Marine organism monitoring method and system based on remote sensing technology

Abstract

The invention relates to the technical field of remote sensing imaging, in particular to a marine organism monitoring method and system based on a remote sensing technology, and the method comprises the following steps: constructing an environment layer based on a remote sensing image and sensing data, analyzing the relation between a spectrum and an environment, monitoring changes, recognizing the spatial distribution of species, extracting a migration path, and superposing environment factors to analyze ecological response. And outputting a remote sensing indication species ecological response interaction layer. According to the method, the remote sensing image and environmental data are sorted and denoised, multi-dimensional factors are effectively integrated, data precision is enhanced, noise interference is eliminated, spectral change and water temperature gradient deviation are combined, abnormal pixels are accurately screened, sea area change trend is insight in advance, key wave bands and hot salt structures are extracted, target organisms are accurately positioned, and identification efficiency is improved. Analyzing a space offset track, restoring a migration path, constructing a dynamic evolution view angle, superposing water temperature and algae changes, analyzing time sequence synchronization characteristics, revealing ecological response turning points, and strengthening early warning and dynamic monitoring capabilities.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing imaging technology, and in particular to a marine life monitoring method and system based on remote sensing technology. Background Art

[0002] The field of remote sensing imaging encompasses various technologies for capturing and processing images of the Earth and its surface features. The core of this technology lies in capturing images of target areas from a distance using sensors, and analyzing and interpreting the resulting image data to identify, classify, or monitor ground features. Remote sensing imaging can be categorized as satellite, aerial, and ground-based, depending on the sensor platform. It relies on multiple electromagnetic wavebands, such as visible light, infrared, and microwaves, for image data acquisition. This field encompasses multiple technical aspects, including image acquisition, preprocessing, geometric correction, image fusion, feature extraction, and target recognition. It is widely used in resource surveys, environmental monitoring, disaster assessment, agricultural observation, and ocean exploration.

[0003] Among them, the marine biological monitoring method refers to a method for observing and identifying the distribution, species changes and quantity dynamics of biological populations in marine ecosystems based on remote sensing imaging technology. This method mainly targets plankton, algae, fish schools and other objects in specific marine areas, and uses multispectral or hyperspectral images to perform image band combination analysis. The type and density of organisms are determined by image color characteristics and spatial distribution characteristics, and their habitats and migration paths are determined by combining environmental parameters such as temperature and salinity. Its implementation method generally includes collecting specific band images, performing image color clustering analysis, comparing and classifying based on known samples, establishing distribution models based on historical records, and performing time series data comparison to complete the monitoring and identification of target marine organisms.

[0004] Existing remote sensing technologies for marine biological monitoring often rely on single-dimensional judgments based on image color and spatial distribution characteristics, failing to fully consider the multidimensional interaction of complex environmental factors, resulting in inaccurate identification of biological distribution. Band combination analysis, limited by inherent image data noise and environmental fluctuations, is prone to misidentification and omissions. This is particularly true in areas of drastic environmental fluctuations, where the distinction between target organisms and the background becomes blurred, reducing identification accuracy. The lack of integration of dynamic environmental factors such as water temperature, salinity, and current velocity prevents effective reflection of the coupled relationship between biological distribution and environmental conditions, leading to inconsistent and inaccurate estimates of species habitats and migration paths. For monitoring species distribution changes, existing technologies often rely on static point-in-time comparisons, lacking in-depth analysis of continuous spatial displacements and migration trajectories. This makes it difficult to capture the spatiotemporal dynamics of biological populations, reducing the ability to analyze trends in species ecological responses. The failure to establish a framework for simultaneous correlation analysis between environmental trends and biological distribution, coupled with a lack of a holistic understanding of multi-factor coupling, results in a fragmented interaction between environmental fluctuations and biological responses, limiting the ability to timely perceive and respond to ecological risks. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a marine life monitoring method and system based on remote sensing technology.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a marine biological monitoring method based on remote sensing technology, comprising the following steps:

[0007] S1: Based on the spectral reflectance and geographic positioning information of the monitored sea area in remote sensing satellite images, data recorded by ocean sensing devices are collected. Combined with time and regional tags, the images and environmental data are formatted and denoised to generate a remote sensing environmental element composite layer.

[0008] S2: Based on the remote sensing environmental element combination layer, analyze the combined relationship between regional spectral reflectance values and environmental elements, identify the corresponding intervals between the spectral value variation range and the water temperature gradient deviation, filter out pixels outside the variation interval, and eliminate the associated environmental data to generate a sea area change monitoring layer;

[0009] S3: Based on the sea area change monitoring layer, extract spectral band data and water temperature and salinity values, identify key band combinations and pixel sets of thermohaline structure, select pixels that meet the spectral template of known species, and overlay geographic coordinates to mark the corresponding areas to generate a spatial distribution map of remote sensing identified species;

[0010] S4: Based on the continuous position change information between multiple regions in the remote sensing species spatial distribution map, the spatial offset distance of adjacent regions in the time period is identified, the offset direction and path trajectory are extracted, and the regional sections with consistent continuous migration directions are screened to obtain the remote sensing species tracking path segments.

[0011] As a further solution of the present invention, the remote sensing environmental element combination layer includes a spectral reflectance characteristic value set, a water body physical and chemical property grid, and a hydrological dynamic parameter surface; the sea area change monitoring layer includes a spectral anomaly identification result, a water temperature gradient deviation mark, and an environmental data elimination mask; the remote sensing species identification spatial distribution map includes a target species candidate pixel set, a key band response template, and a regional positioning code; the remote sensing species tracking path segment includes a continuous migration trend trajectory, a path direction vector set, and a regional offset node.

[0012] As a further solution of the present invention, the steps for obtaining the remote sensing environmental element combination layer are specifically as follows:

[0013] S111: Based on the spectral reflectance values and geographic positioning information of the monitored sea area in remote sensing satellite images, frame sequence matching is performed in combination with time tags. Pixels whose adjacent frame differences in the reflectance value sequence exceed the critical threshold of spectral change are eliminated to generate a multi-band pixel spectral matrix.

[0014] S112: extracting water temperature, salinity, and ocean current velocity data under the same time and region labels based on the multi-band pixel spectral matrix, filtering out records with null values, and generating a valid regional environmental factor data set;

[0015] S113: Call the valid regional environmental factor data set, identify the standard deviation of water temperature, salinity, and ocean current speed in the pixel group with the same regional label, and perform normalization processing using the formula:

[0016]

[0017] Calculate the normalized spectral difference intensity value, reclassify the pixels in the area, and establish a remote sensing environmental element composite layer;

[0018] Among them, L represents the normalized spectral difference intensity value, R i Represents the sum of multi-band reflectance values of image i in the area, T i is the corresponding water temperature of image i in the area, S i is the salinity of image i in the region, V i is the ocean current velocity of image i in the region, n is the number of pixels in the region, R i+1 Represents the sum of the multi-band reflectance values of image i+1 in the area.

