Marine ecological evaluation system and method based on big data analysis

Abstract

The invention provides a marine ecological evaluation system and method based on big data analysis, and the method comprises the steps: obtaining marine biological diversity data and image data of a marine monitoring region, the marine biological diversity data comprising historical species distribution data and ecological niche indexes; performing streaming processing on the image data by using an edge computing architecture to obtain a biological activity feature map; according to the spatial resolution of the biological activity feature map and historical species distribution data, grid division is performed on the marine monitoring area, a reference species distribution density value of each grid unit is generated, and meanwhile, a species distribution density value of each grid unit is calculated; the negative deviation between the species distribution density value of each grid unit and the corresponding reference species distribution density value is calculated, and when the negative deviation is larger than a preset extinction threshold value, the corresponding grid unit is determined as a species distribution density attenuation area; according to the technical scheme provided by the invention, dynamic evaluation and early warning of the health condition of the marine ecosystem are realized.

Description

Technical Field

[0001] This application relates to the field of big data analysis technology, and particularly to an ocean ecological evaluation system and method based on big data analysis. Background Art

[0002] With the increasing impact of human activities and climate change on the marine ecosystem, there is an urgent need for an ecological monitoring and evaluation method that can achieve large-scale, high-frequency, and precise monitoring. Especially in key ecosystems such as coral reefs and seagrass beds, how to quickly identify changes in species distribution, judge the trend of ecological degradation, and provide dynamic basis for scientific restoration has become the core technical requirement of current ecological protection work.

[0003] In response to the above technical requirements, existing solutions combine multi-spectral optical remote sensing technology and cloud computing platforms to remotely sense and centrally analyze the ecological characteristics of the ocean surface. Existing solutions obtain multi-spectral images of ocean areas through satellites or drones, and combine with historical biodiversity databases to construct regional ecological health indices in the cloud, regularly generate ecological evaluation reports, and assist management departments in making macro decisions. However, existing solutions have some defects. For example, remote sensing images are difficult to accurately capture the distribution changes of small or hidden species, and the process of uploading data to the cloud for processing results in a lag in evaluation results, making it impossible to detect sudden ecological anomalies in a timely manner; the centralized computing architecture is prone to data backlog and processing bottlenecks when facing high-frequency monitoring tasks, affecting the timeliness and refinement of ecological evaluation. Summary of the Invention

[0004] This application provides an ocean ecological evaluation system and method based on big data analysis to solve the problems in the prior art, such as the inability to detect sudden ecological anomalies in a timely manner; the easy occurrence of data backlog and processing bottlenecks, and the insufficient timeliness and refinement of ecological evaluation.

[0005] In a first aspect, this application provides an ocean ecological evaluation method based on big data analysis, including: Obtaining marine biodiversity data and image data of a marine monitoring area, where the marine biodiversity data includes historical species distribution data and niche indicators; Using an edge computing architecture deployed in the marine monitoring area to perform streaming processing on the image data to obtain a biological activity feature map; According to the spatial resolution of the biological activity feature map and the historical species distribution data, dividing the marine monitoring area into grids, generating a benchmark species distribution density value for each grid unit, and calculating the species distribution density value for each grid unit; Calculate the negative deviation between the species distribution density value of each grid cell and the corresponding reference species distribution density value. When the negative deviation is greater than a preset extinction threshold, determine the corresponding grid cell as a species distribution density attenuation area; Generate an evaluation result of marine ecological integrity based on all the species distribution density attenuation areas.

[0006] Optionally, use the edge computing architecture deployed in the marine monitoring area to perform streaming processing on the image data to obtain a biological activity feature map, including: Use the edge computing architecture to separate the multi-spectral bands in the image data to obtain the reflection intensity data of each spectral channel; Strengthen the characteristic bands related to biological activities in the reflection intensity data of all spectral channels to generate enhanced spectral data for each spectral channel; Perform weighted superposition on the preset set of biological weight coefficients and the enhanced spectral data of different spectral channels to generate a preliminary biological activity intensity distribution map; Correct the discontinuous areas in the preliminary biological activity intensity distribution map to generate a corrected biological activity intensity distribution map; Convert the corrected biological activity intensity distribution map into a biological activity feature map according to the correlation between the historical species distribution data and spectral characteristics.

[0007] Optionally, perform weighted superposition on the preset set of biological weight coefficients and the enhanced spectral data of different spectral channels to generate a preliminary biological activity intensity distribution map, including: Based on the geographic coordinate data of the marine monitoring area, determine the ecosystem type to which the marine monitoring area belongs, and the ecosystem type includes coral reef ecosystem and seagrass bed ecosystem; Call the preset set of biological weight coefficients corresponding to the ecosystem type, and perform a multiplication operation on the enhanced spectral data of each pixel point in the enhanced spectral data of different spectral channels and the biological weight coefficients of the corresponding spectral channels in the preset set of biological weight coefficients to obtain the weighted spectral values of different spectral channels for each pixel point; Superpose all the weighted spectral values of the same pixel point to obtain the biological activity intensity value of each pixel point; Convert the biological activity intensity values of all pixel points into distribution image pixel values to generate a preliminary biological activity intensity distribution map according to the distribution image pixel values.

[0008] Optionally, convert the corrected biological activity intensity distribution map into a biological activity feature map according to the correlation between the historical species distribution data and spectral characteristics, including: Group the historical species distribution data based on the correlation between the species density and the spectral characteristics of the historical species distribution data, and the ecosystem type, to obtain a grouping result; Based on the grouping result, construct a species density spectral response lookup table, where the species density spectral response lookup table contains conversion rules corresponding to different ecosystem types; Based on the conversion rules, convert the biological activity intensity value of each pixel into a species density characteristic value, so as to generate a biological activity intensity distribution map according to all the species density characteristic values.

[0009] Optionally, according to the spatial resolution of the biological activity characteristic map and the historical species distribution data, divide the marine monitoring area into grids, generate the reference species distribution density value of each grid cell, and calculate the species distribution density value of each grid cell, including: According to the spatial resolution of the biological activity characteristic map, use a preset grid division rule to divide the marine monitoring area into multiple grid cells; Extract the historical observation points located within the grid cell from the historical species distribution data, and parse the data records of the historical observation points to obtain a set of species density measurement values corresponding to each grid cell; Sort the species density measurement values in each set of species density measurement values, select the target species density measurement value at the median, and use the target species density measurement value corresponding to each grid cell as the corresponding reference species distribution density value; Align the coordinate systems of the grid cell and the biological activity characteristic map, extract the set of biological activity characteristic values corresponding to each grid cell, and use the mean value of the characteristic values in each set of biological activity characteristic values as the species distribution density value of the corresponding grid cell.

