the mineral particles. De-noising can use image denoising methods such as Gaussian filtering or median filtering. The collected mineral image data is input into image processing software (such as MATLAB, Python's OpenCV library, etc.), and the selected denoising algorithm is applied to eliminate noise interference in the image, such as random noise, spots, etc., and the clarity of the mineral boundary is improved by comparing the images before and after denoising. Image enhancement can adjust the brightness, contrast, saturation and other parameters of the image according to the characteristics of the mineral image; use the image enhancement function in the image processing software to enhance the denoised image to make the boundaries of the mineral particles more significant and the texture clearer; visually inspect or calculate image quality indicators (such as signal-to-noise ratio, contrast, etc.) to ensure that the image enhancement effect meets the requirements of subsequent segmentation and recognition. The preprocessed mineral image data are classified and sorted according to mineral type, optical system (orthogonal polarization, single polarization), etc. to form a structured data set.

[0048] Step 2: construct an initial joint model based on SAM and the fractal model, train the initial joint model, and obtain a joint model. SAM stands for Segment Anything Model, which is an image segmentation model based on deep learning.

[0049] In this embodiment, the specific process of constructing the initial joint model based on SAM and fractal model is as follows:

[0050] The SAM model is used as the basic segmentation model and combined with the analysis model to construct an initial joint model. The initial joint model includes an image encoding module, a hint encoding module, a mask decoding module and a fractal analysis module. The outputs of the image encoding module and the hint encoding module are connected to the input of the mask decoding module, and the output of the mask decoding module is connected to the input of the fractal analysis module. Among them, the image encoding module is used to extract the characteristic information of the mineral image; the hint encoding module is used to extract the hint information of the mineral image; the mask decoding module is used to decode the characteristic information and hint information of the mineral image into a segmentation mask image; and the fractal analysis module is used to perform fractal analysis on the segmentation mask image.

[0051] In this embodiment, the specific process of training the initial joint model to obtain the joint model is as follows:

[0052] The mineral image dataset is divided into training set, validation set and test set; the training set is used for model training, the validation set is used for model tuning, and the test set is used to evaluate model performance.

[0053] The boundaries of the mineral particles in the training set are annotated, and these annotations will be used as the real labels for model training; the point hint information and bounding box hint information of the mineral image are extracted through the hint encoding module, and the annotated mineral particles are segmented. The point hint information and bounding box hint information will help the model to segment the mineral particles more accurately; the training set is input into the joint model, the loss is calculated through forward propagation, and the model parameters are updated through back propagation. During the training process, the validation set is used to monitor the performance of the model and adjust the hyperparameters in time. Common segmentation loss functions such as cross entropy loss or Dice loss can be used to measure the difference between the segmentation mask image and the real label.

[0054] In each iteration, the segmentation mask image output by the model is subjected to fractal analysis. The fractal dimension of each segmented mineral particle is calculated by the fractal analysis module to quantify the morphological complexity and spatial distribution characteristics of the mineral particles. In order to make the model generate segmentation results with reasonable fractal dimensions, additional fractal loss functions can be designed. For example, the mean square error between the predicted fractal dimension and the true fractal dimension can be calculated as the fractal loss.

[0055] Step 3: Optimize the joint model through the mineral image data set to obtain the optimized joint model.

[0056] In this embodiment, the specific process of optimizing the joint model by using the image data set to obtain the optimized joint model is as follows:

[0057] When fine-tuning the joint model, the parameters of the mask decoding module are set to fixed values ​​to reduce computing resource consumption and accelerate the training process; the parameters of the image encoding module and the prompt encoding module are updated through the back propagation algorithm through the mask image of the real mineral particles and the corresponding center point prompt information, and the optimized joint model is obtained. The real mask image is used as supervision information, and the center point prompt information is used as additional guidance information to help the model segment mineral particles more accurately. In order to prevent the model from overfitting, data enhancement techniques (such as random rotation, flipping, scaling, etc.) are used to enhance the mineral images in the training set. This helps to generate diverse training samples and improve the generalization ability of the model.