[0019] As a further solution of the present invention, the steps for obtaining the sea area change monitoring layer are specifically as follows:

[0020] S211: Based on the remote sensing environmental element combined layer, extracting spectral reflectance values and water temperature gradient data in the layer, matching and comparing pixel spectral values with water temperature gradient intervals, analyzing corresponding relationships, and obtaining corresponding intervals of spectral values;

[0021] S212: Call the corresponding interval of the spectral value, combine the spectral value of the pixel in the combined layer, determine whether it falls into the specified interval, and identify the difference value of the pixel not in the interval using the formula:

[0022]

[0023] Calculate the pixel deviation that exceeds the variation range, determine the objects to be eliminated based on the deviation, and obtain the list of eliminated pixel numbers;

[0024] Among them, ΔX represents the pixel deviation beyond the variation interval, ρ x is the spectral reflectance value of the pixel x to be detected, μ is the spectral mean value in the same band, θ is the water temperature gradient change rate in the band, φ x is the background interference term of the pixel x to be detected, τ x is the environmental disturbance term of the pixel x to be detected, and m is the total number of pixels to be detected;

[0025] S213: Call the list of eliminated pixel numbers, delete the environmental factor data corresponding to the corresponding numbers in the combined layer, reorganize the spatial distribution of the remaining pixels, and generate a sea area change monitoring layer.

[0026] As a further solution of the present invention, the steps for obtaining the spatial distribution map of species identified by remote sensing are specifically as follows:

[0027] S311: Based on the sea area change monitoring layer, extract the spectral reflectance band, water temperature and salinity values of the layer pixels, combine the differentiated band combination with the corresponding water temperature and salinity data, identify the typical ocean thermohaline structure area, and obtain the thermohaline structure representation combination;

[0028] S312: Call the thermal-salinity structure characterization combination, detect the difference in pixel spectral characteristics and the degree of coordinated fluctuation of water temperature and salinity within the combination area, aggregate and compare the reflection response of the pixels in the key band, and use the formula:

[0029]

[0030] Calculate the coordination factor of the thermohaline spectrum response, calibrate the area of biological activity potential, and obtain the indicator set of sensitive areas for biological monitoring;

[0031] Among them, Q represents the coordination factor of the thermal-salinity spectrum response, β is the main band spectrum value, δ is the reference band spectrum value, ζ is the water temperature variation amplitude, γ is the salinity variation amplitude, κ is the key band difference value, ε is the local interference term, and η is the instrument noise term;

[0032] S313: Calling the biological monitoring sensitive area indicator set, matching the sensitive area pixels with the marine biological spectral template, screening the pixels that meet the feature requirements, and marking the longitude and latitude information of the corresponding area to generate a remote sensing identification species spatial distribution map.

[0033] As a further solution of the present invention, the steps for obtaining the path segments of species tracked by remote sensing are specifically as follows:

[0034] S411: Based on the continuous position change information between multiple regions in the remote sensing species spatial distribution map, identifying the geometric center offset of adjacent regions in the time series, extracting the offset direction vector and displacement length, analyzing the direction angle difference of the consistency of the offset direction between adjacent regions, and obtaining the continuous offset direction data of the species;

[0035] S412: Calling the continuous offset direction data of the species, extracting the regional segments with stable offset directions according to the changing trend of the direction difference, screening the continuous segments with direction changes below the direction angle threshold, and obtaining stable offset segment information;

[0036] S414: Call the stable offset segment information, combine the offset direction angle, inter-regional biological remote sensing intensity and time series continuity index in each segment, and use the formula:

[0037] P = |(d+W)·YO|;

[0038] Calculate the path stability fitness value, identify the trajectory sections with strong continuity and concentrated biological signals, and obtain the path sections of remote sensing tracked species;

[0039] Among them, P represents the path stability suitability value, d represents the angle difference of the direction angle series, W represents the time series continuity index, Y represents the remote sensing biological intensity between regions, and O represents the direction change benchmark angle difference.

[0040] As a further embodiment of the present invention, the method further comprises step S5:

[0041] S5: Invoke the water temperature change trend and salinity level in the species path segment tracked by remote sensing, overlay the phytoplankton concentration layer, compare the time-synchronized change characteristics between environmental factors, identify the coordinates of the segment where the turning point occurs at the same time and the overlapping position of the species distribution, and output the remote sensing indicator species ecological response interaction layer;

[0042] The remote sensing indicator species ecological response interaction layer includes environmental factor linkage change areas, species response intersection points, and ecological anomaly indicator groups.

[0043] As a further solution of the present invention, the steps for obtaining the remote sensing indicator species ecological response interaction layer are specifically as follows:

[0044] S511: Based on the water temperature change trend and salinity level in the remote sensing species tracking path segment, extract the daily change value of the water temperature and the daily average value of the salinity in the segment, analyze the water temperature change slope and salinity fluctuation amplitude, and generate a synchronous environmental factor change characteristic value group;

[0045] S512: Calling the synchronous environmental factor change characteristic value group, extracting the inflection points of continuous time periods based on the remote sensing phytoplankton concentration layer grid concentration sequence, combining the water temperature slope, salinity fluctuation amplitude and algae concentration, screening the segment numbers of simultaneous turning features and summarizing them to obtain a mutation synchronization segment index list;

[0046] S513: According to the mutation synchronization segment index list, the species distribution layer coordinate pixels corresponding to the segment number are matched, overlapping grids with consistent indexes are screened and marked, and the remote sensing indicator species ecological response interaction layer is obtained.

[0047] The marine life monitoring system based on remote sensing technology is used to implement the marine life monitoring method based on remote sensing technology, and the system includes:

[0048] The environmental parameter construction module extracts ocean sensor data, including water temperature, salinity, and current velocity, based on the spectral reflectance and geographic positioning information of the monitored sea area from remote sensing satellite images. It then de-noises and formats the spectral reflectance and environmental data by combining them with time and regional tags to generate a set of ocean environmental remote sensing parameters.

[0049] The marine anomaly detection module, based on the marine environment remote sensing parameter set, filters out pixels outside the range according to spectral reflectance and water temperature data, removes salinity and current velocity data associated with anomalies, integrates the remaining spectral reflectance values and geographic information, marks the abnormal areas, and generates a marine environment anomaly mark map;

[0050] The biological hotspot extraction module extracts spectral reflectance values, band data, water temperature and salinity based on the marine environmental anomaly marker map, identifies pixels that meet the thermal and saline characteristics, matches the marine biological spectral template, superimposes geographic location and time tags, performs spatial positioning, and generates marine biological hotspot distribution results;

[0051] The population path tracking module extracts time tags and geographic information based on the distribution results of marine biological hotspots, identifies the spatial distance between hotspot areas in adjacent time periods, distinguishes the offset direction and continuous trajectory, and integrates path segments with consistent offset directions to obtain remote sensing tracking species path segments;

[0052] The ecological change analysis module is based on the remote sensing tracking species path segment, and performs time series comparison according to the water temperature change trend, salinity level and phytoplankton concentration of the path segment, identifies the synchronous turning point position, locates the overlapping position of the turning point area and the hotspot, and outputs the remote sensing indicator species ecological response interaction layer.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are:

