Deep-sea mining ecological restoration capability and potential evaluation system based on biological substratum

Abstract

The invention relates to the technical field of deep-sea mining ecological restoration, in particular to a biological substratum-based deep-sea mining ecological restoration capability and potential evaluation system, which comprises a data acquisition module, a biological substratum identification module and a deep-sea ecological dynamic monitoring module, the data acquisition module acquires water quality parameters of a deep-sea mining area; the biological substratum identification module identifies and analyzes population distribution, density and form of the biological substratum and evaluates growth and health conditions of the biological substratum; the deep sea ecology dynamic monitoring module analyzes the microclimate and sediment dynamic change of a mining area, and evaluates the influence of environment change on the restoration capability of the biological substratum. According to the method, the progress and potential of ecological restoration can be accurately predicted, and powerful data support is provided for scientific decision and environmental protection measures of deep sea ecological restoration.

Description

Technical Field

[0001] The invention relates to the technical field of deep-sea mining ecological restoration, and in particular to a deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment bases. Background Art

[0002] With the increasing demand for deep-sea mineral resource development, the scale and frequency of deep-sea mining activities have gradually increased. However, the environmental disturbance and pollution problems caused by deep-sea mining have had a significant impact on the health and stability of marine ecosystems, especially biological adhesion substrates. As an important component of marine ecosystems, biological adhesion substrates not only provide habitats for a variety of marine organisms, but also play a key role in material circulation and biodiversity maintenance. Therefore, conducting effective ecological restoration capacity assessments in deep-sea mining activities to ensure the sustainable recovery of biological adhesion substrates is an important research direction to ensure the healthy development of deep-sea ecosystems.

[0003] At present, deep-sea ecological restoration assessment systems mostly use single-parameter monitoring methods, lacking dynamic collection and analysis of multi-dimensional environmental data, making it difficult to fully reflect ecological changes in deep-sea mining areas. At the same time, traditional monitoring systems generally have problems with data lag and insufficient accuracy, and are unable to reflect changes in key environmental factors such as water quality and sediments in real time, resulting in deviations between assessment results and actual conditions. In addition, there are limited analytical methods for the health and recovery status of biological attachment bases, and most assessment systems fail to comprehensively consider the growth rate, distribution density and environmental changes of biological attachment bases, resulting in difficulty in accurately predicting ecological restoration potential, affecting the ecological management and decision-making of deep-sea mining activities.

[0004] The purpose of the present invention is to provide a deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base, which not only improves the timeliness and accuracy of monitoring, but also provides scientific and objective decision-making support for the ecological protection and sustainable development of deep-sea mining activities. Summary of the invention

[0005] The invention provides a deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base.

[0006] The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base includes a data acquisition module, a biological attachment base identification module and a deep-sea ecological dynamic monitoring module, among which; The data acquisition module collects water quality parameters of the deep-sea mining area through multiple sensors, including the temperature, depth and salinity of the ocean water body; The biological attachment base identification module identifies and analyzes the population distribution, density, and morphology of the biological attachment base, and evaluates the growth and health status of the biological attachment base in combination with the collected water quality parameters; The deep-sea ecological dynamic monitoring module uses sensors to monitor the environmental data of the deep-sea mining area in real time, including microclimate (dissolved oxygen, pH) and sediment (particle size, heavy metal ions). By analyzing the dynamic changes of the microclimate and sediment in the mining area, the module evaluates the impact of environmental changes on the recovery capacity of biological attachment bases, including: Environmental data collection: Use deep-sea sensors to collect various environmental data of deep-sea mining areas in real time, including dissolved oxygen, pH, particle size, and heavy metal ions; Data preprocessing: preprocess the collected environmental data, including noise filtering and standardization; Analysis and assessment of environmental changes: Analyze the pre-processed environmental data to evaluate the impact of dynamic changes in microclimate and sediments in deep-sea mining areas on the recovery capacity of biofouling substrates.