[0010] Optionally, generate a marine ecological integrity evaluation result based on all the species distribution density attenuation regions, including: Calculate the proportion of the area of all the species distribution density attenuation regions in the marine monitoring area, and use the proportion as the attenuation area proportion value; Extract the corresponding negative deviation values from all the species distribution density attenuation regions, calculate the average value of all the negative deviation values, and generate an overall attenuation intensity value; Compare the attenuation area proportion value with a preset health level threshold table to obtain a health level determination result; Based on a preset risk index calculation rule, convert the overall attenuation intensity value into a risk index; Combine the health level determination result and the risk index as the marine ecological integrity evaluation result.

[0011] Optionally, based on a preset risk index calculation rule, converting the overall attenuation intensity value into a risk index includes: Determining a target conversion numerical range corresponding to the overall attenuation intensity value according to a preset attenuation intensity risk index mapping table; According to the ecosystem type, calling a benchmark proportionality coefficient corresponding to the target conversion numerical range in the preset risk index calculation rule to perform proportional scaling on the overall attenuation intensity value to obtain a scaled attenuation intensity value; Based on the marine biodiversity data, selecting a target compensation parameter from the environmental dynamic compensation parameters of the preset risk index calculation rule; Superimposing the scaled attenuation intensity value and the target compensation parameter to generate a risk index.

[0012] In a second aspect, the present application provides a marine ecological evaluation system based on big data analysis, including: An acquisition module for acquiring marine biodiversity data and image data of a marine monitoring area, where the marine biodiversity data includes historical species distribution data and niche indicators; A processing module for performing streaming processing on the image data by using an edge computing architecture deployed in the marine monitoring area to obtain a biological activity feature map; A partitioning module for partitioning the marine monitoring area according to the spatial resolution of the biological activity feature map and the historical species distribution data to generate a reference species distribution density value for each grid unit, and simultaneously calculating the species distribution density value for each grid unit; A determination module for calculating a negative deviation between the species distribution density value of each grid unit and the corresponding reference species distribution density value, and when the negative deviation is greater than a preset extinction threshold, determining the corresponding grid unit as a species distribution density attenuation area; An evaluation module for generating a marine ecological integrity evaluation result based on all the species distribution density attenuation areas.

[0013] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a marine ecological evaluation method based on big data analysis as described in the first aspect above.

[0014] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it implements a marine ecological evaluation method based on big data analysis as described in the first aspect.

[0015] In this application, marine biodiversity data and image data of a marine monitoring area are obtained. The marine biodiversity data includes historical species distribution data and niche indicators. The edge computing architecture deployed in the marine monitoring area is used to perform streaming processing on the image data to obtain a biological activity feature map. According to the spatial resolution of the biological activity feature map and the historical species distribution data, the marine monitoring area is divided into grids to generate the baseline species distribution density value of each grid cell, and at the same time, the species distribution density value of each grid cell is calculated. The negative deviation between the species distribution density value of each grid cell and the corresponding baseline species distribution density value is calculated. When the negative deviation is greater than a preset extinction threshold, the corresponding grid cell is determined as a species distribution density attenuation area. Based on all the species distribution density attenuation areas, a marine ecological integrity evaluation result is generated. The technical solution provided by this application obtains marine biodiversity data and image data of a marine monitoring area; uses the edge computing architecture deployed near the monitoring area to perform streaming processing on the image data, improving the real-time performance and efficiency of data processing and reducing the latency problem caused by traditional centralized processing; divides the monitoring area into grids according to the spatial resolution of the biological activity feature map and the historical species distribution data, achieving refined modeling and dynamic comparison on the spatial scale; by comparing the negative deviation between the current density value and the baseline value and identifying the decreasing grid cells, all attenuation areas are obtained to generate a marine ecological integrity evaluation result. Among them, through band separation and enhancement processing of multi-spectral image data, key spectral information related to biological activities is extracted, and weighted superposition is performed in combination with a preset biological weight coefficient to form a preliminary biological activity intensity distribution map; subsequently, the discontinuous areas in the image are corrected to generate a biological activity feature map with ecological semantics, improving the extraction accuracy and ecological interpretation ability of biological activity information in remote sensing images and solving the problem that it is difficult to identify small or hidden species due to insufficient spatial resolution in traditional remote sensing means; by completing key image processing tasks at the edge, the processing latency caused by uploading the original image to the cloud is avoided, improving the response speed of the system to sudden ecological anomaly events and the local detail recognition ability, thus breaking through the technical barriers of existing centralized remote sensing monitoring in terms of timeliness and fineness.

[0016] These aspects or other aspects of this application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of this application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 shows a flowchart of a marine ecological assessment method provided by the present application based on big data analysis; Figure 2 shows a schematic structural diagram of a marine ecological assessment system provided by the present application based on big data analysis; Figure 3 shows a schematic structural diagram of a computing device provided by the present application. Detailed implementation manners

[0019] In order to enable those skilled in the art of the present technology to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application.

[0020] In some processes described in the specification, claims and the above-mentioned drawings of the present application, a plurality of operations that appear in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0021] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0022] Aiming at the limitations of the existing technical solutions based on multispectral remote sensing and cloud computing in terms of response speed, spatial resolution and processing efficiency, it is difficult to meet the requirements of high-frequency and refined monitoring of the dynamic changes in species distribution. To solve these problems, the present application proposes a marine ecological assessment method based on big data analysis, which realizes near-real-time processing of the entire process from data collection to ecological assessment by integrating edge computing and biodiversity data analysis technologies, improves the timeliness and accuracy of ecological anomaly recognition, and makes up for the deficiencies of the traditional centralized processing mode in dealing with complex and changeable marine environments.

[0023] Figure 1 For the embodiments of the present application, a flowchart of a marine ecological assessment method based on big data analysis is provided, asFigure 1 As shown in Figure 1 , the method includes: Step 101: Obtain the marine biodiversity data and image data of the marine monitoring area, where the marine biodiversity data includes historical species distribution data and niche indicators.

