Method, System and Medium for Extracting Remote Sensing Ecological Index Considering Human Activity Intensity

Abstract

The present invention discloses a method, system and medium for extracting a remote sensing ecological index considering human activity intensity, belonging to the technical field of remote sensing ecological environment monitoring and evaluation. The steps include: S1, aiming at the problems of difficult acquisition of multi-source data, low accuracy and insufficient calculation efficiency in large-scale habitat quality assessment in alpine arid areas, an improved method for remote sensing ecological index based on the Google Earth Engine cloud platform and human activity intensity is proposed; S2, a salinity index is introduced, and an improved method for extracting remote sensing ecological index is proposed; S3, combining four indicators of night light, percentage of built-up land, population density and landscape development index, a method for calculating human activity intensity is proposed; S4, coupling the improved remote sensing ecological index, a remote sensing ecological index considering human activity intensity is established to realize rapid and refined calculation of the remote sensing ecological index. The present invention provides strong data support for ecological management and scientific decision-making in alpine mountainous areas.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing ecological environment monitoring and evaluation, and in particular to a method, system and medium for extracting remote sensing ecological indices considering the intensity of human activities, which are used for long-term habitat quality monitoring and assessment of the impact of human activities. Background Art

[0002] Remote sensing ecological indices are a method widely used for evaluating the quality of the ecological environment. They can extract key ecological indicators such as greenness, dryness, humidity and heat from remote sensing data, and use the principal component analysis method for comprehensive evaluation. However, the calculation of traditional remote sensing ecological indices has problems such as low calculation efficiency and difficulty in long-term monitoring, which limits their application in large-scale ecological environment assessment. In addition, the ecological environment in alpine and arid regions has unique characteristics, and traditional remote sensing ecological indices are difficult to accurately reflect the changes in habitat quality in such regions. For example, the applicability of the heat index in cold and arid regions is relatively low. Therefore, the present invention introduces the Google Earth Engine cloud platform to establish a spatio-temporal database of habitat quality, improving the calculation efficiency and timeliness of remote sensing ecological indices. At the same time, aiming at the ecological characteristics of the Altai Mountains, a salinity index is introduced to optimize the remote sensing ecological index and improve its applicability. On the other hand, existing remote sensing ecological indices do not fully consider the impact of human activities on the ecological environment. The present invention proposes a method for calculating the intensity of human activities based on night lights, built-up land, population density and landscape development index, and couples it with the improved remote sensing ecological index to form a remote sensing ecological index considering the intensity of human activities, achieving a more comprehensive evaluation of the ecological environment quality. Summary of the Invention

[0003] The object of the present invention is to provide a method, system and medium for extracting remote sensing ecological indices considering the intensity of human activities, which proposes a method for calculating the intensity of human activities based on night lights, built-up land, population density and landscape development index, and couples it with the improved remote sensing ecological index to form a remote sensing ecological index considering the intensity of human activities, achieving a more comprehensive evaluation of the ecological environment quality, and can provide strong data support for ecological management and scientific decision-making in alpine mountainous areas.

[0004] To achieve the above object, the present invention provides a method, system and medium for extracting remote sensing ecological indices considering the intensity of human activities, including the following steps:

[0005] Step S1, based on the Google Earth Engine cloud platform, extract cloud, cloud shadow and cirrus information through operations, generate a cloud-free mask, and remove the pixels affected by atmospheric pollution;

[0006] Crop the data according to a preset summer time window to lock the spatial range of the study area;

[0007] Calculate the water body distribution using the improved Normalized Difference Water Index, create a non-water body mask through the threshold method, and exclude the influence of water areas on the terrestrial ecological assessment;

[0008] Integrate the preprocessed images of different years in summer according to the time series to construct a long-term spatio-temporal database of habitat quality;