[0007] Optionally, the data acquisition module includes: Temperature data collection: Use temperature sensors to measure the temperature of water in deep-sea mining areas in real time and record changes in water temperature at each designated location and depth; Depth data collection: Use depth sensors to monitor the water depth of deep-sea mining areas in real time and capture depth changes in deep-sea waters; Salinity data collection: Salinity sensors are used to measure the salinity of water in deep-sea mining areas.

[0008] Optionally, the biological attachment base identification module includes: Biological adhesion base image acquisition and preliminary processing: The image data of biological adhesion base in the deep-sea mining area is collected in real time through underwater image acquisition equipment, and the collected image data is preliminarily processed, including denoising, image enhancement and segmentation; Identification of biological adhesion bases: The preliminarily processed image data is identified through the convolutional neural network (CNN) model to analyze the population distribution, density, and morphology of biological adhesion bases to determine the distribution of biological adhesion bases in different areas; Analysis of the growth and health status of biological attachments: Based on the results of biological attachment identification, the growth rate and health status of biological attachments under different water quality conditions are evaluated by analyzing the correlation between water quality parameters and the growth of biological attachments.

[0009] Optionally, the biological attachment base image acquisition and preliminary processing include: Underwater image acquisition: Real-time image data of biological attachment bases in deep-sea mining areas are collected through underwater cameras; Image denoising: Gaussian filtering algorithm is used to suppress noise in the collected image data; Image enhancement: Use histogram equalization to improve the contrast and details of image data; Image segmentation: The Otsu threshold method was used to separate the biofouling area from the background in the image data.

[0010] Optionally, the convolutional neural network (CNN) model includes: Multi-scale feature extraction: through convolution kernels of different scales ( , , ) to extract the multi-scale features of the image, and concatenate the convolution results of all scales to obtain a multi-scale feature map ; Fully connected layer: multi-scale feature maps Flatten it into a one-dimensional feature vector F, and input the flattened feature vector F into the fully connected layer to generate a feature representation ; Pixel classification: for each pixel in the comprehensive feature map , use Softmax classifier for To classify; Segmentation result: The segmentation result is obtained by taking the maximum classification probability of each pixel ; Distribution and density calculation: The distribution and density information of biological attachment bases are calculated based on the segmentation results. The category information of each pixel is directly used to determine the spatial distribution of biological adhesion bases, and the density D of biological adhesion bases is calculated by counting the number of pixels belonging to the biological adhesion base category.

[0011] Optionally, the growth and health status analysis of the biological attachment base includes: Growth rate calculation: By analyzing the changes in the density of biological attachment bases at different time points, the growth rate G is calculated; Water quality correlation analysis: Pearson correlation coefficient r was used to analyze the correlation between water quality parameters and biofouling density; Health status assessment: Combine growth rate and water quality correlation results to determine the health threshold of growth rate. and unhealthy threshold and the correlation threshold Compare and comprehensively evaluate the health status of the biological attachment base. and , indicating health, if or , indicating unhealthy.

[0012] Optionally, the environmental data collection includes: Dissolved oxygen data collection: Real-time collection of dissolved oxygen concentration in water through dissolved oxygen sensors; pH data collection: Use pH sensors to monitor the pH of water in real time; Particle size data collection: The particle size distribution of suspended particles in water is measured by particle size analysis sensors to provide water turbidity and particle information; Heavy metal ion data collection: Use heavy metal ion sensors to detect the concentration of heavy metal ions in water, including cadmium, lead, and mercury.

[0013] Optionally, the data preprocessing includes: Noise filtering: Use moving average filtering algorithm to filter noise from collected environmental data; Standardization processing: Standardize the environmental data after noise filtering.

[0014] Optionally, the environmental change analysis and assessment includes: Environmental change trend analysis: A linear regression algorithm is used to perform trend analysis on the pre-processed environmental data to capture the dynamic changes of microclimate and sediments in the deep-sea mining area; Evaluation of the impact of dynamic changes on biological attachment bases: Based on the results of environmental change trend analysis and combined with the correlation analysis of the recovery capacity of biological attachment bases, the impact of environmental changes on biological attachment bases is evaluated. The correlation coefficient between environmental changes and the recovery capacity (growth rate) of biological attachment bases is calculated using the correlation analysis method. .