[0024] In this step, the marine biodiversity data refers to the data set reflecting the species composition and distribution characteristics of the marine ecosystem, including historical species distribution data (such as coral coverage rate, fish abundance) and niche indicators (such as species competition index, habitat fitness score). The image data refers to the multi-spectral marine surface images obtained by optical remote sensing equipment, which are used to capture the chlorophyll concentration of plankton, the characteristics of coral symbiotic algae, and the characteristics of seagrass canopies. The historical species distribution data refers to the verified species spatial distribution records by authoritative institutions, including coordinate positions, survey times, and species density values (such as coral coverage percentage, seagrass biomass g / m²). The niche indicators refer to the data parameters quantifying the functional status of species in the ecosystem, including trophic level index, niche width, and competition exclusion coefficient, which are calculated through the species interaction relationships recorded in historical surveys.

[0025] In the embodiment of the present application, through the satellite remote sensing receiving station and the marine monitoring buoy network, the multi-spectral optical image data of the marine monitoring area and the historical ecological survey database are synchronously collected. The marine biodiversity data is obtained by calling the national marine species database, including historical species distribution data and niche indicators. The image data is provided by high-resolution satellites and unmanned aerial vehicle aerial photography systems, covering the visible light to near-infrared bands.

[0026] Step 102: Use the edge computing architecture deployed in the marine monitoring area to perform streaming processing on the image data to obtain a biological activity feature map.

[0027] In this step, the edge computing architecture refers to the distributed computing system deployed in the marine monitoring area, which is composed of anti-salt corrosion treatment nodes and provides four functional modules: spectral decomposition, feature enhancement, data fusion, and spatial optimization. The biological activity feature map refers to the spatial distribution map characterizing the biological activity intensity per unit area of the ocean, which is generated by converting the image data into a single-channel intensity value matrix.

[0028] In the embodiment of the present application, use the spectral decomposition module in the edge computing architecture deployed in the marine monitoring area to decompose the image data into independent channels, and use the marine biological feature enhancement technology to enhance the reflection intensity of some bands in the coral reef area and seagrass bed area. The enhanced multi-channel data is processed by the weighted fusion module, and then the wave interference and light gradient are eliminated through the spatial optimization module, and finally a biological activity feature map is generated.

[0029] Step 103: According to the spatial resolution of the biological activity feature map and the historical species distribution data, divide the marine monitoring area into grids to generate the baseline species distribution density value of each grid cell, and at the same time calculate the species distribution density value of each grid cell.

[0030] In this step, the spatial resolution refers to the actual sea surface size corresponding to a single pixel in the biological activity feature map, which is used to constrain the grid division granularity. The baseline species distribution density value refers to the statistical aggregation result of the historical species density measurement values within the grid cell. For the coral reef area, the median value of the live coral coverage rate is taken, and for the seagrass bed area, the water depth weighted average value of the leaf biomass is taken. The species distribution density value refers to the species density of the grid cell calculated in real time through the biological activity feature map, which is generated by the arithmetic mean of all pixel feature values within the cell. In the coral reef area, it is associated with the reflectance intensity of symbiotic algae, and in the seagrass bed area, it is associated with the near-infrared reflectance of the canopy.

[0031] In the embodiment of the present application, grid division is performed according to the spatial resolution of the biological activity feature map and the historical species distribution data. The grid size is set according to the resolution of the feature map, and then the density measurement values of the historical species distribution data within each grid cell are extracted. At the same time, the arithmetic mean of all pixel feature values within each grid cell is extracted from the biological activity feature map to generate the baseline species distribution density value and the species distribution density value of each grid cell respectively.

[0032] Step 104: Calculate the negative deviation between the species distribution density value of each grid cell and the corresponding baseline species distribution density value. When the negative deviation is greater than the preset extinction threshold, the corresponding grid cell is determined as the species distribution density attenuation area.

[0033] In this step, the negative deviation refers to the relative deviation amplitude of the current species distribution density value lower than the baseline value. The preset extinction threshold refers to the critical value matrix for determining species attenuation, including the coral reef dynamic threshold and the seagrass bed stratification threshold. The species distribution density attenuation area refers to the set of grid cells with negative deviation exceeding the threshold, carrying metadata such as coordinate range, attenuation intensity, and species type, which is used to generate the coral bleaching heat map or the seagrass degradation distribution map.

[0034] In the embodiment of the present application, for each grid cell, the density difference is obtained by subtracting the species distribution density value from the baseline species distribution density value. When the density difference is greater than 0, calculate the proportion of the density difference to the baseline species distribution density value, and use this proportion as the negative deviation. Compare the negative deviation with the preset extinction threshold. If it exceeds the threshold, mark the grid as the species distribution density attenuation area.

[0035] Step 105: Generate the marine ecological integrity evaluation result based on all the species distribution density attenuation areas.

[0036] In this step, the marine ecological integrity evaluation result refers to a structured report including the health level and risk index, encapsulating parameters such as the proportion of the attenuation area, the attenuation rate of core species, and the weights of environmental factors.

[0037] In the embodiment of the present application, the ratio of the total area of the species distribution density attenuation region to the area of the marine monitoring region is calculated to obtain the proportion value of the attenuation area, and the negative deviation values of all species distribution density attenuation regions are extracted and averaged as the overall attenuation intensity value. The area proportion value is mapped to the health level through the health level threshold table, and the attenuation intensity value is converted to the risk index through the risk index calculation rule. Finally, the health level and the risk index are combined to obtain the marine ecological integrity evaluation result.

[0038] The embodiment of the present application solves the problem that the traditional method cannot balance large-scale monitoring and species-level accuracy, and provides near-real-time ecological risk early warning for marine protected areas.

[0039] The present application provides a specific embodiment, step 102, using the edge computing architecture deployed in the marine monitoring region to perform streaming processing on the image data to obtain a biological activity feature map, which specifically includes the following steps: Step 201: Using the edge computing architecture, separate the multi-spectral bands in the image data to obtain the reflection intensity data of each spectral channel.

[0040] In this step, the multi-spectral band refers to the spectral interval divided by wavelength in the image data. The reflection intensity data of each spectral channel refers to the light reflection energy data set of a single band after separation, which is used to quantify the spectral response characteristics of different ground objects.

[0041] In the embodiment of the present application, when separating the multi-spectral bands in the image data using the edge computing architecture, the satellite image file is parsed through the spectral processing module of the edge node, and the composite spectral data is decomposed into independent spectral channels to generate a reflection intensity data matrix for each spectral channel. Each matrix element stores the reflection intensity data of each spectral channel.