[0054] In the present invention, by formatting and denoising remote sensing images and environmental data, image data is effectively integrated with multi-dimensional environmental factors such as water temperature, salinity, and ocean current speed, thereby improving the availability and accuracy of original data, eliminating the interference of environmental noise on the results, and achieving a more accurate data foundation. In the combined relationship between regional spectral reflectance values and environmental factors, the correspondence between spectral changes and water temperature gradient deviations is deeply explored, abnormal change pixels are accurately screened, and the trend of marine environmental variation is identified in advance, avoiding the limitations of single parameter analysis and improving the ability of multivariate coupling analysis. By extracting spectral band data and combining it with thermal salt structure, a composite judgment standard for key bands and environmental factors is established, and target pixels that meet the characteristics of specific biological populations are accurately located, thereby improving the efficiency of target biological identification and the depth of spatial distribution interpretation. By extracting continuous position change information and calculating spatial offset trajectories, the migration path of species is clearly restored, breaking through the traditional static distribution model, constructing a dynamic spatiotemporal evolution perspective, and enhancing the understanding of the laws of biological group activity. Superimpose the changing trends of water temperature and phytoplankton concentrations, analyze the temporal synchronization characteristics among multiple environmental factors, clarify the dynamic interaction between species and the ecological environment, reveal the key turning points of ecological responses, achieve in-depth analysis of the linkage trends between biological distribution and environmental factors, and further strengthen ecological early warning and dynamic monitoring capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0056] Figure 2 This is a flowchart for obtaining a remote sensing environmental element combination layer in the present invention;

[0057] Figure 3 This is a flow chart for obtaining the sea area change monitoring layer in the present invention;

[0058] Figure 4 This is a flowchart for obtaining a spatial distribution map of species identified by remote sensing in the present invention;

[0059] Figure 5 A flowchart for obtaining a path segment for remote sensing species tracking in the present invention;

[0060] Figure 6 This is a flowchart for obtaining the interactive layer of ecological response of remote sensing indicator species in the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0063] Example 1

[0064] See also Figure 1 The present invention provides a technical solution: a method for monitoring marine organisms based on remote sensing technology, comprising the following steps:

[0065] S1: Based on the spectral reflectance and geographic positioning information of the monitored sea area in remote sensing satellite images, water temperature, salinity, and current velocity data recorded by ocean sensors are collected. Combined with time and regional tags, the images and environmental data are formatted and denoised to generate a remote sensing environmental element composite layer.

[0066] S2: Based on the remote sensing environmental element composite layer, the combined relationship between regional spectral reflectance values and environmental elements is analyzed, the corresponding intervals between the spectral value variation range and the water temperature gradient deviation are identified, pixels outside the variation interval are screened, and the associated environmental data are removed to generate a sea area change monitoring layer;

[0067] S3: Based on the sea area change monitoring layer, spectral band data, water temperature, and salinity values are extracted to identify pixel sets with key band combinations and thermohaline structures. Pixels that meet the spectral templates of known species are screened, and the corresponding areas are marked with geographic coordinates to generate a spatial distribution map of remotely sensed species.

[0068] S4: Based on remote sensing, we can identify the continuous position change information between multiple regions in the species spatial distribution map, identify the spatial offset distance between adjacent regions in the time period, extract the offset direction and path trajectory, and select the regional segments with consistent continuous migration direction to obtain the remote sensing species tracking path segments;

[0069] S5: Use remote sensing to track the changing trends of water temperature and salinity levels in the species path segments, overlay the phytoplankton concentration layer, compare the time-synchronized change characteristics between environmental factors, identify the coordinates of the segments where turning points occur simultaneously and the overlapping positions of species distribution, and output the remote sensing indicator species ecological response interaction layer.

[0070] The remote sensing environmental element combination layer includes a spectral reflectance characteristic value set, a water body physical and chemical property grid, and a hydrological dynamic parameter surface. The sea area change monitoring layer includes spectral anomaly identification results, water temperature gradient deviation marks, and environmental data elimination masks. The remote sensing species identification spatial distribution map includes a target species candidate pixel set, a key band response template, and a regional positioning code. The remote sensing species tracking path segment includes a continuous migration trend trajectory, a path direction vector set, and a regional offset node. The remote sensing indicator species ecological response interaction layer includes an environmental factor linkage change zone, a species response intersection point, and an ecological anomaly indicator group.

[0071] See also Figure 2 , the specific steps for obtaining the remote sensing environmental element combination layer are:

[0072] S111: Based on the spectral reflectance values and geographic positioning information of the monitored sea area in remote sensing satellite images, frame sequence matching is performed in combination with time tags. Pixels whose adjacent frame differences in the reflectance value sequence exceed the critical threshold of spectral change are eliminated to generate a multi-band pixel spectral matrix.

[0073] The basic data is obtained based on the spectral reflectance value of each pixel in different bands and its corresponding longitude and latitude coordinates. This process is the basis of data collection. For example, for a specific ocean area, its reflectance in the infrared band and visible light band is monitored. By obtaining the spectral reflectance value of each pixel, the vegetation distribution in the area or the salinity of the water body can be preliminarily determined. This data is collected at a specific time point through satellite remote sensing technology to ensure the timeliness and accuracy of the data. The image frame sequence is matched by combining the time tag. For the processing of time series data, for example, the difference in reflectance value of a certain area in two consecutive days is compared. If the difference exceeds the set critical threshold of spectral change, it can be judged that environmental changes have occurred in the area, such as the appearance of oil pollution or red tide. The setting of this threshold is based on historical data analysis. By performing statistical analysis on the data at the same time point in the past, a threshold reflecting significant changes is set. The specific data used in this process is the change value of spectral reflectance, which is obtained by directly subtracting the previous and next frames of data. For example, if the reflectance changes from 0.3 to 0.45, the change value is 0.15. If the threshold is set to 0.1, the change is significant and the pixel needs to be further analyzed or eliminated to generate a multi-band pixel spectral matrix.

[0074] S112: Extracting water temperature, salinity, and ocean current velocity data under the same time and region labels based on the multi-band pixel spectral matrix, filtering out records with null values, and generating a valid regional environmental factor data set;

[0075] Based on the multi-band pixel spectral matrix extracted from remote sensing images, we further call on marine environmental sensor data with the same time tag and geographic tag as the image, such as water temperature, salinity, and current speed. The purpose of this step is to combine image data with actual environmental parameters to enhance the explanatory power of the data. For example, in a marine protected area, by collecting water temperature data synchronized with the image, we can further analyze the impact of water temperature changes on coral reefs based on spectral analysis. Salinity data can help understand changes in the salinity of water bodies. For salinity measurements, they are obtained through conductivity sensors and matched with image data. Through the data screening stage, any records with null values in the environmental sensor data are eliminated to ensure the integrity and accuracy of the data for subsequent analysis and generate a valid regional environmental factor data set.