[0009] Step S2: Calculate four ecological parameters, namely the Normalized Vegetation Index, Humidity Index, Salinity Index, and Dryness Index, from the bands of the remote sensing dataset. Perform range normalization on each of them to eliminate the dimension difference. Extract the first principal component of the covariance matrix of the four indicators through principal component analysis to capture the maximum variation information of multi-dimensional ecological characteristics. Quadratically normalize the first principal component to generate a remote sensing ecological index in the range of 0 - 1;

[0010] Step S3: Integrate four indicators, namely night-time light, percentage of built-up land, population density, and landscape development index. Use the Analytic Hierarchy Process to calculate the weights of the four indicators of human activity intensity, and generate a human activity intensity value in the range of 0 - 1 through weighted normalization;

[0011] Step S4: Use the entropy method to obtain the weights of human activity intensity and the remote sensing ecological index, calculate the remote sensing ecological index model considering human activity intensity, and divide it into five grades: poor, bad, medium, good, and excellent according to the equal value method;

[0012] Adopt the difference method to calculate the difference of the remote sensing ecological index considering human activity intensity within different time periods, and obtain the change map of the overall ecological environment quality level in different time periods of the study area.

[0013] Preferably, step S1 includes the following steps:

[0014] S101: Based on the Google Earth Engine cloud platform, through calling the API and functions of Google Earth Engine, the acquisition, processing, and analysis of remote sensing data are realized;

[0015] S102: Cloud removal and image preprocessing. Define the cloud mask extraction function bitwiseExtract: Parse the StateQA band to extract cloud status, cloud shadow, and cirrus cloud information;

[0016] Create a cloud-free image function to retain only the pixel data without clouds, cloud shadows, and cirrus clouds, and remove the influence of atmospheric pollution;

[0017] S103: Set the research period at intervals of 3 years;

[0018] Set the research season as July - September;

[0019] Generate a DateRange list with each 3-year period as a time segment;

[0020] S104: Calculate the improved normalized difference water index;

[0021] Set the water threshold: Identify water using the normalized difference water index < 0.2;

[0022] Generate a water mask to filter out the interference of water areas on the ecological index;

[0023] S105: Image cropping and water mask application Define the GetIMG image acquisition function: Filter the MODIS image and extract data for the study area according to DateRange;

[0024] Calculate the normalized difference water index and apply the water mask: Generate a water mask using mndwicol.mean().clip(sa).lt(0.2);

[0025] Calculate the minimum - maximum value sts_minmax of the image: Calculate the minimum and maximum values of each image for normalization processing.

[0026] Preferably, step S2 includes the following steps:

[0027] S201: Perform index calculation and normalization calculation, and the calculation formula is:

[0028] Salinity index S6 The calculation formula is: ;

[0029] In the formula, NIR : Near - infrared band, Red : Red - light band, Green : Green - light band;

[0030] Normalized difference vegetation index NDVI The calculation formula is: ;

[0031] In the formula, NIR : Near - infrared band, Red : Red - light band;

[0032] Humidity index WET The calculation formula is:

[0033] ;

[0034] In the formula, Red : Red - light band, NIR : Near - infrared band, Blue : Blue - light band, Green : Green - light band, MIR : Middle - infrared band, SWIR1 : Short - wave infrared 1, SWIR2 : Short - wave infrared 2;

[0035] Normalized Building Soil Index NDBSI The calculation formula is as follows: ;

[0036] In the formula, BI : Bare soil index, IBI : Improved building index, Red : Red light band, NIR : Near-infrared band, Blue : Blue light band, Green : Green light band, MIR : Mid-infrared band, SWIR1 : Short-wave infrared 1, SWIR2 : Short-wave infrared 2;

[0037] S202: Perform range normalization on different indices so that each index is between 0 and 1: ;

[0038] S203: Perform principal component analysis, calculate the covariance matrix of the four standardized ecological parameters, and extract the first principal component PC1 , calculate the weight of the first principal component, and the calculation method is: set the input standardized data matrix as X , through principal component analysis, obtain the first principal component PC1 = XW1 , where W1 is the weight vector of the first principal component obtained by principal component analysis calculation;

[0039] S204, perform Min-Max normalization on the first principal component PC1 , adjust its value to between 0 and 1 to obtain the remote sensing ecological index RSEI value, and the calculation formula is as follows: ;

[0040] S205: Record the remote sensing ecological index value every three years, and make a line chart of the remote sensing ecological index value.