[0042] Step 202: Strengthen the characteristic bands related to biological activities in the reflection intensity data of all spectral channels to generate enhanced spectral data for each spectral channel.

[0043] In this step, the characteristic bands related to biological activities refer to the sensitive spectral intervals that characterize biological activity in a specific ecosystem. The enhanced spectral data of each spectral channel refers to the spectral data matrix after radiation enhancement processing, which is achieved through dynamic range expansion and gain amplification.

[0044] In the embodiments of the present application, the ecosystem type is first identified, and infrared bands are selected for the coral reef area and the seagrass bed area respectively. Through the gain amplification technology, the reflection intensity data of all spectral channels are linearly enhanced, and at the same time, the dynamic range of non-characteristic bands is compressed, and the enhanced data is used to generate enhanced spectral data for each spectral channel.

[0045] Step 203: Weightedly superimpose the preset biological weight coefficient set and the enhanced spectral data of different spectral channels to generate a preliminary biological activity intensity distribution map.

[0046] In this step, the preset biological weight coefficient set refers to a set of spectral fusion parameters pre-configured according to the ecosystem, which is stored in the local database of the edge node. The preliminary biological activity intensity distribution map refers to a single-channel intensity spatial distribution map generated by weighted superposition.

[0047] In the embodiments of the present application, the preset biological weight coefficient set is called. The red light dominant coefficient set is applied to the coral reef area, and the near-infrared dominant coefficient set is applied to the seagrass bed area. Channel value weighted calculation is performed with the enhanced spectral data of different spectral channels. The formula is to multiply the enhanced spectral data value of each channel by the corresponding weight coefficient, and then add up the weighted values of all channels to generate the biological activity intensity value at this position. Finally, all position intensity values are integrated to obtain a preliminary biological activity intensity distribution map.

[0048] Step 204: Correct the discontinuous regions in the preliminary biological activity intensity distribution map to generate a corrected biological activity intensity distribution map.

[0049] In this step, the corrected biological activity intensity distribution map refers to an optimized distribution map after eliminating environmental interference, which is achieved by median filtering to eliminate wave mutations and radial gradient compensation.

[0050] In the embodiments of the present application, first, the mutations of adjacent pixels caused by the undulation of sea waves in the preliminary biological activity intensity distribution map are detected, and median filtering smoothing processing of the pixel window is adopted; secondly, a radial brightness compensation model is established for the radiation gradient caused by the change of the solar altitude angle to correct the discontinuous regions and generate a corrected biological activity intensity distribution map.

[0051] Step 205: Convert the corrected biological activity intensity distribution map into a biological activity feature map according to the correlation between the historical species distribution data and the spectral characteristics.

[0052] In this step, the spectral characteristics refer to the corresponding relationship between the reflectivity and biomass recorded in the historical species distribution data, including the linear model of coral reef red light reflectivity and coverage rate and the logarithmic model of seagrass bed near-infrared reflectivity and biomass, which are used for the conversion from intensity value to biological parameters.

[0053] In the embodiments of the present application, historical species distribution data and spectral characteristics are called to construct a conversion table for red light intensity and coral coverage rate, and a conversion table for near-infrared intensity and seagrass biomass. The intensity values at each position in the distribution map are substituted into the linear formula of the corresponding conversion table through coordinate matching, and finally the corrected biological activity intensity distribution map is converted into a biological activity characteristic map.

[0054] In the embodiments of the present application, edge computing is used to achieve real-time hierarchical processing of multispectral data, breaking through two major bottlenecks of traditional remote sensing monitoring.

[0055] The present application provides a specific embodiment, step 203, of weighted superposition of a preset set of biological weight coefficients and enhanced spectral data of different spectral channels to generate a preliminary biological activity intensity distribution map, which specifically includes the following steps: Step 211: Based on the geographical coordinate data of the marine monitoring area, determine the type of ecosystem to which the marine monitoring area belongs, and the types of ecosystems include coral reef ecosystems and seagrass bed ecosystems.

[0056] In this step, the type of ecosystem refers to the ecological classification of the marine monitoring area, including coral reef ecosystems and seagrass bed ecosystems. A coral reef ecosystem refers to an ecosystem with scleractinian corals as the framework, characterized by a three-dimensional reef structure, and includes reef-building species such as Acropora and Porites. A seagrass bed ecosystem refers to an ecosystem with hydrocharitaceae plants as the dominant species, forming a strip-shaped or patchy meadow.

[0057] In the embodiments of the present application, when determining the type of ecosystem based on the geographical coordinate data of the marine monitoring area, by querying a preset geographical information database, the coordinates are matched with the distribution maps of coral reef ecosystems and seagrass bed ecosystems. If the coordinates fall within the spatial boundary of the coral reef, it is marked as a coral reef ecosystem, and if they fall within the isobath range of the seagrass bed, it is marked as a seagrass bed ecosystem.

[0058] Step 212: Call the preset set of biological weight coefficients corresponding to the type of ecosystem, and perform a multiplication operation on the enhanced spectral data of each pixel point in the enhanced spectral data of different spectral channels and the biological weight coefficients of the corresponding spectral channels in the preset set of biological weight coefficients to obtain the weighted spectral values of different spectral channels for each pixel point.

[0059] In this step, the enhanced spectral data of each pixel point refers to a four-dimensional vector after radiation enhancement processing, and the data format is 32-bit floating-point type. The biological weight coefficient refers to the weight parameter of spectral fusion. The weighted spectral values of different spectral channels for each pixel point refer to the intermediate data generated by the multiplication operation and are used for subsequent accumulation calculations.

[0060] In the embodiments of the present application, when calling the preset biological weight coefficient set corresponding to the ecosystem type, the coral reef ecosystem loads the red light dominant coefficient set, and the seagrass bed ecosystem loads the near-infrared dominant coefficient set. Perform a product operation on the enhanced spectral data of each pixel point in the enhanced spectral data of different spectral channels. The formula is to multiply the value of each spectral channel by the corresponding weight coefficient to generate the weighted spectral values of different spectral channels for this pixel point.

[0061] Step 213: Superimpose all the weighted spectral values of the same pixel point to obtain the biological activity intensity value of each pixel point.

[0062] In this step, the biological activity intensity value of each pixel point refers to a scalar representing biological activity.

[0063] In the embodiments of the present application, when superimposing all the weighted spectral values of the same pixel point, an arithmetic accumulation algorithm is adopted. The formula is that the biological activity intensity value is equal to the red light weighted value plus the green light weighted value plus the blue light weighted value plus the near-infrared weighted value. The calculation result is stored as the biological activity intensity value of each pixel point.