[0076] S113: Call the valid regional environmental factor data set, identify the standard deviation of water temperature, salinity, and ocean current speed in the pixel group with the same regional label, and perform normalization processing using the formula:

[0077]

[0078] Calculate the normalized spectral difference intensity value, reclassify the pixels in the area, and establish a remote sensing environmental element composite layer;

[0079] Among them, L represents the normalized spectral difference intensity value, R i Represents the sum of multi-band reflectance values of image i in the area, T i is the corresponding water temperature of image i in the area, S i is the salinity of image i in the region, V i is the ocean current velocity of image i in the region, n is the number of pixels in the region, R i+1 It represents the sum of multi-band reflectance values of image i+1 in the area;

[0080] After calling the valid regional environmental factor data set, a more in-depth data analysis is performed. The core of this stage is to calculate the standard deviation of water temperature, salinity, and ocean current speed, and perform normalization to unify the measurement. For example, when analyzing a specific sea area, the standard deviation of water temperature in certain areas is abnormal, which indicates that there are significant temperature fluctuations or uneven temperature layer structure in the area. Such data is collected by ocean research vessels or buoys, recording hourly water temperature data, and then using statistical software to calculate the standard deviation of water temperature for all monitoring points in the area. The same method is applied to the data collection and processing of salinity and ocean current speed;

[0081] Normalization is performed to make the data comparable. The specific meaning and acquisition process of each parameter are as follows:

[0082] R i It represents the sum of the multi-band spectral reflectance values of pixel i. This value is obtained by accumulating the spectral data directly recorded by satellite sensors. For example, the reflectance values of a pixel in the red, green, and blue bands may be 0.3, 0.4, and 0.35 respectively. i Then it is 0.3+0.4+0.35=1.05;

[0083] T i represents the water temperature value recorded at the pixel i. Assume that the water temperature obtained through data collection is 20℃;

[0084] S i It indicates the salinity value corresponding to the location, which is set to 35‰ and is collected by the salinity sensor at the same time and location;

[0085] V i Indicates the current velocity value at this location. If the data obtained by the water velocity meter is 0.5m / s;

[0086] n represents the total number of pixels that belong to the same region label as pixel i. For example, there are 25 pixels in a 5x5 pixel region.

[0087] R i+1It represents the sum of the spectral reflectance values of the other pixel i+1 in the same area. For a simple example, it is set to 1.1;

[0088] Now let's bring in specific values for calculation. First, let's calculate the first part of the formula:

[0089] Then calculate the second part of the formula:

[0090] The final normalized spectral difference intensity value L is L = 5.21 + 0.05 = 5.26;

[0091] This calculation process demonstrates how to analyze environmental parameters through actual data and further derive the specific environmental conditions of each area. Such data can help researchers assess and monitor the health of marine ecosystems and take necessary environmental protection measures based on this data.

[0092] See also Figure 3 , the specific steps for obtaining the sea area change monitoring layer are as follows:

[0093] S211: Based on the remote sensing environmental element combination layer, extract the spectral reflectance value and water temperature gradient data in the layer, match and compare the pixel spectral value and the water temperature gradient interval, analyze the corresponding relationship, and obtain the spectral value corresponding interval;

[0094] When monitoring is performed based on a combination of remote sensing environmental element layers, the matching and comparison of spectral reflectance values and water temperature gradients is a key process. This process involves comparing the spectral data recorded in the layer with the ambient water temperature data to identify outliers or deviations in the spectral data. For example, suppose in a typical ocean monitoring project, remote sensing data shows that the spectral reflectance value of an area suddenly increases, while the water temperature record of the area shows a normal range. This difference suggests that some biological activity or environmental change has occurred in the area, such as the occurrence of red tides. Through comprehensive analysis of the data, it is possible to match the range of different spectral values and water temperature gradient deviations, partition the data for further research, and finally generate a corresponding interval of spectral values. This is a specific numerical interval extracted from the basic data of spectral reflectance and water temperature gradient. The setting of this interval is based on statistical analysis of similar data in the past and on-site monitoring results.

[0095] S212: Call the corresponding interval of the spectral value, combine the spectral value of the pixel in the combined layer, determine whether it falls into the specified interval, and identify the difference value of the pixels that are not in the interval using the formula:

[0096]

[0097] Calculate the pixel deviation that exceeds the variation range, determine the objects to be eliminated based on the deviation, and obtain the list of eliminated pixel numbers;

[0098] Among them, ΔX represents the pixel deviation beyond the variation interval, ρ x is the spectral reflectance value of the pixel x to be detected, μ is the spectral mean value in the same band, θ is the water temperature gradient change rate in the band, φ x is the background interference term of the pixel x to be detected, τ x is the environmental disturbance term of the pixel x to be detected, and m is the total number of pixels to be detected;

[0099] Combined with the pixel spectral reflectance values in the actual remote sensing layer, it is necessary to determine whether the spectral value of each pixel is within the interval. This process is achieved by traversing each pixel in the layer and comparing its spectral reflectance value with the upper and lower limits of the interval. For example, if the interval is set to [0.18, 0.34], if the reflectance value of a pixel is 0.41, it is determined to be outside the interval and needs to be included in further analysis. For such pixels, the difference amount needs to be calculated in combination with the changing characteristics of seawater temperature and the background of environmental disturbance to confirm whether the degree of deviation meets the elimination standard;

[0100] The parameter acquisition process and value setting are as follows:

[0101] Pixel reflectance value ρ x Read directly from remote sensing images, the value is 0.41;

[0102] The band reflectance mean μ is calculated by all pixels in the area, and the average of the band in the area is set to 0.26;

[0103] The water temperature gradient change rate θ is the temperature change amplitude within a unit space. Assuming that the average temperature change in the current area is 3.8℃, the corresponding area span is 20km, and the calculation is

[0104] Background interference term φ x and the environmental disturbance term τ x Both were calculated by superimposing the variance of historical remote sensing data, with values of 0.0065 and 0.0082, respectively, and a total disturbance value of 0.0147;

[0105] The number of pixels in the region is set to m = 6, then:

[0106] Substitute the formula into the calculation as follows:

[0107] Calculate the difference part: ρ x -μ=0.41-0.26=0.15;

[0108] Multiply by the temperature gradient: (ρx -μ)×θ=0.15×0.19=0.0285;

[0109] Find the square root of the denominator:

[0110] Complete the integer operation:

[0111] This value is the pixel deviation outside the variation range, indicating that the pixel has a deviation of 0.096 units relative to the spectral and water temperature trends. According to the set rejection baseline value range [0.08, 0.1], this pixel is at the edge of the critical deviation range and needs to be added to the rejection list according to the set logic.

[0112] Among them, Δ x is the deviation (dimensionless) calculated for pixel x in the current band, ρ x is the spectral reflectance value of the corresponding band of pixel x (unit: dimensionless reflectance), μ is the average reflectance value of the band (unit: dimensionless reflectance), θ is the water temperature gradient change rate of the area (unit: ℃ / km), φ x is the background interference value of pixel x (unit: dimensionless), τ x is the environmental disturbance value of pixel x (unit: dimensionless), and m is the total number of pixels in the calculation area (unit: piece); all the above dimensions have been unified in the formula structure, and are consistent in the form of dimensionless or unit ratio, making the formula practically operable;

[0113] By coupling the spectral reflectance deviation with the water temperature gradient and taking into account the weighted influence of the disturbance term in the region, the sensitivity to changes in environmental factors can be comprehensively reflected to avoid misjudgment of local changes. The current pixel deviation value is at the critical position of the preset elimination interval, and it is judged that it has the basis for elimination, so the number will be included in the elimination pixel number list.