[0058] In this embodiment, the optimization process also includes iterative optimization; the embedding vector of the mineral image data set is obtained through the image encoding module; the prompt information of the mineral image data set is generated through the prompt encoding module; the embedding vector and the prompt information are input into the mask decoding module to generate a predicted mask image; the total loss function is set, and a total loss function including segmentation loss and fractal analysis loss is defined. The segmentation loss is used to evaluate the difference between the segmentation mask generated by the model and the true mask, and the fractal analysis loss is used to evaluate the difference between the fractal dimension calculated by the model and the true fractal dimension; the parameters of the joint model are updated by the gradient descent method, and the parameters of the joint model are updated using the gradient descent method (or its variants, such as the Adam optimizer). In each iteration, the gradient of the total loss function is calculated, and the model parameters are adjusted according to the gradient to minimize the total loss; the number of iterations or the loss function threshold is set as the stopping condition until the loss function value is lower than the preset parameter threshold, and the preset parameter threshold is the preset number of iterations or the loss function threshold, and the iteration is stopped. The performance of the optimized joint model is evaluated on the validation set. The performance of the model is evaluated by calculating indicators such as segmentation accuracy, recall rate, F1 score, and the calculation accuracy and stability of the fractal dimension; based on the evaluation results on the validation set, the model parameters, data enhancement strategies, etc. are adjusted and optimized to improve the model performance.

[0059] Step 4: Segment the mineral image to be processed by optimizing the joint model and mark the boundaries of mineral particles in the mineral image to be processed.

[0060] In this embodiment, the specific process of segmenting the mineral image to be processed by optimizing the joint model and marking the boundaries of the mineral particles in the mineral image to be processed is as follows:

[0061] The mineral image to be processed is input into the image encoding module of the optimized joint model to extract the feature information of the mineral image to be processed. The mineral image to be processed can be a mineral sample image from a microscope, covering various mineral forms and structures; the image encoding module receives the input mineral image and starts to extract the feature information in the image, including color, texture, shape and other feature information; the prompt information of the mineral image to be processed is extracted through the prompt encoding module, and the prompt information is used to guide the subsequent segmentation process. The point prompt information is used to determine the approximate position of the mineral particles, and the boundary box prompt information is used to further limit the segmentation area; the extracted feature information and prompt information are input into the mask decoding module to generate the segmentation mask image of the mineral image to be processed. The mask image is a binary image, in which the white area represents the mineral particles and the black area represents the background; according to the segmentation mask image, the boundary of each mineral particle in the mineral image to be processed is marked, and the boundary of the mineral particle can be obtained by calculating the outline of the white area in the mask image. The marked mineral particle boundary is output as a visualization result, which can be an original mineral image superimposed with the segmentation boundary, or an image containing only the segmentation boundary.

[0062] Step 5: Perform fractal analysis on the labeled mineral particles by optimizing the joint model, and output the segmentation mask and fractal quantitative information of the mineral particles.

[0063] In this embodiment, the specific process of performing fractal analysis on the labeled mineral particles by optimizing the joint model and outputting the segmentation mask and fractal quantitative information of the mineral particles is as follows:

[0064] The segmentation mask image of the annotated mineral particles is input into the fractal analysis module of the optimized joint model; the fractal analysis module calculates the fractal dimension of each mineral particle, and the box counting method, radius dimension method and other calculation methods can be used to calculate the fractal dimension; according to the calculated fractal dimension, the morphological complexity and spatial distribution characteristics of each mineral particle are quantified. The higher the fractal dimension, the more complex the morphology of the mineral particles and the more irregular the edges. Conversely, the lower the fractal dimension, the simpler the morphology of the mineral particles and the smoother the edges; the spatial distribution characteristics between different mineral particles, such as the degree of aggregation and dispersion, can also be inferred by analyzing the differences in fractal dimensions between different mineral particles; the quantitative information containing the segmentation mask and fractal dimension of the mineral particles is output, and this information can be presented in the form of images, tables or texts for subsequent analysis and processing. For example, an image containing the segmentation mask and fractal dimension of each mineral particle can be generated, where different colors or grayscales represent different fractal dimension values; or a table can be generated to list the segmentation mask coordinates and fractal dimension values ​​of each mineral particle.