[0064] Step 214: Convert the biological activity intensity values of all pixel points into distribution image pixel values to generate a preliminary biological activity intensity distribution map based on the distribution image pixel values.

[0065] In this step, the distribution image pixel value refers to the gray value of the preliminary biological activity intensity distribution map, which is obtained by linearly mapping the biological activity intensity value to an integer between 0 and 255.

[0066] In the embodiments of the present application, first linearly scale the biological activity intensity values of all pixel points to obtain the distribution image pixel values, that is, the distribution image pixel value = intensity value ÷ maximum intensity value × A, where A is a constant with a range between 0 and 255. Using the digital elevation model, project the geographical coordinates of the coral reef area onto a two-dimensional image grid, and at the same time assign values to the seagrass bed area according to the isobaths, and finally generate a preliminary biological activity intensity distribution map with the same spatial resolution as the original image.

[0067] The embodiments of the present application break through the triple technical barriers of traditional remote sensing fusion through dynamic loading of the weight set by the ecosystem, pixel-level product operation to retain spatial heterogeneity, and edge node fixed-point number optimization technology.

[0068] The present application provides a specific embodiment, step 205. According to the correlation between the historical species distribution data and the spectral characteristics, convert the corrected biological activity intensity distribution map into a biological activity characteristic map, which specifically includes the following steps: Step 221: Group the historical species distribution data based on the correlation between the species density of the historical species distribution data and the spectral characteristics, and the ecosystem type to obtain a grouping result.

[0069] In this step, species density refers to the number or coverage ratio of target species per unit area, including the percentage of coverage reflecting the area of live corals in the coral reef area, and grams per square meter reflecting leaf biomass in the seagrass bed area. The grouping result refers to a structured data set divided by ecosystem type and spectral intensity range, including a coral reef data set and a seagrass bed data set. Each group contains intensity boundary values, average species density, and the number of samples.

[0070] In the embodiment of the present application, the spectral reflection intensity values of the species density and its corresponding spectral characteristics recorded in the historical species distribution data are extracted; according to the ecosystem type, the historical species distribution data is divided into independent data sets; divided into a preset interval group according to the spectral reflection intensity values; finally, a grouping result including ecosystem attributes and intensity intervals is output.

[0071] Step 222: Based on the grouping result, construct a species density spectral response lookup table, and the species density spectral response lookup table contains conversion rules corresponding to different ecosystem types.

[0072] In this step, the species density spectral response lookup table refers to a database storing the mapping relationship between spectral intensity and species density, including coral reef type entries and seagrass bed type entries, generated through statistical analysis based on the grouping result, and used to guide pixel-level density conversion. The conversion rules corresponding to different ecosystem types refer to ecosystem-specific intensity and density conversion formulas, including the coral reef rule that the red light intensity value is multiplied by a coefficient of 0.25 and then added with a constant of 5, and the seagrass bed rule that the near-infrared intensity value is multiplied by a coefficient of 0.3 and then added with a constant of 10.

[0073] In the embodiment of the present application, based on each grouping result, the average value of the measured species density of each group is calculated as the reference density; the mapping relationship between the intensity interval boundary value and the reference density is established; the mapping relationship is integrated into lookup table entries according to the ecosystem type; finally, a species density spectral response lookup table containing coral reef and seagrass bed specific conversion rules is generated.

[0074] Step 223: Based on the conversion rule, convert the biological activity intensity value of each pixel point into a species density characteristic value, so as to generate a biological activity intensity distribution map according to all species density characteristic values.

[0075] In this step, the species density characteristic value refers to a quantitative value representing the species distribution density per unit area, including the percentage of coverage of coral reef pixels and the biomass value of seagrass bed pixels. The biological activity intensity distribution map refers to a geographical distribution map in the form of a two-dimensional space matrix, including a grid structure with the same resolution as the original image, and each grid unit stores the species density characteristic value, which is used to visually display the species distribution heat map.

[0076] In the embodiment of the present application, based on the conversion rule, for each pixel point of the corrected biological activity intensity distribution map, its biological activity intensity value is read; the ecosystem type is determined according to the pixel coordinates; the conversion rule corresponding to the intensity interval is matched from the species density spectral response lookup table; the rule is applied to linearly convert the biological activity intensity value into a species density characteristic value (for example, the biological activity intensity value of 120 is converted into a coral coverage rate of 30%); the species density characteristic values of all pixel points are reorganized according to the original spatial position to generate a biological activity intensity distribution map.

[0077] The embodiment of the present application realizes the accurate conversion from pixel-level intensity to species density, and solves the core problem that the biological activity intensity in marine remote sensing monitoring cannot directly reflect the real species distribution; through the differential processing of the red light response of coral reefs and the near-infrared response of seagrass beds, the inversion accuracy of coral coverage and seagrass biomass is improved, providing a reliable data basis for ecological evaluation.

[0078] The present application provides a specific embodiment, step 103, according to the spatial resolution of the biological activity characteristic map and the historical species distribution data, divide the marine monitoring area into grids, generate the reference species distribution density value of each grid unit, and calculate the species distribution density value of each grid unit at the same time, specifically including the following steps: Step 301: According to the spatial resolution of the biological activity characteristic map, use a preset grid division rule to divide the marine monitoring area into multiple grid units.

[0079] In this step, the preset grid division rule refers to a grid generation strategy that dynamically adapts to the characteristics of the ecosystem, and is an automatic selection rule based on the terrain database. The multiple grid units refer to the spatial segmentation units covering the marine monitoring area, including closed polygons with unique geographical codes. Each unit stores boundary coordinates, ecosystem types and associated data, and is used to organize density distribution data.

[0080] In the embodiment of the present application, according to the spatial resolution of the biological activity characteristic map, the pixel size of the biological activity characteristic map is read, and a preset grid division rule library is called. For the coral reef ecosystem, an irregular triangular grid is adopted, and for the seagrass bed ecosystem, a rectangular grid is adopted; the marine monitoring area is cut into specified grids through a geographic information system engine, and multiple grid units covering the whole area are output.

[0081] Step 302: Extract the historical observation points located in the grid unit from the historical species distribution data, and analyze the data records of the historical observation points to obtain the set of species density measurement values corresponding to each grid unit.