[0114] S213: calling the list of eliminated pixel numbers, deleting the environmental factor data corresponding to the corresponding numbers in the combined layer, reorganizing the spatial distribution of the remaining pixels, and generating a sea area change monitoring layer;

[0115] Using the list of deleted pixel numbers, operators can delete the environmental factor data corresponding to the numbers in the layer and reorganize the spatial distribution of the remaining pixels, which is directly related to the accuracy and practicality of the monitoring results. For example, when monitoring red tides, after removing pixels that reflect less impact from red tides or low data quality, the remaining pixel data will be recombined to form a new layer. This layer more accurately depicts the distribution and intensity of red tides, providing decision makers with real-time and reliable monitoring data. In this way, a sea area change monitoring layer is generated, which not only reflects the current state of the ocean, but also provides a basis for predicting future changes. The results are based on continuous observation of marine ecological changes and real-time data analysis, ensuring the accuracy and practicality of the monitoring layer data.

[0116] See also Figure 4 The specific steps for obtaining the spatial distribution map of species identified by remote sensing are as follows:

[0117] S311: Based on the sea area change monitoring layer, extract the spectral reflectance bands, water temperature, and salinity values of the layer pixels. Combine the differentiated band combinations with the corresponding water temperature and salinity data to identify typical ocean thermohaline structure areas and obtain thermohaline structure characterization combinations.

[0118] The algorithm automatically selects pixels containing key thermohaline structure features. For example, if the salinity and water temperature in certain areas of the monitored sea area change suddenly, and the changes indicate the presence of specific marine biological activities, such as fish migration in a specific season, the spectral values of different bands will be compared with the corresponding thermohaline data to identify spectral band combinations that can represent the thermohaline structure. This process involves arithmetic calculations on multiple band values, such as comparing the reflectance values of the main band with those of the auxiliary band, and considering the combined effects of water temperature and salinity to obtain spectral combinations that can significantly represent the thermohaline interaction. This calculation not only relies on the raw data collected by the sensor, but also involves data preprocessing, such as correcting the spectral data to eliminate instrument bias and environmental noise. Through detailed steps, the thermohaline structure characterization combination is finally obtained, providing an important data foundation for subsequent biological distribution analysis.

[0119] S312: Call the thermal-salinity structure characterization combination to detect the difference in pixel spectral characteristics and the degree of coordinated fluctuation of water temperature and salinity within the combination area, aggregate and compare the reflection response of the pixels in the key bands, and use the formula:

[0120]

[0121] Calculate the coordination factor of the thermohaline spectrum response, calibrate the area of biological activity potential, and obtain the indicator set of sensitive areas for biological monitoring;

[0122] Among them, Q represents the coordination factor of the thermal-salinity spectrum response, β is the main band spectrum value, δ is the reference band spectrum value, ζ is the water temperature variation amplitude, γ is the salinity variation amplitude, κ is the key band difference value, ε is the local interference term, and η is the instrument noise term;

[0123] After confirming the combined area that can characterize the thermohaline structure, it is necessary to conduct a joint collaborative evaluation of the spectral response of the pixels in this area with water temperature and salinity to identify the existing hotspots of biological activity. The specific method is to extract the spectral reflectance values of each pixel in the key main band and auxiliary band, and perform numerical calculations based on the variation of water temperature and salinity. Taking the actual marine monitoring scene as an example, a coral reef distribution area near the southern coast was selected for analysis. The main band reflectance value of a certain pixel was β = 0.42, the auxiliary band reflectance value was δ = 0.31, the water temperature gradient variation range was ζ = 2.8 ° C, the salinity variation range was γ = 1.6%, the offset difference between the neighboring key bands was κ = 0.07, the local area background disturbance was ε = 0.012, and the remote sensing instrument noise term was η = 0.005;

[0124] The calculation is as follows: spectral difference multiplied by the change in water temperature:

[0125] (β-δ)·ζ=(0.42-0.31)·2.8=0.11·2.8=0.308;

[0126] Adding salinity change and band difference:

[0127] (β-δ)·ζ+(γ+κ)=0.308+1.6+0.07=1.978;

[0128] Calculate the square root term in the denominator:

[0129]

[0130] The final formula is calculated as:

[0131] This value is the thermohaline spectral response coordination factor. The higher the value of this factor, the tighter the coupling between spectral differences, water temperature changes, and salinity fluctuations in the area where the pixel is located. According to the analysis of biological response data, the Q index of the coral reef growth sensitive area is between 10 and 20. The current result is 15.215, which is within the response sensitive range. Therefore, it can be determined that there is a significant trend in marine biological activity in this area, and it is then included in the biological monitoring sensitive area indicator set.

[0132] Where Q represents the thermal-salinity spectral response coordination factor (dimensionless), β is the pixel main band reflectance (reflectance, unit is dimensionless), δ is the auxiliary band reflectance (reflectance, unit is dimensionless), ζ is the water temperature change value (unit is Celsius), γ is the salinity change amplitude (unit is ‰), κ is the key band difference (unit is reflectance), ε is the local perturbation error (dimensionless), and η is the instrument noise amplitude (dimensionless). The dimensions of each parameter have been processed into the same system in the formula to ensure the consistency of units.

[0133] By introducing the joint structure of spectral difference, water temperature change and salinity amplitude, and incorporating the disturbance term into the denominator to adjust the fluctuation coefficient, the synergistic characteristics can be dynamically characterized without the need for weight adjustment. The current pixel has high synergistic responsiveness, which is consistent with the characteristics of highly active marine biological areas, and can therefore be directly attributed to the biological monitoring sensitive area indicator set.

[0134] S313: Call the biological monitoring sensitive area indicator set, match the sensitive area pixels with the marine biological spectral template, select the pixels that meet the characteristic requirements, and mark the longitude and latitude information of the corresponding area to generate the remote sensing identification species spatial distribution map;

[0135] Pixels that conform to known spectral templates are screened, and the spatial distribution of organisms is marked on the remote sensing layer. For example, when monitoring seagrass in a certain sea area, researchers can identify the seagrass growth area through remote sensing data. The process involves matching the spectral reflectance characteristics in the remote sensing data with the spectral template of the seagrass, and then spatially locating and marking the successfully matched areas. This can not only help scientific researchers quickly identify the distribution range of seagrass, but also provide accurate geographic information for subsequent ecological protection and resource management, and generate remote sensing identification species spatial distribution maps, which will intuitively display the specific distribution of various marine organisms in the study area, becoming an important tool in marine ecological monitoring and environmental management.

