Remote sensing evaluation method for artemia cyst resource quantity

Abstract

The invention relates to the technical field of remote sensing application and biological resource evaluation, in particular to an artemia cyst resource quantity remote sensing evaluation method, which comprises the following steps: firstly, acquiring a multi-temporal optical remote sensing image covering a research area and completing radiation, atmosphere and geometry unification preprocessing to obtain surface reflectance data; secondly, extracting a water body range based on the improved water body index, and constructing a water depth inversion model in combination with the actually measured water depth to obtain pixel-level water depth; screening sensitive spectral characteristics and establishing an artemia cyst biomass inversion model by utilizing an actually measured artemia cyst wet weight sample and the corresponding reflectivity, and generating pixel-level biomass density; finally, the pixel water depth and the pixel biomass are coupled and accumulated in the uniform space grid, and an artemia cyst resource quantity evaluation result is output. According to the method, the defects of water depth deficiency, rough density and scale inconsistency in the prior art are overcome, and rapid, fine and dynamic monitoring of the salt lake artemia cyst resource quantity can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing application and biological resource assessment, in particular to a remote sensing assessment method for Artemia egg resource quantity. Background Art

[0002] Artemia eggs are a crucial biological resource for aquaculture and salt lake ecological monitoring. Accurate assessment of their abundance is directly linked to sustainable industrial development and the development of ecological regulation plans. Existing assessment methods often rely on fixed-point sampling and statistical extrapolation, or simply use single-period remote sensing imagery to estimate lake area and average Artemia egg density, then multiply by empirical coefficients to infer abundance. Due to a lack of pixel-level water depth information, the failure to establish a spectral inversion model for Artemia egg biomass, and the neglect of density differences caused by temporal and spatial variations in lake water, resource estimates are highly biased and inefficient, making them difficult to meet the needs of dynamic monitoring.

[0003] Existing technologies usually use a single water body index to extract the scope of the lake, and then estimate the Artemia egg reserves through empirical proportions or fixed coefficients: ① Water depth varies with time and space and is not effectively portrayed, resulting in distorted estimates of the Artemia egg habitat volume; ② Artemia egg biomass density is often replaced by historical means, which cannot reflect actual distribution differences; ③ Depth and biomass are not coupled within a unified spatial grid, resulting in inconsistent scales of results and error transmission and amplification. Summary of the Invention

[0004] In response to the many problems existing in the above-mentioned prior arts, the present invention provides a remote sensing assessment method for Artemia egg resources. The present invention uses optical remote sensing surface reflectance as the information source, first inverts the water depth of lake pixels to determine the habitat volume, then inverts the Artemia egg biomass density to characterize the biological distribution intensity, and finally multiplies and accumulates the two types of pixel data within a unified spatial grid to output the regional-level resource quantity, thereby realizing non-destructive, efficient and dynamic Artemia egg resource assessment.

[0005] A remote sensing assessment method for Artemia egg resources comprises the following steps: Acquiring optical remote sensing satellite images covering the study area, and preprocessing the optical remote sensing satellite images to obtain surface reflectance data; Extracting the water body range of the study area based on the surface reflectivity data, and establishing a water depth inversion model based on the surface reflectivity and the corresponding measured water depth to generate pixel-level water depth data; Establishing an Artemia egg biomass inversion model based on the surface reflectance and the corresponding measured Artemia egg biomass to generate pixel-level Artemia egg biomass data; The pixel-level water depth data and the pixel-level Artemia egg biomass data are compounded and accumulated in the same spatial grid to obtain an Artemia egg resource assessment result of the study area.

[0006] Preferably, multi-temporal and multi-source optical remote sensing satellite images are used, and the optical remote sensing satellite images contain both green light bands and near-infrared bands to support water body range extraction based on the normalized difference water index, and have the ability to perform repeated observations to form time series image data of the study area.