[0082] In this step, the historical observation points refer to the discrete geographical location points recorded in historical surveys, including the track coordinates of the survey ship and the diving observation stations, carrying time stamps, measured values of species density, and water depth information, which are used to provide a benchmark density data source. The set of species density measurement values refers to the density data groups of all historical observation points within the same grid cell, including the percentage coverage sequence of the coral reef unit or the biomass sequence of the seagrass bed unit, which are used for statistical aggregation.

[0083] In the embodiment of the present application, traverse the geographical boundaries of each grid cell, and screen the historical observation points whose coordinates fall within the boundaries in the historical species distribution data; analyze the species density fields recorded in the historical observation point data, and summarize all valid values into the set of species density measurement values corresponding to each grid cell.

[0084] Step 303: Sort the species density measurement values in each set of species density measurement values to select the target species density measurement value at the median, and use the target species density measurement value corresponding to each grid cell as the corresponding benchmark species distribution density value.

[0085] In this step, the target species density measurement value refers to the median value after sorting the set of species density measurement values, which reflects the typical ecological state of the grid cell.

[0086] In the embodiment of the present application, sort the set of species density measurement values of each grid cell in ascending order of numerical value. If the data volume is odd, take the middle value; if it is even, take the average of the two middle values; exclude the interference of outliers, and use the target species density measurement value corresponding to each grid cell as the corresponding benchmark species distribution density value.

[0087] Step 304: Align the coordinate systems of the grid cell and the biological activity feature map to extract the set of biological activity feature values corresponding to each grid cell, and use the feature mean value in each set of biological activity feature values as the species distribution density value of the corresponding grid cell.

[0088] In this step, the set of biological activity feature values refers to all pixel values in the image area corresponding to the grid cell, including the reflection intensity sequence of the pixels in the biological activity feature map, which is used to calculate the real-time density. The feature mean value refers to the arithmetic mean value of the set of biological activity feature values, which represents the overall biological activity intensity of the unit.

[0089] In the embodiment of the present application, through coordinate system conversion, map the grid cell boundary onto the biological activity feature map; extract all biological activity feature values within the coverage of each grid; calculate the feature mean value of the pixel values. The calculation formula of the feature mean value is: the sum of all pixel values within the grid divided by the number of pixels. For example, (152 + 168 + 142) / 3 = 154, to generate the current species distribution density value.

[0090] The embodiments of this application achieve the spatial alignment of multi-source data through ecosystem-adaptive grid division, eliminate the abnormal interference of historical data by median statistics, and accurately quantify the real-time species distribution density by combining pixel mean calculation, solving the data deviation problem caused by grid mismatch in traditional methods; the differential design of coral reef triangular grids and seagrass bed rectangular grids improves the spatial representation accuracy of coral coverage and seagrass biomass.

[0091] This application provides a specific embodiment. Step 105: Generate the marine ecological integrity evaluation result based on all species distribution density attenuation regions, which specifically includes the following steps: Step 501: Calculate the proportion of the area of all species distribution density attenuation regions in the marine monitoring region, and take this proportion as the attenuation area proportion value.

[0092] In this step, the attenuation area proportion value refers to a quantitative indicator of the proportion of the species decline region, including the percentage value reflecting the bleached area in the coral reef region.

[0093] In the embodiments of this application, the grid cell areas of all species distribution density attenuation regions are accumulated through a geographic information system, and the accumulated value is divided by the total area of the marine monitoring region to obtain the attenuation area proportion value, that is, the attenuation area proportion value = (total area of the attenuation region ÷ total area of the monitoring region)%; where the coral reef region is calculated according to the three-dimensional surface area of the reef flat, and the seagrass bed region is calculated according to the plane area corrected by the isobath.

[0094] Step 502: Extract the corresponding negative deviation values from all species distribution density attenuation regions, calculate the average value of all negative deviation values, and generate an overall attenuation intensity value.

[0095] In this step, the negative deviation value refers to the deviation amplitude of the real-time density of the grid cell being lower than the historical benchmark, including the negative deviation of coral coverage, which is used to quantify the single-point decline intensity. The overall attenuation intensity value refers to a comprehensive representation value of the overall decline degree of the region, including the arithmetic mean of all negative deviation values. The seagrass bed region is corrected by the tidal phase, reflecting the overall pressure level of the ecosystem.

[0096] In the embodiments of this application, traverse each attenuation region metadata field, and read the pre-stored negative deviation values from all species distribution density attenuation regions (such as taking the symbiotic algae deviation value in the coral reef region and the frond biomass deviation value in the seagrass bed region); perform arithmetic mean calculation on all negative deviation values, and additionally introduce a tidal influence coefficient weighting in the seagrass bed region to obtain an overall attenuation intensity value reflecting the overall decline degree.

[0097] Step 503: Compare the attenuation area proportion value with a preset health level threshold table to obtain a health level determination result.

[0098] In this step, the preset health level threshold table refers to the area ratio and health status mapping standard specific to the ecosystem, including the three-level threshold table for coral reefs and the four-level threshold table for seagrass beds, which are used for objective grading. The health level determination result refers to the qualitative evaluation output by threshold comparison, including text conclusions such as healthy, degraded, and endangered, and defines the semantics in combination with the standards of the International Union for Conservation of Nature.

[0099] In the embodiment of the present application, according to the current ecosystem type, the preset health level threshold table is loaded, and the attenuation area ratio value is matched to the corresponding interval (if the coral reef ratio is 18%, it is determined as the degraded level), and the health level determination result.

[0100] Step 504: Based on the preset risk index calculation rule, convert the overall attenuation intensity value into a risk index.

[0101] In this step, the preset risk index calculation rule refers to the conversion logic from the intensity value to the risk score, including the linear function for coral reefs and the piecewise function for seagrass beds. The risk index refers to the numerical expression of the ecological risk degree, including a continuous scale from 0 to 100 points. A yellow warning is triggered when the score is above 60, and a red emergency response is triggered when the score is above 80.

[0102] In the embodiment of the present application, based on the ecosystem-specific piecewise linear conversion in the preset risk index calculation rule, for example, in the coral reef area, the intensity value is multiplied by the coefficient 100 to generate the basic risk value, and the real-time water temperature compensation is superimposed (add 5 points for each 1℃ exceeding the benchmark); in the seagrass bed area, the intensity value is multiplied by the coefficient 80 and then the turbidity compensation is added (add 10 points for each 5NTU exceeding), and the quantitative risk index is output.

[0103] Step 505: Combine the health level determination result and the risk index to obtain the marine ecological integrity evaluation result.