[0015] Beneficial effects of the present invention: The present invention comprehensively monitors key environmental parameters of deep-sea mining areas through a data acquisition module, a biological attachment base identification module and a deep-sea ecological dynamic monitoring module. The data acquisition module can efficiently and real-timely collect water quality data such as temperature, depth, and salinity, ensuring the timeliness and accuracy of the data, avoiding the data lag problem in traditional monitoring methods, and providing scientific and reliable basic data support for the evaluation of ecological restoration.

[0016] The present invention, through the biological attachment base recognition module based on image processing and convolutional neural network model, can not only collect and process the image data of the biological attachment base in real time, but also can identify the population distribution, density, morphology and other information of the biological attachment base through multi-scale feature extraction and pixel-level segmentation, and analyze the growth rate and health status of the biological attachment base in combination with water quality data, provide efficient and accurate distribution and recovery status of the biological attachment base, can accurately predict the progress and potential of ecological restoration, and provide strong data support for scientific decision-making and environmental protection measures for deep-sea ecological restoration.

[0017] The present invention, through environmental change analysis and evaluation, combined with real-time monitoring of microclimate and sediment dynamic changes, can dynamically analyze the impact of environmental changes on the recovery capacity of biological attachment bases, quantify the promoting or inhibiting effects of environmental factors through trend analysis and correlation evaluation, and timely identify key influencing factors, thereby achieving dynamic and accurate evaluation of ecological restoration status. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention; Figure 2 Schematic diagram of a deep-sea ecological dynamic monitoring module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0021] like Figure 1-Figure 2 As shown, the deep-sea mining ecological restoration capacity and potential evaluation system based on biological adhesion base includes a data acquisition module, a biological adhesion base identification module and a deep-sea ecological dynamic monitoring module, wherein; The data acquisition module collects water quality parameters in the deep-sea mining area through multiple sensors, including the temperature, depth, and salinity of the ocean water body; The biological attachment base identification module identifies and analyzes the population distribution, density, and morphology of biological attachment bases, and evaluates the growth and health status of biological attachment bases in combination with the collected water quality parameters; The deep-sea ecological dynamic monitoring module uses sensors to monitor the environmental data of the deep-sea mining area in real time, including microclimate (dissolved oxygen, pH) and sediment (particle size, heavy metal ions). By analyzing the dynamic changes of the microclimate and sediment in the mining area, the impact of environmental changes on the recovery capacity of biological attachment bases is evaluated, including: Environmental data collection: Use deep-sea sensors to collect various environmental data of deep-sea mining areas in real time, including dissolved oxygen, pH, particle size, and heavy metal ions; Data preprocessing: preprocess the collected environmental data, including noise filtering and standardization; Environmental change analysis and assessment: Analyze the pre-processed environmental data to assess the impact of dynamic changes in microclimate and sediments in deep-sea mining areas on the recovery capacity of biofouling substrates; Through the above content, environmental data of deep-sea mining areas are collected and analyzed in real time, covering microclimate changes and sediment changes. Combined with the type, distribution and growth stability analysis of biological attachment bases, a comprehensive assessment of ecological restoration is made, and dynamic analysis of environmental changes is conducted. This can accurately assess the impact of deep-sea mining activities on the recovery capacity of biological attachment bases, improve the timeliness, accuracy and foresight of the assessment, and provide more scientific and practical decision-making support for the ecological restoration of deep-sea mining areas.

[0022] The data acquisition module includes: Temperature data collection: Use temperature sensors to measure the temperature of water in deep-sea mining areas in real time and record changes in water temperature at each designated location and depth; Depth data collection: Use depth sensors to monitor the water depth of deep-sea mining areas in real time and capture depth changes in deep-sea waters; Salinity data collection: The salinity of water in deep-sea mining areas is measured using salinity sensors; Through the above, the temperature, depth changes and salinity fluctuations of the water body can be continuously monitored at different depths and locations. Through an efficient data acquisition system, environmental changes can be captured in real time to ensure the timeliness and accuracy of the data, avoid data lags and inaccuracies in traditional monitoring methods, and provide a scientific basis for the ecological impact assessment of deep-sea mining activities.