[0007] Preferably, the preprocessing of the optical remote sensing satellite imagery sequentially implements image band calculation, orthorectification, radiometric calibration, image cropping and atmospheric correction to obtain surface reflectance data.

[0008] Preferably, when extracting the water body range in the study area, an improved normalized difference water index is used in combination with a dynamic threshold segmentation algorithm to determine the water body boundary.

[0009] Preferably, when establishing a water depth inversion model, Pearson correlation analysis is first used to screen spectral features that are sensitive to water depth changes, and then a statistical regression algorithm is used to construct a water depth inversion model to generate pixel-level water depth data.

[0010] Preferably, after generating pixel-level water depth data, a water storage capacity estimation model is established based on a pixel statistical method of the relationship between area and water depth, and the lake water storage capacity is obtained through pixel-by-pixel water volume calculation and spatial integration operations.

[0011] Preferably, when establishing the Artemia egg biomass inversion model, Pearson correlation analysis is used to screen spectral features related to Artemia egg biomass, and a statistical regression algorithm is used to construct the Artemia egg biomass inversion model.

[0012] Preferably, the accuracy of the Artemia egg biomass inversion model is verified using independent samples.

[0013] Preferably, the Artemia egg resource quantity is assessed by multiplying the Artemia egg biomass density of a pixel by the water volume of the corresponding pixel in the same spatial grid, and summing the result for all valid pixels.

[0014] Preferably, the impact of lake area changes and meteorological factors on brine shrimp egg resources is further analyzed to improve the conclusion of resource assessment.

[0015] Compared with the prior art, the advantages and beneficial effects of the present invention are: Through optical remote sensing image preprocessing and surface reflectance acquisition, the physical consistency of multi-source and multi-temporal data was achieved; by improving the water body index to extract the water body range and establish a water depth inversion model, the dynamic acquisition of pixel-level water depth was achieved; through spectral feature screening and statistical regression to establish an Artemia egg biomass inversion model, spatially continuous biomass density estimation was achieved; by coupling water depth data and biomass data within the same spatial grid and accumulating them, a refined assessment of resource quantity was achieved, which significantly reduced the workload of field sampling and improved the estimation accuracy and timeliness. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0017] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0018] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise," "include," etc. used herein indicate the presence of the features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0019] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0020] like Figure 1 As shown, a remote sensing assessment method for Artemia egg resources comprises the following steps: Acquiring optical remote sensing satellite images covering the study area, and preprocessing the optical remote sensing satellite images to obtain surface reflectance data; Preferably, multi-temporal and multi-source optical remote sensing satellite images are used, and the optical remote sensing satellite images contain both green light bands and near-infrared bands to support water body range extraction based on the normalized difference water index, and have the ability to perform repeated observations to form time series image data of the study area.

[0021] This application comprehensively utilizes multi-source satellite remote sensing data to conduct lake depth inversion, construct models for estimating water storage capacity and Artemia egg biomass, and thus assess and predict Artemia egg resources. This application not only fills the research gap in remote sensing monitoring of Artemia egg resources in salt lakes, but also provides new ideas for resource assessment in similar hypersaline ecosystems, with important scientific value and application prospects for achieving sustainable resource utilization.

[0022] In the specific implementation, data source acquisition and preprocessing are carried out, where the data sources include images of the study area obtained from the Gaofen-1 satellite, the Landsat series (Landsat-5 / 7 / 8) and the Sentinel-2 imagery. Data source preprocessing includes image band calculation, orthorectification, radiometric calibration, image cropping and atmospheric correction.

[0023] Preferably, the preprocessing of the optical remote sensing satellite imagery sequentially implements image band calculation, orthorectification, radiometric calibration, image cropping and atmospheric correction to obtain surface reflectance data.

[0024] In the implementation, the water body mentioned above was selected as the research area in Lake Aibi, Xinjiang. Before the fishing ban in 2017, the Artemia resources in Lake Aibi ranked first among many salt lakes in my country, and its output nearly reached two-thirds of the total output in the country. It is an ideal area for studying the ecology of salt lake Artemia and its resource development.