[0065] Example 2

[0066] Based on Example 1, in this embodiment, the fractal dimension of each segmented mineral particle is calculated by the fractal analysis module, and the specific process of quantifying the morphological complexity and spatial distribution characteristics of the mineral particles is as follows:

[0067] The box dimension method is used to calculate the fractal dimension. The boundary area of ​​the segmented mineral particles is divided into grids of different step sizes, such as 1 pixel, 2 pixels, 4 pixels, etc. The step size of the grid is defined as ∈. For each step size ∈, the minimum number of boxes N(∈) required to cover the mineral particles is calculated on the boundary of the mineral particles. The calculation expression of the fractal dimension is as follows (1):

[0068]

[0069] Where D represents the fractal dimension; the fractal dimension D is obtained by fitting with different step sizes ∈ and corresponding N(∈) values;

[0070] By calculating the fractal dimension distribution of all mineral particles, the average fractal dimension and standard deviation of different types of mineral particles are counted, and the distribution characteristics of the fractal dimension of mineral particles are extracted. The fractal dimension distribution histogram or kernel density estimation diagram of each type of mineral particles can be drawn to intuitively display its distribution characteristics, analyze the degree of overlap of the fractal dimensions of different types of mineral particles, and whether there are obvious distribution differences.

[0071] The threshold range is set according to the distribution characteristics of the fractal dimension of mineral particles. The threshold range can be used to distinguish mineral particles with different morphological complexities; mineral particles with matching morphological complexity are classified into the same category, which helps to further understand the morphological complexity and spatial distribution characteristics of mineral particles and provide an important basis for subsequent mineral identification and analysis.

[0072] Different types of mineral particles are classified and processed by setting the threshold range.

[0073] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A large model-fractal joint assisted mineral image intelligent segmentation and recognition method, characterized in that: Includes steps: Acquire mineral image data, and preprocess the mineral image data to obtain a mineral image data set; An initial joint model is constructed based on the SAM and the fractal model, and the initial joint model is trained to obtain a joint model; The joint model is optimized through the mineral image data set to obtain an optimized joint model; The mineral image to be processed is segmented by optimizing the joint model, and the boundaries of the mineral particles in the mineral image to be processed are marked; The fractal analysis of the annotated mineral particles is performed by optimizing the joint model, and the segmentation mask and fractal quantitative information of the mineral particles are output.

2. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination as claimed in claim 1, characterized in that: The specific process of obtaining mineral image data and preprocessing the mineral image data to obtain an image data set is as follows: The mineral image data including several mineral samples are obtained; the obtained mineral image data are denoised; the denoised mineral image data are enhanced to obtain a mineral image data set.

3. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination as claimed in claim 2, characterized in that: The mineral image data is obtained by performing microscopic imaging of the mineral sample under an orthogonal polarization system or a single polarization system through a high-resolution microscope or a microscopic imaging system.

4. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination according to claim 1, characterized in that: The specific process of constructing the initial joint model based on SAM and fractal model is as follows: The SAM model is used as the basic segmentation model and combined with the analysis model to construct an initial joint model. The initial joint model includes an image encoding module, a hint encoding module, a mask decoding module and a fractal analysis module. The outputs of the image encoding module and the hint encoding module are connected to the input of the mask decoding module, and the output of the mask decoding module is connected to the input of the fractal analysis module. Among them, the image encoding module is used to extract the characteristic information of the mineral image; the hint encoding module is used to extract the hint information of the mineral image; the mask decoding module is used to decode the characteristic information and hint information of the mineral image into a segmentation mask image; and the fractal analysis module is used to perform fractal analysis on the segmentation mask image.

5. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination as claimed in claim 4 is characterized in that: The specific process of training the initial joint model to obtain the joint model is as follows: Divide the mineral image dataset into training set, validation set and test set; The boundaries of the mineral particles in the training set are marked, and the point prompt information and bounding box prompt information of the mineral image are extracted through the prompt encoding module, and the marked mineral particles are segmented; The fractal analysis module is used to calculate the fractal dimension of each segmented mineral particle and quantify the morphological complexity and spatial distribution characteristics of the mineral particles.

6. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination as claimed in claim 5, characterized in that: The specific process of calculating the fractal dimension of each segmented mineral particle through the fractal analysis module and quantifying the morphological complexity and spatial distribution characteristics of the mineral particles is as follows: The box dimension method is used to calculate the fractal dimension. The boundary area of ​​the segmented mineral particles is divided into grids of different step sizes, and the step size of the grid is defined as ∈. For each step size ∈, the minimum number of boxes N(∈) required to cover the mineral particles is calculated on the boundary of the mineral particles. The calculation expression of the fractal dimension is as follows (1): Where D represents the fractal dimension; the fractal dimension D is obtained by fitting with different step sizes ∈ and corresponding N(∈) values; By calculating the fractal dimension distribution of all mineral particles, the average fractal dimension and standard deviation of different types of mineral particles are counted, and the distribution characteristics of the fractal dimension of mineral particles are extracted; The threshold range is set according to the distribution characteristics of the fractal dimension of mineral particles, and mineral particles with matching morphological complexity are classified into the same category; Different types of mineral particles are classified and processed by setting the threshold range.

7. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination as claimed in claim 4, characterized in that: The specific process of optimizing the joint model through the image data set to obtain the optimized joint model is: When fine-tuning the joint model, the parameters of the mask decoding module are set to fixed values; the parameters of the image encoding module and the prompt encoding module are updated through the mask image of the real mineral particles and the corresponding center point prompt information, and the optimized joint model is obtained.

8. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination as claimed in claim 7, characterized in that: The optimization process also includes iterative optimization; acquiring the embedding vector of the mineral image data set through the image encoding module; generating prompt information of the mineral image data set through the prompt encoding module; inputting the embedding vector and the prompt information into the mask decoding module to generate a predicted mask image; setting a total loss function, updating the parameters of the joint model through the gradient descent method, until the loss function value is lower than the preset parameter threshold, and stopping the iteration.

9. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination according to claim 5, characterized in that: The specific process of segmenting the mineral image to be processed by optimizing the joint model and marking the boundaries of mineral particles in the mineral image to be processed is: The mineral image to be processed is input into the image encoding module of the optimized joint model to extract the characteristic information of the mineral image to be processed; the prompt information of the mineral image to be processed is extracted through the prompt encoding module; the extracted characteristic information and prompt information are input into the mask decoding module to generate a segmentation mask image of the mineral image to be processed; according to the segmentation mask image, the boundary of each mineral particle in the mineral image to be processed is marked.

10. The method for intelligent segmentation and recognition of mineral images assisted by large model-fractal combination according to claim 9, characterized in that: The specific process of performing fractal analysis on the labeled mineral particles by optimizing the joint model and outputting the segmentation mask and fractal quantitative information of the mineral particles is as follows: The segmentation mask image of the annotated mineral particles is input into the fractal analysis module of the optimized joint model; the fractal analysis module calculates the fractal dimension of each mineral particle; and according to the calculated fractal dimension, the morphological complexity and spatial distribution characteristics of each mineral particle are quantified; The output contains mineral grain segmentation masks and quantitative information of fractal dimensions.

Images