[0023] The biofouling base identification module includes: Biological adhesion base image acquisition and preliminary processing: The image data of biological adhesion base in the deep-sea mining area is collected in real time through underwater image acquisition equipment, and the collected image data is preliminarily processed, including denoising, image enhancement and segmentation; Identification of biological adhesion bases: The preliminarily processed image data is identified through the convolutional neural network (CNN) model to analyze the population distribution, density, and morphology of biological adhesion bases to determine the distribution of biological adhesion bases in different areas; Analysis of the growth and health status of biological attachments: Based on the results of biological attachment identification, the correlation between water quality parameters and the growth of biological attachments is analyzed to evaluate the growth rate and health status of biological attachments under different water quality conditions; Through the above content, the population distribution, density, and morphology of biological attachment bases can be identified and analyzed in real time, thereby effectively evaluating their growth and stability. By correlating water quality parameters with the growth status of biological attachment bases, the progress and potential of ecological restoration can be accurately predicted. More objective and real-time monitoring data can be provided to support scientific decision-making and environmental protection measures for ecological restoration in deep-sea mining areas, and promote the sustainable development of deep-sea mining activities.

[0024] Biological attachment base image acquisition and preliminary processing include: Underwater image acquisition: Real-time image data of biological attachment bases in deep-sea mining areas are collected through underwater cameras; Image denoising: Gaussian filtering algorithm is used to suppress noise in the collected image data to improve image quality and reduce interference caused by factors such as water flow and light changes, which can be expressed as: ; in, is the pixel value of the denoised image, is the relative pixel position in the image The neighborhood pixel value of is the Gaussian kernel, k is the radius of the filter window; Image enhancement: Histogram equalization is used to improve the contrast and details of the image data, making the morphological characteristics of the biological attachment base more obvious, expressed as: ; Among them, r is the input image pixel value, s is the enhanced image pixel value, is the maximum pixel value in the input image, is the minimum pixel value in the input image, and L is the grayscale level, which is used to determine the brightness range of the image; Image segmentation: The Otsu threshold method is used to separate the biological attachment base area in the image data from the background, which is expressed as: ; in, is the between-class variance, and are the class weights, and is the mean of the class; Through the above content, clear biological attachment base image data can be obtained in real time in complex deep-sea environments. Through denoising, image enhancement and segmentation, the image quality and clarity of biological features are improved, effectively overcoming insufficient deep-sea light and noise interference, providing higher reliability and accuracy for ecological monitoring, and improving the efficiency of identifying the distribution, density and health status of biological attachment bases.

[0025] The Convolutional Neural Network (CNN) model consists of: Multi-scale feature extraction: through convolution kernels of different scales ( , , ) to extract the multi-scale features of the image, and concatenate the convolution results of all scales to obtain a multi-scale feature map , the convolution operation at each scale is expressed as: ; in, is the convolution output of scale s, is the convolution kernel weight of scale s, is the input image at pixel location The value at Fully connected layer: multi-scale feature maps Flatten it into a one-dimensional feature vector F, and input the flattened feature vector F into the fully connected layer to generate a feature representation , expressed as: ; ; in, is the output feature vector of the fully connected layer, used for classification, is the weight matrix of the fully connected layer, F is the flattened multi-scale feature vector, is the bias of the fully connected layer; Pixel classification: for each pixel in the comprehensive feature map , use Softmax classifier for Classification is performed, expressed as: ; in, is the probability that the input image belongs to category c, is the score of the feature vector of the fully connected layer on category c, and k represents all categories; Segmentation result: The segmentation result is obtained by taking the maximum classification probability of each pixel , expressed as: ; in, Represents pixels The final classification result (biofouling matrix or background); Distribution and density calculation: The distribution and density information of biological attachment bases are calculated based on the segmentation results. The category information of each pixel is directly used to determine the spatial distribution of biological adhesion bases, and the density D of biological adhesion bases is calculated by counting the number of pixels belonging to the biological adhesion base category, which is expressed as: ; Where D is the density of the biofouling matrix, H and W are the height and width of the image, respectively. is the indicator function, when the pixel The value is 1 if it belongs to the biological attachment base category c, otherwise it is 0; Through the above content, efficient recognition and accurate segmentation of biological attachment bases in deep-sea environments are achieved. Multi-scale feature extraction can capture features of different scales and adapt to the morphological diversity of biological attachment bases. The fully connected layer further integrates these features to make classification more accurate. The Softmax classifier realizes category segmentation at the pixel level, ensuring the fine division of the image. It can not only identify the type of biological attachment base, but also output its distribution and density information, providing comprehensive ecological monitoring data support.