[0025] Based on ENVI5.6, 54 multi-source remote sensing images from 2012 to 2024 were preprocessed. The main process is described as follows by sensor: The data level selected by GF-1WFV is L1A. The data requires that the image is cloudless or has little cloud cover, does not block the Aibi Lake area, and has high data quality. The data used are all from its wide-field multispectral camera (WFV, WideFieldView). This sensor has the characteristics of wide coverage and high revisit period, and is suitable for large-scale surface dynamic monitoring research. Preprocessing process: (1) Radiation correction, including two steps: radiation calibration and atmospheric correction. First, the DN value recorded by the sensor is converted into absolute radiation brightness using calibration parameters. Then, the Quick Atmospheric Correction (QUAC) algorithm is used to eliminate or weaken atmospheric interference and obtain surface reflectance data. (2) Orthorectification. Based on the RPC model, strict geometric correction is implemented in combination with 1:50,000 DEM data to eliminate projection differences caused by terrain. (3) Band operation. Calculate the NDWI index and clip the study area based on the vector boundary. (4) Clipping of the study area.

[0026] The data level selected for Landsat-7ETM+ / 8OLI is L1, with strip number 146 / 029 and a spatial resolution of 30m. Data requirements are the same as above, considering that only Landsat-7 and Landsat-8 were used during the study period. Preprocessing process: (1) Strip repair (only for Landsat-7ETM+). For the regular strip missing problem in images acquired after May 31, 2003, the Landsat_gapfill tool in ENVI is used to repair it and maximize the recovery of missing pixel information. (2) Radiometric calibration and atmospheric correction steps are the same as above. (3) Band calculation. (4) Study area cropping.

[0027] Sentinel-2A / BMSI uses L2A data products. This level of data has been atmospherically corrected and includes surface reflectance products, making it directly usable for quantitative remote sensing analysis. Preprocessing involves direct band calculations and cropping of the study area.

[0028] Extracting the water body range of the study area based on the surface reflectivity data, and establishing a water depth inversion model based on the surface reflectivity and the corresponding measured water depth to generate pixel-level water depth data; Preferably, when extracting the water body range in the study area, an improved normalized difference water index is used in combination with a dynamic threshold segmentation algorithm to determine the water body boundary.

[0029] The water body boundary is extracted based on the water boundary information required for the improved NDWI extraction. Based on the measured water depth data and its synchronous satellite data, the Pearson correlation analysis method is used to screen the optimal spectral feature combination sensitive to water depth changes and establish a water depth inversion model.

[0030] Water has a high reflectivity in the green band (500-600nm) and exhibits strong absorption characteristics in the near-infrared band (800-900nm). Therefore, the Normalized Difference Water Index (NDWI) is used to extract the boundaries of water bodies. This index significantly improves the separability of water bodies from background objects by enhancing the difference in reflectivity between water bodies in the visible and near-infrared bands. The specific expression is:

[0031] In the formula 、 Representing the surface reflectance in the green and near-infrared bands, respectively. NDWI feature analysis of multispectral remote sensing images reveals that the water areas of Lake Aibi appear prominently bright in the imagery, while the non-water areas appear distinctly dark. This distinct difference facilitates high-precision extraction of instantaneous water boundaries. Using a normalized ratio calculation and a dynamic threshold segmentation algorithm, the water boundary was successfully extracted. The calculation formula is: Overall accuracy = number of correctly classified sample points / total number of sample points × 100%. The final overall accuracy of water boundary extraction was 95.70%, which met the research requirements.

[0032] Preferably, when establishing a water depth inversion model, Pearson correlation analysis is first used to screen spectral features that are sensitive to water depth changes, and then a statistical regression algorithm is used to construct a water depth inversion model to generate pixel-level water depth data.