[0136] See also Figure 5 The specific steps for obtaining the path segments of species tracked by remote sensing are as follows:

[0137] S411: Identify the continuous position change information between multiple regions in the species spatial distribution map based on remote sensing, identify the geometric center offset of adjacent regions in the time series, extract the offset direction vector and displacement length, analyze the direction angle difference of the consistency of the offset direction between adjacent regions, and obtain the continuous offset direction data of species;

[0138] First, the center coordinates of each area are calculated, and the coordinate changes between two consecutive time points are obtained by the difference method. This change represents the movement trend and speed of the species. For example, suppose a marine species moves from coordinate A (100, 200) to coordinate B (150, 250). The calculated coordinate offset is (50, 50). This offset is further vector analyzed to determine the direction and speed of the species' movement. Based on the extracted offset direction vector and displacement length value, the data is clustered and analyzed to screen out species groups with the same or similar movement directions. This is achieved by comparing the angular difference of the displacement direction vectors of different groups. The angle difference calculation uses the cosine theorem to quantitatively describe the directional consistency. If the angular difference between the two vectors is less than 10 degrees, they are considered to have the same offset direction. Through this process, the migration path of species can be effectively monitored and predicted, which provides important data for studying the distribution changes of species in marine ecology and their environmental adaptability, and ultimately obtains continuous species offset direction data.

[0139] S412: Calling the species continuous migration direction data, extracting the regional segments with stable migration direction according to the change trend of the direction difference, screening the continuous segments with direction changes below the direction angle threshold, and obtaining the stable migration segment information;

[0140] After obtaining the continuous migration direction data of the species, the data is further analyzed to determine whether the migration of the species has a certain degree of continuity and stability. The process first involves the calculation of the direction difference, which is accomplished by comparing the direction vector of each segment in the continuous migration data and calculating the change in its direction angle. For example: if the migration directions of three consecutive segments are 10 degrees north-east, 12 degrees north-east, and 11 degrees north-east, and the change in direction angle is stable within 2 degrees, it can be considered that the migration direction of the species is highly consistent. The segments with consistent directions are marked and extracted to construct the migration path map of the species. The marked segments are very critical in marine biological monitoring because they help researchers identify the key paths and stopover points of species migration. By identifying the key paths, scientific researchers can better understand the response of species to changes in the marine environment and their ecological niches, and ultimately obtain information on stable migration segments.

[0141] S414: Call the stable offset segment information, combine the offset direction angle in each segment, the inter-regional biological remote sensing intensity and the time series continuity index, and use the formula:

[0142] P = |(d+W)·YO|;

[0143] Calculate the path stability fitness value, identify the trajectory sections with strong continuity and concentrated biological signals, and obtain the path sections of remote sensing tracked species;

[0144] Among them, P represents the path stability suitability value, d represents the angle difference of the direction angle series, W represents the time series continuity index, Y represents the remote sensing biological intensity between regions, and O represents the direction change benchmark angle difference;

[0145] After confirming the stable offset segment information, a comprehensive assessment of the species' migration path segments is conducted. The stable suitability value of the path is calculated through a combination of multiple factors, and its actual tracking significance in remote sensing monitoring is judged based on this. First, the azimuth angle difference d, remote sensing biological intensity Y, time series continuity index W, and the set azimuth offset reference angle difference O in each stable path segment are extracted. Each parameter must have dimension consistency. The azimuth angle difference d is in degrees, with a value range of 0°–180°. The remote sensing biological intensity Y is the normalized reflectance value (between 0 and 1). The time series continuity index W is a dimensionless ratio (ranging from 0 to 1). The azimuth offset reference angle difference O is in degrees. The setting value must be set based on empirical thresholds or species historical trajectory assessments. For example, for a certain type of mid-water fish, if the migration path direction does not deviate from the reference angle by ±7°, then O is set to 7.

[0146] The process of obtaining each parameter is as follows:

[0147] Direction angle difference d: obtained by calculating the heading angle change between any two points in each stable path segment. The heading angle is calculated by constructing a vector from the species distribution points in the remote sensing image. For example, the angle between two points A (120, 130) and B (150, 160) can be obtained by It was calculated that the maximum deviation within a segment obtained by sequence alignment is d, which is assumed to be 5°;

[0148] Remote sensing biological intensity Y: extract the reflectance or fluorescence signal from the high-resolution remote sensing image within the coverage area of the route segment, and obtain the value after mean normalization, assuming it is 0.85;

[0149] Time series continuity index W: It indicates the proportion of regional locations without breaks in consecutive time points. If the monitoring period is 10 days and the path segment covers 9 consecutive days, then W = 9 / 10 = 0.9;

[0150] Directional deviation base angle difference O: Based on historical monitoring data, the deviation angles of this species are mostly concentrated within ±7°. To enhance the comparability of the formula results, O=7 is set;

[0151] Substituting the above parameters into the formula: P = |(d+W)·YO| = |(5+0.9)·0.85-7| = 1.985;

[0152] Wherein, d is the direction angle difference, unit is degree, W is the temporal continuity index, which is a dimensionless ratio, Y is the normalized remote sensing biomass intensity, dimensionless reflectivity value, O is the reference angle difference, unit is degree, and P is the path stability suitability value, unit is degree;

[0153] The results indicate that the current path segment is close to the set benchmark path suitability interval (0–2) in terms of directional stability and biological signal continuity, and therefore has good tracking characteristics and can be used as a remote sensing species path segment for subsequent migration analysis, resource deployment, or ecological regulation research.

[0154] Through comprehensive quantitative analysis of three factors: remote sensing intensity, directional consistency, and time coverage, a new path segment stability evaluation standard is constructed. This standard not only considers the position change trend, but also combines the intensity and time characteristics of biological reactions, thereby achieving the simultaneous integration of path tracking reliability evaluation and dynamic screening.

[0155] See also Figure 6 The specific steps for obtaining the remote sensing indicator species ecological response interaction layer are as follows:

[0156] S511: Tracking water temperature trends and salinity levels along species pathways using remote sensing, extracting daily water temperature variations and daily average salinity values, analyzing water temperature variation slopes and salinity fluctuations, and generating synchronized environmental factor variation feature sets.

[0157] First, the daily water temperature change and salinity average daily value data for each path section are obtained. For example, in a specific area of the North Sea, satellite remote sensing data shows that the water temperature rose from 15 degrees to 19 degrees and the salinity increased from 3.5‰ to 3.7‰. Then, the data is statistically analyzed to calculate the slope of the water temperature change and the average fluctuation amplitude of the salinity. For example, the ratio of the water temperature change value to the time interval is used to obtain the slope of the change, and the stability of the salinity fluctuation is described by the standard deviation. After the data is synchronously numbered, it can be directly mapped to a specific path section. For example, the data of section number 001 represents a small environment with a sudden increase in water temperature. This information is critical for predicting the short-term migration of fish schools, and provides researchers with reference values for specific operations. For example, for every 1 degree increase in water temperature, the probability of fish migration increases by 10%, and finally a set of characteristic values of synchronous environmental factor changes is generated.