[0026] Analysis of biofouling growth and health status includes: Growth rate calculation: By analyzing the changes in the density of biological attachment bases at different time points, the growth rate G is calculated. represents the density of biological attachment base at time t, and the growth rate G is expressed as: ; Where G is the growth rate, the unit is density change / time, and Represents the density of biological attachment base at two different time points, is the time interval; Water quality correlation analysis: In order to evaluate the effect of environmental conditions on the stability of biological attachment bases, the Pearson correlation coefficient r was used to analyze the correlation between water quality parameters and biological attachment base density, expressed as: ; Among them, X is the water quality parameter, D is the density of biological attachment base, and are the means of X and D respectively; Health status assessment: Combine growth rate and water quality correlation results to determine the health threshold of growth rate. and unhealthy threshold and the correlation threshold Compare and comprehensively evaluate the health status of the biological attachment base. and , indicating health, if or , indicating unhealthy; Health threshold and unhealthy threshold It is determined by analyzing the growth rate distribution of biological attachment substrates under different environmental conditions, including: Data collection: Under various environmental conditions, collect the growth rate G data of the biological attachment base to form a sample data set ,in represents the growth rate of the i-th sampling; Calculate the mean and standard deviation of growth rate: Calculate the mean of growth rate based on the collected growth rate data set and standard deviation , expressed as: ; ; Where N is the number of data samples, is the growth rate of the i-th sample; Threshold setting: The healthy threshold is set to the mean plus one standard deviation, and the unhealthy threshold is set to the mean minus one standard deviation, expressed as: ; ; Relevance Threshold The settings include: Data collection: Collect data on the correlation between water quality parameters (temperature, salinity, pH) and biofouling density ,in is the correlation coefficient of the i-th sample; Calculate the mean and standard deviation of the correlation coefficient: Calculate the mean of the correlation coefficient based on the collected data and standard deviation , expressed as: ; ; Where M is the number of correlation samples, is the correlation coefficient of the i-th sample; Correlation threshold setting: Set the correlation threshold to the mean plus half the standard deviation, expressed as: ; Through the above content, the correlation between the growth rate of biological attachment base and environmental conditions is taken into account, which can dynamically respond to changes in different environments and accurately judge the health status of biological attachment base. The threshold setting based on data-driven can effectively avoid the deviation caused by human settings, and provide a reliable and accurate assessment basis for deep-sea ecological monitoring, which is helpful to make more scientific ecological protection and restoration decisions.

[0027] Environmental data collection includes: Dissolved oxygen data collection: Real-time collection of dissolved oxygen concentration in water through dissolved oxygen sensors; pH data collection: Use pH sensors to monitor the pH of water in real time; Particle size data collection: The particle size distribution of suspended particles in water is measured by particle size analysis sensors to provide water turbidity and particle information; Heavy metal ion data collection: Use heavy metal ion sensors to detect the concentration of heavy metal ions in water, including cadmium, lead, and mercury; Through the above content, not only can the changes in oxygen content and pH in the water be reflected, but also the suspended particles in the water and the level of heavy metal pollution can be captured, thereby providing comprehensive and in-depth support for deep-sea ecological monitoring, helping to timely detect environmental anomalies and providing strong data basis for ecological protection and risk management.