[0033] Water depth inversion model construction: Based on blue ( ),green( ),red( ) and near infrared ( ) band reflectance data, a multidimensional spectral feature set was constructed, including the original band reflectance, mathematical transformation features (logarithmic transformation, inverse transformation, exponential transformation), and multi-band combination features (band ratio, normalized index, differential features, etc.). The Pearson correlation analysis method was used to systematically evaluate the correlation between each spectral feature and the measured water depth value, and the correlation pattern between key features and water depth was visualized through the Origin heat map. A water depth inversion model was constructed based on the statistical regression method, and 6 highly correlated band combinations identified by the heat map analysis were selected. 、 、 、 、 and As candidate independent variables, a water depth statistical regression inversion model is established.

[0034] Preferably, after generating pixel-level water depth data, a water storage capacity estimation model is established based on a pixel statistical method of the relationship between area and water depth, and the lake water storage capacity is obtained through pixel-by-pixel water volume calculation and spatial integration operations.

[0035] A water storage capacity estimation model was constructed and the pixel statistics method based on the area-water depth relationship was used for calculation. The specific process is as follows: (1) Calculation of water volume in a single pixel:

[0036] in, For the The area of ​​the pixel (m 2 ), which is directly determined by the spatial resolution of remote sensing images; For the The water depth (m) of each pixel is obtained through the water depth inversion model.

[0037] (2) Calculation of total water storage capacity:

[0038] This method achieves accurate estimation of lake water storage capacity through pixel-by-pixel water volume calculation and spatial integration operation.

[0039] Establishing an Artemia egg biomass inversion model based on the surface reflectance and the corresponding measured Artemia egg biomass to generate pixel-level Artemia egg biomass data; Preferably, when establishing the Artemia egg biomass inversion model, Pearson correlation analysis is used to screen spectral features related to Artemia egg biomass, and a statistical regression algorithm is used to construct the Artemia egg biomass inversion model.

[0040] Construct and verify the Artemia egg biomass remote sensing inversion model. Specifically, based on the 2023 Artemia egg wet weight field measurement data and its synchronous satellite data, the optimal spectral feature combination sensitive to Artemia egg biomass was selected through Pearson correlation analysis, and then the Artemia egg biomass remote sensing inversion model was constructed and the model accuracy was evaluated. The specific process includes: (1) Establish the spatial correspondence between the wet weight data of the sampling points and the image feature values ​​through pixel-by-pixel matching; (2) Draw a scatter plot of eigenvalues ​​and biomass to reveal their nonlinear distribution characteristics; (3) Referring to the water depth inversion model framework, a biomass prediction model was established using statistical regression analysis functions (power, exponential, linear, logarithmic, and quadratic polynomial) to ensure the consistency of the method system.

[0041] Preferably, the accuracy of the Artemia egg biomass inversion model is verified using independent samples.

[0042] After the remote sensing inversion model of Artemia egg biomass is established, the embodiment randomly divides all field measured wet weight samples into modeling samples and verification samples according to space and time stratification. The former is used for spectral feature screening and parameter fitting, and the latter is completely isolated in the modeling stage; then the inversion model is applied to the corresponding pixels of the verification samples, and the predicted values ​​and measured values ​​are compared point by point, and the error indicators are statistically analyzed to test the generalization ability of the model on unknown data and ensure the reliability of subsequent resource estimation.

[0043] The pixel-level water depth data and the pixel-level Artemia egg biomass data are compounded and accumulated in the same spatial grid to obtain an Artemia egg resource assessment result of the study area.

[0044] Preferably, the Artemia egg resource quantity is assessed by multiplying the Artemia egg biomass density of a pixel by the water volume of the corresponding pixel in the same spatial grid, and summing the result for all valid pixels.