[0158] S512: Calling the synchronous environmental factor change characteristic value group, based on the remote sensing phytoplankton concentration layer grid concentration sequence, combined with the water temperature slope, salinity fluctuation amplitude and algae concentration, extracting the inflection points of the continuous time period, screening the segment numbers of the simultaneous turning features and summarizing them, and obtaining the mutation synchronization segment index list;

[0159] Calling the synchronous environmental factor change characteristic value group and the remote sensing phytoplankton concentration layer grid concentration sequence of the corresponding period, for practical applications in environmental protection and marine biology, such as monitoring the occurrence of red tides, the sea area where red tides occur can be determined through comprehensive data analysis of water temperature, salinity and algae concentration, so as to implement preventive measures. The specific operation includes daily comparative analysis of water temperature slope, salinity fluctuation amplitude and algae concentration. For example, on a certain day, the water temperature sudden increase slope is observed to be 0.2℃ / day, the salinity fluctuation amplitude is 0.1‰ / day, and on the same day, When the algae concentration suddenly increases to 2000 cells / mL, the data is marked and automatically compared and identified, and the segments where all three indicators simultaneously change at the same time point or in a continuous period are identified. The data is processed through examples, such as determining whether the turning point meets the conditions for the occurrence of red tide, that is, the algae concentration threshold is set at 1500 cells / mL, the water temperature rise rate threshold is 0.15℃ / day, and the salinity change threshold is 0.05‰ / day. Based on the comprehensive judgment of the parameters, the synchronous mutation segments are further extracted and listed, and the mutation synchronization segment index list is obtained.

[0160] S513: According to the mutation synchronization segment index list, match the species distribution layer coordinate pixels corresponding to the segment number, select and mark the overlapping grids with consistent indexes, and obtain the remote sensing indicator species ecological response interaction layer;

[0161] Through actual marine protected area management, such as the specific monitoring of a marine protected area, staff can use data to quickly determine the sea area blocks that need to be monitored. The specific process includes matching the coordinate pixels of the corresponding segment number in the species distribution layer. For example, segment number 002 is marked in the red tide warning index, and the GPS coordinates of the corresponding area are 120 degrees east longitude and 30 degrees north latitude. The algae concentration, water temperature and salinity data of the area are all shown to be above the red tide warning value. The coordinates of the segment and the species distribution layer are analyzed for data coverage, and the overlapping coordinate grid points are marked and classified. For example, the data of the point are further integrated and analyzed to calculate the probability of red tide occurrence at the point, providing specific data support for the marine protected area. For example, if the probability of red tide occurrence exceeds 75%, emergency response measures must be initiated immediately. Such detailed data processing and result output provide a decision-making basis for protected area management, and ultimately obtain an interactive layer of remote sensing indicator species ecological responses.

[0162] The marine biological monitoring system based on remote sensing technology is used to implement the above-mentioned marine biological monitoring method based on remote sensing technology, and the system includes:

[0163] The environmental parameter construction module extracts ocean sensor data, including water temperature, salinity, and current velocity, based on the spectral reflectance and geographic positioning information of the monitored sea area from remote sensing satellite images. It then de-noises and formats the spectral reflectance and environmental data by combining them with time and regional tags to generate a set of ocean environmental remote sensing parameters.

[0164] The marine anomaly detection module is based on a set of marine environmental remote sensing parameters. It filters pixels that exceed the range based on spectral reflectance and water temperature data, removes salinity and current velocity data associated with anomalies, integrates the remaining spectral reflectance values with geographic information, marks the anomaly areas, and generates a marine environmental anomaly map.

[0165] The biological hotspot extraction module extracts spectral reflectance values, band data, water temperature and salinity based on the marine environmental anomaly marker map, identifies pixels that meet the thermal and salinity characteristics, matches the marine biological spectral template, superimposes geographic location and time tags, performs spatial positioning, and generates marine biological hotspot distribution results;

[0166] The population path tracking module extracts time tags and geographic information based on the distribution of marine biological hotspots, identifies the spatial distance between hotspot areas in adjacent time periods, distinguishes offset directions and continuous trajectories, and integrates path segments with consistent offset directions to obtain remote sensing species tracking path segments;

[0167] The ecological change analysis module tracks species path segments based on remote sensing. It compares time series according to the water temperature change trends, salinity levels, and phytoplankton concentrations of the path segments, identifies synchronous turning points, locates the overlapping positions of turning points and hotspots, and outputs remote sensing indicator species ecological response interaction layers.

[0168] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for monitoring marine life based on remote sensing technology, characterized in that: The following steps are involved: S1: Based on the spectral reflectance and geographic positioning information of the monitored sea area in remote sensing satellite images, data recorded by ocean sensing devices are collected. Combined with time and regional tags, the images and environmental data are formatted and denoised to generate a remote sensing environmental element composite layer. S2: Based on the remote sensing environmental element combination layer, analyze the combined relationship between regional spectral reflectance values and environmental elements, identify the corresponding intervals between the spectral value variation range and the water temperature gradient deviation, filter out pixels outside the variation interval, and eliminate the associated environmental data to generate a sea area change monitoring layer; S3: Based on the sea area change monitoring layer, extract spectral band data and water temperature and salinity values, identify key band combinations and pixel sets of thermohaline structure, select pixels that meet the spectral template of known species, and overlay geographic coordinates to mark the corresponding areas to generate a spatial distribution map of remote sensing identified species; S4: Based on the continuous position change information between multiple regions in the remote sensing species spatial distribution map, the spatial offset distance of adjacent regions in the time period is identified, the offset direction and path trajectory are extracted, and the regional sections with consistent continuous migration directions are screened to obtain the remote sensing species tracking path segments.

2. The marine biological monitoring method based on remote sensing technology according to claim 1, characterized in that: The remote sensing environmental element combination layer includes a spectral reflectance characteristic value set, a water body physical and chemical property grid, and a hydrological dynamic parameter surface. The sea area change monitoring layer includes a spectral anomaly identification result, a water temperature gradient deviation mark, and an environmental data elimination mask. The remote sensing species identification spatial distribution map includes a target species candidate pixel set, a key band response template, and a regional positioning code. The remote sensing species tracking path segment includes a continuous migration trend trajectory, a path direction vector set, and a regional offset node.

3. The marine biological monitoring method based on remote sensing technology according to claim 1, characterized in that: The steps for obtaining the remote sensing environmental element combination layer are specifically as follows: S111: Based on the spectral reflectance values and geographic positioning information of the monitored sea area in remote sensing satellite images, frame sequence matching is performed in combination with time tags. Pixels whose adjacent frame differences in the reflectance value sequence exceed the critical threshold of spectral change are eliminated to generate a multi-band pixel spectral matrix. S112: extracting water temperature, salinity, and ocean current velocity data under the same time and region labels based on the multi-band pixel spectral matrix, filtering out records with null values, and generating a valid regional environmental factor data set; S113: Call the valid regional environmental factor data set, identify the standard deviation of water temperature, salinity, and ocean current speed in the pixel group with the same regional label, and perform normalization processing using the formula: Calculate the normalized spectral difference intensity value, reclassify the pixels in the area, and establish a remote sensing environmental element composite layer; Among them, L represents the normalized spectral difference intensity value, R i Represents the sum of multi-band reflectance values of image i in the area, T i is the corresponding water temperature of image i in the area, S i is the salinity of image i in the region, V i is the ocean current velocity of image i in the region, n is the number of pixels in the region, R i+1 Represents the sum of the multi-band reflectance values of image i+1 in the area.