[0028] Data preprocessing includes: Noise filtering: The moving average filtering algorithm is used to filter the noise of the collected environmental data to remove abnormal data caused by sensor errors or environmental interference, which is expressed as: ; in, represents the data points after smoothing. is the original data point, q is the window radius, which determines the size of the smoothing window (that is, the window size is ); Standardization: The environmental data after noise filtering is standardized to eliminate the dimensional differences of different parameters, expressed as: ; in, represents the standardized data value, is the data point after noise filtering, and are the mean and standard deviation of the data respectively; Through the above content, the quality and consistency of the collected environmental data are effectively improved. Noise filtering can remove abnormal data caused by sensor errors or environmental interference, making the data more stable and reliable. Standardization processing adjusts the values ​​of different parameters to the same scale, eliminating dimensional differences and ensuring the accuracy and comparability of environmental data.

[0029] Environmental change analysis and assessment includes: Environmental change trend analysis: A linear regression algorithm was used to perform trend analysis on the preprocessed environmental data to capture the dynamic changes in microclimate and sediments in the deep-sea mining area, expressed as: ; in, is the environmental data value at time t, is the intercept, indicating the initial environmental state, The trend coefficient reflects the speed and direction of change of environmental data. , determine the rising or falling trend of environmental parameters; Evaluation of the impact of dynamic changes on biological attachment bases: Based on the results of environmental change trend analysis and combined with the correlation analysis of the recovery capacity of biological attachment bases, the impact of environmental changes on biological attachment bases is evaluated. The correlation coefficient between environmental changes and the recovery capacity (growth rate) of biological attachment bases is calculated using the correlation analysis method. , expressed as: ; in, is the correlation coefficient of environmental change, which is used to measure the trend value of environmental change The effect on the growth rate G of biological attachment base, is the environmental change trend value, indicating the change trend of environmental data within a certain period of time. G is the growth rate of the biological attachment base, defined as the density change rate, which is a key indicator of recovery ability. is the environmental change trend value The mean of The mean value of the growth rate G of the biofouling substrate; Through the above content, it is possible to accurately quantify the promoting or inhibiting effect of environmental changes on the recovery capacity of biological attachment bases, help identify key influencing factors, and through the combination of dynamic trend analysis and correlation assessment, it is possible to capture in real time the complex relationship between environmental conditions and ecological restoration, providing a comprehensive and reliable decision-making basis for the ecological management of deep-sea mining areas.

[0030] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0031] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment, characterized by: It includes data acquisition module, biological attachment base identification module and deep-sea ecological dynamic monitoring module, among which; The data acquisition module collects water quality parameters of the deep-sea mining area through multiple sensors, including the temperature, depth and salinity of the ocean water body; The biological attachment base identification module identifies and analyzes the population distribution, density, and morphology of the biological attachment base, and evaluates the growth and health status of the biological attachment base in combination with the collected water quality parameters; The deep-sea ecological dynamic monitoring module uses sensors to monitor the environmental data of the deep-sea mining area in real time, including microclimate and sediments. By analyzing the dynamic changes of the microclimate and sediments in the mining area, the impact of environmental changes on the recovery capacity of biological attachment bases is evaluated, including: Environmental data collection: Use deep-sea sensors to collect various environmental data of deep-sea mining areas in real time, including dissolved oxygen, pH, particle size, and heavy metal ions; Data preprocessing: preprocess the collected environmental data, including noise filtering and standardization; Analysis and assessment of environmental changes: Analyze the pre-processed environmental data to evaluate the impact of dynamic changes in microclimate and sediments in deep-sea mining areas on the recovery capacity of biofouling substrates.

2. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 1 is characterized in that: The data acquisition module comprises: Temperature data collection: Use temperature sensors to measure the temperature of water in deep-sea mining areas in real time and record changes in water temperature at each designated location and depth; Depth data collection: Use depth sensors to monitor the water depth of deep-sea mining areas in real time and capture depth changes in deep-sea waters; Salinity data collection: Salinity sensors are used to measure the salinity of water in deep-sea mining areas.

3. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 1 is characterized in that: The biological attachment base identification module comprises: Biological adhesion base image acquisition and preliminary processing: The image data of biological adhesion base in the deep-sea mining area is collected in real time through underwater image acquisition equipment, and the collected image data is preliminarily processed, including denoising, image enhancement and segmentation; Identification of biological adhesion bases: The preliminarily processed image data is identified through a convolutional neural network model to analyze the population distribution, density, and morphology of biological adhesion bases to determine the distribution of biological adhesion bases in different areas; Analysis of the growth and health status of biological attachments: Based on the results of biological attachment identification, the growth rate and health status of biological attachments under different water quality conditions are evaluated by analyzing the correlation between water quality parameters and the growth of biological attachments.

4. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 3 is characterized in that: The biological attachment base image acquisition and preliminary processing include: Underwater image acquisition: Real-time image data of biological attachment bases in deep-sea mining areas are collected through underwater cameras; Image denoising: Gaussian filtering algorithm is used to suppress noise in the collected image data; Image enhancement: Use histogram equalization to improve the contrast and details of image data; Image segmentation: The Otsu threshold method was used to separate the biofouling area from the background in the image data.

5. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 4 is characterized in that: The convolutional neural network model includes: Multi-scale feature extraction: extract the multi-scale features of the image through convolution kernels of different scales, and concatenate the convolution results of all scales to obtain a multi-scale feature map ; Fully connected layer: multi-scale feature maps Flatten it into a one-dimensional feature vector F, and input the flattened feature vector F into the fully connected layer to generate a feature representation ; Pixel classification: for each pixel in the comprehensive feature map , use Softmax classifier for To classify; Segmentation result: The segmentation result is obtained by taking the maximum classification probability of each pixel ; Distribution and density calculation: The distribution and density information of biological attachment bases are calculated based on the segmentation results. The category information of each pixel is directly used to determine the spatial distribution of biological adhesion bases, and the density D of biological adhesion bases is calculated by counting the number of pixels belonging to the biological adhesion base category.

6. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 5 is characterized in that: The biofouling base growth and health status analysis includes: Growth rate calculation: By analyzing the changes in the density of biological attachment bases at different time points, the growth rate G is calculated; Water quality correlation analysis: Pearson correlation coefficient r was used to analyze the correlation between water quality parameters and biofouling density; Health status assessment: Combine growth rate and water quality correlation results to determine the health threshold of growth rate. and unhealthy threshold and the correlation threshold Compare and comprehensively evaluate the health status of the biological attachment base. and , indicating health, if or , indicating unhealthy.

7. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 6 is characterized in that: The environmental data collection includes: Dissolved oxygen data collection: Real-time collection of dissolved oxygen concentration in water through dissolved oxygen sensors; pH data collection: Use pH sensors to monitor the pH of water in real time; Particle size data collection: The particle size distribution of suspended particles in water is measured by particle size analysis sensors to provide water turbidity and particle information; Heavy metal ion data collection: Use heavy metal ion sensors to detect the concentration of heavy metal ions in water, including cadmium, lead, and mercury.

8. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 7 is characterized in that: The data preprocessing includes: Noise filtering: Use moving average filtering algorithm to filter noise from collected environmental data; Standardization processing: Standardize the environmental data after noise filtering.

9. The deep-sea mining ecological restoration capacity and potential evaluation system based on biological attachment base according to claim 8 is characterized in that: The environmental change analysis and assessment includes: Environmental change trend analysis: A linear regression algorithm is used to perform trend analysis on the pre-processed environmental data to capture the dynamic changes of microclimate and sediments in the deep-sea mining area; Evaluation of the impact of dynamic changes on biological attachment bases: Based on the results of environmental change trend analysis and combined with the correlation analysis of the recovery capacity of biological attachment bases, the impact of environmental changes on biological attachment bases is evaluated. The correlation coefficient between environmental changes and the recovery capacity of biological attachment bases is calculated using the correlation analysis method. .

Images