[0104] In the embodiment of the present application, a structured data object is created, and the health level determination result and the risk index value are encapsulated as key-value pairs (such as the health level is degraded, and the risk index is 65), and the link to the spatial distribution heat map is attached to generate the marine ecological integrity evaluation result that can be directly output to the supervision platform.

[0105] The embodiment of the present application breaks through the limitations of traditional single indicators through a two-dimensional evaluation mechanism that couples the area ratio and the decline intensity, and combines the ecological system dynamic compensation algorithm; the design of water temperature compensation for coral reefs and turbidity correction for seagrass beds improves the accuracy of early warning for extreme events in tropical oceans and provides a precise determination of the priority of actions for ecological restoration.

[0106] The present application provides a specific embodiment. In step 504, based on the preset risk index calculation rule, convert the overall attenuation intensity value into a risk index, which specifically includes the following steps: Step 511: Determine the target conversion value interval corresponding to the overall attenuation intensity value according to a preset attenuation intensity risk index mapping table.

[0107] In this step, the preset attenuation intensity risk index mapping table refers to a structured standard that defines the relationship between intensity intervals and risk levels, including a three-level interval table for coral reef types and a four-level interval table for seagrass bed types, which is generated based on historical ecological disaster data modeling. The target conversion value interval refers to the continuous numerical range matched by the overall attenuation intensity value, including closed boundaries with interval identification codes, used to activate corresponding calculation rules.

[0108] In the embodiment of the present application, the attenuation intensity risk index mapping table stored in the rule library is called, and the overall attenuation intensity value (such as 0.45) is compared with the boundaries of each interval in the preset attenuation intensity risk index mapping table to lock the corresponding target conversion value interval, such as 0.45 ∈ [0.3, 0.6].

[0109] Step 512: According to the ecosystem type, call the benchmark proportionality coefficient corresponding to the target conversion value interval in the preset risk index calculation rule, and perform proportional scaling on the overall attenuation intensity value to obtain the scaled attenuation intensity value.

[0110] In this step, the benchmark proportionality coefficient refers to a multiplier factor that linearly converts the intensity value into a risk benchmark, which is determined by the joint matching of the interval and the ecosystem. The scaled attenuation intensity value refers to the intermediate result value after being processed by the proportionality coefficient, reflecting the basic risk level.

[0111] In the embodiment of the present application, according to the current ecosystem type (such as coral reef, seagrass bed), the preset risk index calculation rule is loaded, the target conversion value interval (such as the medium-risk area) is matched to obtain the benchmark proportionality coefficient (such as 80 for coral reefs and 70 for seagrass beds); the overall attenuation intensity value is multiplied by the benchmark proportionality coefficient (such as 0.45×80 = 36 in the coral reef area) to generate the scaled attenuation intensity value.

[0112] Step 513: Based on the marine biodiversity data, select the target compensation parameter from the environmental dynamic compensation parameters of the preset risk index calculation rule.

[0113] In this step, the environmental dynamic compensation parameter refers to the adjustment rule of real-time environmental factors on the risk value, which is dynamically activated based on real-time monitoring data. The target compensation parameter refers to the actual compensation numerical quantity used for final risk correction.

[0114] In the embodiments of the present application, real-time environmental parameters in the marine biodiversity data are parsed (such as the surface water temperature is taken for coral reefs, and the turbidity value is taken for seagrass beds); according to the ecosystem type and the current risk range, corresponding parameters are selected from the environmental dynamic compensation parameters of the preset risk index calculation rules (such as when the water temperature in the medium-risk area of coral reefs > 28°C, 5 points are added for every 1°C exceeded), and the target compensation parameter is output.

[0115] Step 514: Superimpose the scaled attenuation intensity value and the target compensation parameter to generate a risk index.

[0116] In the embodiments of the present application, arithmetic addition of the scaled attenuation intensity value and the target compensation parameter is performed (such as 36 + (29 - 28) × 5 = 41 in the coral reef area), and when the result exceeds 100, it is truncated to 100 to generate the final risk index.

[0117] The embodiments of the present application break through the limitations of traditional static risk models through a two-layer mechanism of intervalized intensity mapping and dynamic environmental compensation, providing an accurate quantitative basis for marine protection decision-making.

[0118] Figure 2 The following is a schematic structural diagram of a marine ecological evaluation system based on big data analysis provided for the embodiments of the present application, as Figure 2 shown. The system includes: An acquisition module 21, configured to acquire marine biodiversity data and image data of a marine monitoring area, where the marine biodiversity data includes historical species distribution data and niche indicators; A processing module 22, configured to perform streaming processing on the image data by using an edge computing architecture deployed in the marine monitoring area to obtain a biological activity feature map; A partitioning module 23, configured to partition the marine monitoring area according to the spatial resolution of the biological activity feature map and the historical species distribution data to generate a reference species distribution density value for each grid unit, and at the same time calculate the species distribution density value for each grid unit; A determination module 24, configured to calculate the negative deviation between the species distribution density value of each grid unit and the corresponding reference species distribution density value, and when the negative deviation is greater than a preset extinction threshold, determine the corresponding grid unit as a species distribution density attenuation area; An evaluation module 25, configured to generate a marine ecological integrity evaluation result based on all the species distribution density attenuation areas.

[0119] Figure 2 The described marine ecological evaluation system based on big data analysis can execute Figure 1A method for evaluating marine ecology based on big data analysis described in the illustrated embodiment, the implementation principle and technical effects of which will not be elaborated further. For a marine ecology evaluation system based on big data analysis in the above embodiment, the specific manners in which each module and unit perform operations have been described in detail in the embodiments related to the method, and will not be elaborated herein.

[0120] In a possible design, Figure 2 A marine ecology evaluation system based on big data analysis in the illustrated embodiment can be implemented as a computing device, such as Figure 3 as shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0121] The processing component 32 is used for the above Figure 1 A method for evaluating marine ecology based on big data analysis in the illustrated embodiment.

[0122] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method.

[0123] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0124] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0125] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0126] The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, etc.

[0127] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server. The above-mentioned processing components, storage components, etc. can be basic server resources leased or purchased from a cloud computing platform.

[0128] The embodiment of the present application also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above-mentioned Figure 1 marine ecological evaluation method based on big data analysis shown in the embodiment.