[0045] Constructing an Artemia egg resource assessment model: The Artemia egg resource model is constructed based on the collaborative analysis of multispectral remote sensing data and ground-based measured data. The spatial estimation of regional Artemia egg resources is achieved by coupling the biomass remote sensing inversion model with the water storage estimation model. The data units are biomass (g / 10L) and water storage (m 3 ), while the model inversion requires unit unification preprocessing (kg / m 3 and m 3 ). The mathematical expression of the model is as follows: The total Artemia egg resources are obtained by summing the biomass of each pixel:

[0046] in, is the total Artemia egg resources (kg), For the Artemia egg biomass (density) per pixel (kg / m 3 ), For the The volume of water in each pixel (m 3 ), is the total number of valid pixels in the study area.

[0047] Preferably, the impact of lake area changes and meteorological factors on brine shrimp egg resources is further analyzed to improve the conclusion of resource assessment.

[0048] In order to clarify the regulatory mechanism of the external environment on the amount of Artemia egg resources, the embodiment further constructs a time series dataset of lake area and meteorological elements, and conducts multivariate statistical analysis based on the obtained Artemia egg resource results. By quantifying the coupling relationship between factors such as lake area fluctuations, temperature, precipitation and resource amount, the main driving factors and their relative contributions are identified, thereby providing a scientific basis for the sustainable development of resources.

[0049] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware.

[0050] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included within the scope of the claims of the present application.

Claims

1. A remote sensing assessment method for Artemia egg resources, characterized in that: The following steps are involved: Acquiring optical remote sensing satellite images covering the study area, and preprocessing the optical remote sensing satellite images to obtain surface reflectance data; Extracting the water body range of the study area based on the surface reflectivity data, and establishing a water depth inversion model based on the surface reflectivity and the corresponding measured water depth to generate pixel-level water depth data; Establishing an Artemia egg biomass inversion model based on the surface reflectance and the corresponding measured Artemia egg biomass to generate pixel-level Artemia egg biomass data; The pixel-level water depth data and the pixel-level Artemia egg biomass data are compounded and accumulated in the same spatial grid to obtain an Artemia egg resource assessment result of the study area.

2. The method according to claim 1, characterized in that Multi-phase and multi-source optical remote sensing satellite images are used. The optical remote sensing satellite images contain both green light bands and near-infrared bands to support water body range extraction based on the normalized difference water index, and have the ability to perform repeated observations to form time series image data of the study area.

3. The method according to claim 1, characterized in that The optical remote sensing satellite image is preprocessed by sequentially performing image band calculation, orthorectification, radiation calibration, image cropping and atmospheric correction to obtain surface reflectance data.

4. The method according to claim 1, wherein When extracting the water body range in the study area, the improved normalized difference water index is used in combination with the dynamic threshold segmentation algorithm to determine the water body boundary.

5. The method according to claim 1, wherein When establishing the water depth inversion model, Pearson correlation analysis is first used to screen spectral features that are sensitive to water depth changes, and then a statistical regression algorithm is used to construct the water depth inversion model to generate pixel-level water depth data.

6. The method according to claim 1, characterized in that After generating pixel-level water depth data, a water storage capacity estimation model is established based on the pixel statistical method of the relationship between area and water depth, and the lake water storage capacity is obtained through pixel-by-pixel water volume calculation and spatial integration operations.

7. The method according to claim 1, characterized in that When establishing the Artemia egg biomass inversion model, Pearson correlation analysis was used to screen the spectral features related to Artemia egg biomass, and a statistical regression algorithm was used to construct the Artemia egg biomass inversion model.

8. The method according to claim 1, characterized in that The accuracy of the Artemia egg biomass inversion model was verified using independent samples.

9. The method according to claim 1, characterized in that The assessment of Artemia egg resources is achieved by multiplying the Artemia egg biomass density of a pixel by the water volume of the corresponding pixel in the same spatial grid and summing the results for all valid pixels.

10. The method according to claim 1, characterized in that Further analysis will be conducted on the impact of lake area changes and meteorological factors on Artemia egg resources to improve the conclusions of resource assessment.

Images