4. The marine biological monitoring method based on remote sensing technology according to claim 3, characterized in that: The steps for obtaining the sea area change monitoring layer are as follows: S211: Based on the remote sensing environmental element combined layer, extracting spectral reflectance values and water temperature gradient data in the layer, matching and comparing pixel spectral values with water temperature gradient intervals, analyzing corresponding relationships, and obtaining corresponding intervals of spectral values; S212: Call the corresponding interval of the spectral value, combine the spectral value of the pixel in the combined layer, determine whether it falls into the specified interval, and identify the difference value of the pixel not in the interval using the formula: Calculate the pixel deviation that exceeds the variation range, determine the objects to be eliminated based on the deviation, and obtain the list of eliminated pixel numbers; Among them, ΔX represents the pixel deviation beyond the variation interval, ρ x is the spectral reflectance value of the pixel x to be detected, μ is the spectral mean value in the same band, θ is the water temperature gradient change rate in the band, φ x is the background interference term of the pixel x to be detected, τ x is the environmental disturbance term of the pixel x to be detected, and m is the total number of pixels to be detected; S213: Call the list of eliminated pixel numbers, delete the environmental factor data corresponding to the corresponding numbers in the combined layer, reorganize the spatial distribution of the remaining pixels, and generate a sea area change monitoring layer.

5. The marine biological monitoring method based on remote sensing technology according to claim 4, characterized in that: The steps for obtaining the spatial distribution map of species identified by remote sensing are specifically as follows: S311: Based on the sea area change monitoring layer, extract the spectral reflectance band, water temperature and salinity values of the layer pixels, combine the differentiated band combination with the corresponding water temperature and salinity data, identify the typical ocean thermohaline structure area, and obtain the thermohaline structure representation combination; S312: Call the thermal-salinity structure characterization combination, detect the difference in pixel spectral characteristics and the degree of coordinated fluctuation of water temperature and salinity within the combination area, aggregate and compare the reflection response of the pixels in the key band, and use the formula: Calculate the coordination factor of the thermohaline spectrum response, calibrate the area of biological activity potential, and obtain the indicator set of sensitive areas for biological monitoring; Among them, Q represents the coordination factor of the thermal-salinity spectrum response, β is the main band spectrum value, δ is the reference band spectrum value, ζ is the water temperature variation amplitude, γ is the salinity variation amplitude, κ is the key band difference value, ε is the local interference term, and η is the instrument noise term; S313: Calling the biological monitoring sensitive area indicator set, matching the sensitive area pixels with the marine biological spectral template, screening the pixels that meet the feature requirements, and marking the longitude and latitude information of the corresponding area to generate a remote sensing identification species spatial distribution map.

6. The marine biological monitoring method based on remote sensing technology according to claim 5, characterized in that: The steps for obtaining the path segments of species tracked by remote sensing are specifically as follows: S411: Based on the continuous position change information between multiple regions in the remote sensing species spatial distribution map, identifying the geometric center offset of adjacent regions in the time series, extracting the offset direction vector and displacement length, analyzing the direction angle difference of the consistency of the offset direction between adjacent regions, and obtaining the continuous offset direction data of the species; S412: Calling the continuous offset direction data of the species, extracting the regional segments with stable offset directions according to the changing trend of the direction difference, screening the continuous segments with direction changes below the direction angle threshold, and obtaining stable offset segment information; S414: Call the stable offset segment information, combine the offset direction angle, inter-regional biological remote sensing intensity and time series continuity index in each segment, and use the formula: P = |(d+W)·YO|; Calculate the path stability fitness value, identify the trajectory sections with strong continuity and concentrated biological signals, and obtain the path sections of remote sensing tracked species; Among them, P represents the path stability suitability value, d represents the angle difference of the direction angle series, W represents the time series continuity index, Y represents the remote sensing biological intensity between regions, and O represents the direction change benchmark angle difference.

7. The marine biological monitoring method based on remote sensing technology according to claim 1, characterized in that: The method further comprises step S5: S5: Invoke the remote sensing tracking of water temperature change trends and salinity levels in the species path segment, overlay a phytoplankton concentration layer, compare the time-synchronized change characteristics between environmental factors, identify the coordinates of the segments where simultaneous turning points occur and the overlapping positions of species distribution, and output a remote sensing indicator species ecological response interaction layer; The remote sensing indicator species ecological response interaction layer includes environmental factor linkage change areas, species response intersection points, and ecological anomaly indicator groups.

8. The marine biological monitoring method based on remote sensing technology according to claim 7, characterized in that: The steps for obtaining the remote sensing indicator species ecological response interaction layer are as follows: S511: Based on the water temperature change trend and salinity level in the remote sensing species tracking path segment, extract the daily change value of the water temperature and the daily average value of the salinity in the segment, analyze the water temperature change slope and salinity fluctuation amplitude, and generate a synchronous environmental factor change characteristic value group; S512: Calling the synchronous environmental factor change characteristic value group, extracting the inflection points of continuous time periods based on the remote sensing phytoplankton concentration layer grid concentration sequence, combining the water temperature slope, salinity fluctuation amplitude and algae concentration, screening the segment numbers of simultaneous turning features and summarizing them to obtain a mutation synchronization segment index list; S513: According to the mutation synchronization segment index list, the species distribution layer coordinate pixels corresponding to the segment number are matched, overlapping grids with consistent indexes are screened and marked, and the remote sensing indicator species ecological response interaction layer is obtained.

9. Marine biological monitoring system based on remote sensing technology, characterized by: The system is used to implement the marine biological monitoring method based on remote sensing technology according to any one of claims 1 to 8, and the system comprises: The environmental parameter construction module extracts ocean sensor data, including water temperature, salinity, and current velocity, based on the spectral reflectance and geographic positioning information of the monitored sea area from remote sensing satellite images. It then de-noises and formats the spectral reflectance and environmental data by combining them with time and regional tags to generate a set of ocean environmental remote sensing parameters. The marine anomaly detection module, based on the marine environment remote sensing parameter set, filters out pixels outside the range according to spectral reflectance and water temperature data, removes salinity and current velocity data associated with anomalies, integrates the remaining spectral reflectance values and geographic information, marks the abnormal areas, and generates a marine environment anomaly mark map; The biological hotspot extraction module extracts spectral reflectance values, band data, water temperature and salinity based on the marine environmental anomaly marker map, identifies pixels that meet the thermal and saline characteristics, matches the marine biological spectral template, superimposes geographic location and time tags, performs spatial positioning, and generates marine biological hotspot distribution results; The population path tracking module extracts time tags and geographic information based on the distribution results of marine biological hotspots, identifies the spatial distance between hotspot areas in adjacent time periods, distinguishes the offset direction and continuous trajectory, and integrates path segments with consistent offset directions to obtain remote sensing tracking species path segments; The ecological change analysis module is based on the remote sensing tracking species path segment, and performs time series comparison according to the water temperature change trend, salinity level and phytoplankton concentration of the path segment, identifies the synchronous turning point position, locates the overlapping position of the turning point area and the hotspot, and outputs the remote sensing indicator species ecological response interaction layer.

Images