[0129] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0130] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0131] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An ocean ecological evaluation method based on big data analysis, characterized in that, Including: Obtaining marine biodiversity data and image data of a marine monitoring area, where the marine biodiversity data includes historical species distribution data and niche indicators; Using an edge computing architecture deployed in the marine monitoring area to perform streaming processing on the image data to obtain a biological activity feature map; According to the spatial resolution of the biological activity feature map and the historical species distribution data, dividing the marine monitoring area into grids, generating a benchmark species distribution density value for each grid cell, and calculating the species distribution density value for each grid cell; Calculating the negative deviation between the species distribution density value of each grid cell and the corresponding benchmark species distribution density value, and when the negative deviation is greater than a preset extinction threshold, determining the corresponding grid cell as a species distribution density attenuation area; Generating a marine ecological integrity evaluation result based on all the species distribution density attenuation areas; 2. The method according to claim 1, wherein Using an edge computing architecture deployed in the marine monitoring area to perform streaming processing on the image data to obtain a biological activity feature map, including: Using the edge computing architecture to separate the multi-spectral bands in the image data to obtain the reflection intensity data of each spectral channel; Strengthening the characteristic bands related to biological activities in the reflection intensity data of all spectral channels to generate enhanced spectral data for each spectral channel; Performing weighted superposition on a preset set of biological weight coefficients and the enhanced spectral data of different spectral channels to generate a preliminary biological activity intensity distribution map; Correcting the discontinuous areas in the preliminary biological activity intensity distribution map to generate a corrected biological activity intensity distribution map; Converting the corrected biological activity intensity distribution map into a biological activity feature map according to the correlation between the historical species distribution data and spectral characteristics; 3. The method according to claim 2, wherein Performing weighted superposition on a preset set of biological weight coefficients and the enhanced spectral data of different spectral channels to generate a preliminary biological activity intensity distribution map, including: Based on the geographic coordinate data of the marine monitoring area, determining the ecosystem type to which the marine monitoring area belongs, where the ecosystem type includes coral reef ecosystems and seagrass bed ecosystems; Invoking a preset set of biological weight coefficients corresponding to the ecosystem type, and performing a multiplication operation on the enhanced spectral data of each pixel point in the enhanced spectral data of different spectral channels and the biological weight coefficients of the corresponding spectral channels in the preset set of biological weight coefficients to obtain the weighted spectral values of different spectral channels for each pixel point; Superposing all the weighted spectral values of the same pixel point to obtain the biological activity intensity value of each pixel point; Converting the biological activity intensity values of all pixel points into distribution image pixel values to generate a preliminary biological activity intensity distribution map according to the distribution image pixel values; 4. The method according to claim 2, wherein Converting the corrected biological activity intensity distribution map into a biological activity feature map according to the correlation between the historical species distribution data and spectral characteristics, including: Grouping the historical species distribution data based on the correlation between the species density of the historical species distribution data and the spectral characteristics, and the ecosystem type to obtain a grouping result; Based on the grouping results, construct a lookup table for the spectral response of species density, where the lookup table for the spectral response of species density contains transformation rules corresponding to different ecosystem types; Based on the transformation rules, convert the biological activity intensity value of each pixel into a species density characteristic value, so as to generate a biological activity intensity distribution map according to all the species density characteristic values.

5. The method according to claim 1, wherein According to the spatial resolution of the biological activity characteristic map and the historical species distribution data, divide the marine monitoring area into grids to generate the reference species distribution density value of each grid unit, and at the same time calculate the species distribution density value of each grid unit, including: According to the spatial resolution of the biological activity characteristic map, use the preset grid division rules to divide the marine monitoring area into multiple grid units; Extract the historical observation points located within the grid unit from the historical species distribution data, and parse the data records of the historical observation points to obtain a set of species density measurement values corresponding to each grid unit; Sort the species density measurement values in each set of species density measurement values to select the target species density measurement value at the median, and use the target species density measurement value corresponding to each grid unit as the corresponding reference species distribution density value; Align the coordinate systems of the grid unit and the biological activity characteristic map to extract a set of biological activity characteristic values corresponding to each grid unit, and use the mean value of the characteristic values in each set of biological activity characteristic values as the species distribution density value of the corresponding grid unit.

6. The method according to claim 1, wherein Based on all the species distribution density attenuation regions, generate an evaluation result of marine ecological integrity, including: Calculate the proportion of the area of all species distribution density attenuation regions in the marine monitoring area, and use the proportion as the attenuation area proportion value; Extract the corresponding negative deviation values from all the species distribution density attenuation regions, calculate the average value of all the negative deviation values, and generate an overall attenuation intensity value; Compare the attenuation area proportion value with the preset health level threshold table to obtain a health level determination result; Based on the preset risk index calculation rule, convert the overall attenuation intensity value into a risk index; Combine the health level determination result and the risk index into an evaluation result of marine ecological integrity.

7. The method according to claim 6, characterized in that Based on the preset risk index calculation rule, convert the overall attenuation intensity value into a risk index, including: According to the preset attenuation intensity risk index mapping table, determine the target conversion numerical interval corresponding to the overall attenuation intensity value; According to the ecosystem type, call the reference proportionality coefficient corresponding to the target conversion numerical interval in the preset risk index calculation rule to scale the overall attenuation intensity value to obtain a scaled attenuation intensity value; Based on the marine biodiversity data, select a target compensation parameter from the environmental dynamic compensation parameters of the preset risk index calculation rule; Superimpose the scaled attenuation intensity value and the target compensation parameter to generate a risk index.

8. An ocean ecological evaluation system based on big data analysis, characterized in that, Including: An acquisition module for acquiring marine biodiversity data and image data of a marine monitoring area, where the marine biodiversity data includes historical species distribution data and niche indicators; A processing module, configured to perform streaming processing on the image data by using an edge computing architecture deployed in the marine monitoring area to obtain a biological activity feature map; A partitioning module, configured to partition the marine monitoring area according to the spatial resolution of the biological activity feature map and the historical species distribution data, generate a reference species distribution density value for each grid cell, and calculate the species distribution density value for each grid cell; A determination module, configured to calculate the negative deviation between the species distribution density value of each grid cell and the corresponding reference species distribution density value, and when the negative deviation is greater than a preset extinction threshold, determine the corresponding grid cell as a species distribution density attenuation area; An evaluation module, configured to generate a marine ecological integrity evaluation result based on all the species distribution density attenuation areas.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a marine ecological evaluation method according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, it implements a marine ecological evaluation method according to any one of claims 1 to 7.

Images