[0041] Preferably, step S3 includes the following steps:

[0042] S301: Obtain data on night lights, building land percentage, population density, and landscape development index, so that the obtained data covers the study area;

[0043] Perform projection transformation on data from different sources, unify them to the same geographic coordinate system or projection coordinate system, and perform spatial matching;

[0044] Based on the boundary of the study area, crop all data to ensure that the data coverage is consistent;

[0045] S302: Perform extreme value normalization on each indicator so that each indicator is between 0 and 1: ;

[0046] S303: To determine the indicator weights for the evaluation of human activity intensity, use the analytic hierarchy process to construct a hierarchical structure model;

[0047] This hierarchical structure model includes three levels: the goal level, the criterion level, and the indicator level. The goal level is human activity intensity. The criterion level consists of four factors: night-time light, population density, percentage of built-up land, and landscape development index. The indicator level uses night-time light, percentage of built-up land, population density, and landscape development index as data sources for support;

[0048] On this basis, construct a pairwise comparison judgment matrix according to the literature analysis and score it according to the Saaty 1-9 scale principle;

[0049] By calculating the eigenvector of the judgment matrix, obtain the normalized weights of each indicator to quantify the relative importance of each factor in the evaluation system;

[0050] S304: Calculate the human activity intensity index by weighting, and sum up the four normalized indicators according to their weights:

[0051] ;

[0052] XNL : Night-time light, XPD : Population density, XLU : Built-up land, XLDI : Landscape development intensity, H : Human activity intensity value within the range of 0-1.

[0053] Preferably, step S4 includes the following steps:

[0054] S401: Use the entropy method to evaluate the weights of human activity intensity and remote sensing ecological index, couple the human activity intensity value with the remote sensing ecological index, and construct a coupling model;

[0055] When using Arcgis 10.7, use the equal value method to classify the results of the remote sensing ecological index considering human activity intensity into five levels: poor, bad, medium, good, and excellent;

[0056] Through spatial statistical methods, calculate the area and proportion of ecological environment quality at different levels and different time periods;

[0057] S403: Statistically calculate the average value and standard deviation of the remote sensing ecological index values considering human activity intensity;

[0058] S404: The difference method is used to calculate the differences in the remote sensing ecological index considering human activity intensity over different time periods. The remote sensing ecological index considering human activity intensity is divided into five levels: significantly deteriorated, slightly deteriorated, unchanged, slightly improved, and significantly improved.

[0059] S405: It is statistically shown that the grades of the remote sensing ecological index considering human activity intensity present a dynamic change process, which can be effectively visualized through a Sankey diagram.

[0060] The present invention provides a system for extracting the remote sensing ecological index considering human activity intensity. By applying the above method for extracting the remote sensing ecological index considering human activity intensity, it includes the following modules:

[0061] Data acquisition and preprocessing module: Integrate data on landscape development index, night lights, population density, and built-up land. After cloud removal, radiometric correction, atmospheric correction, and resampling processing, store them in the Google Earth Engine platform.

[0062] Remote sensing ecological index calculation module: Calculate the remote sensing ecological index on the Google Earth Engine platform, including the normalized difference vegetation index, humidity, dryness index, and introduce a salinity index optimization algorithm.

[0063] Human activity intensity calculation module: Process the night light data, combine the percentage of built-up land, landscape development index, and population density, and use the multi-factor coupling method to quantify the human activity intensity.

[0064] Remote sensing ecological index calculation and analysis module considering human activity intensity: Couple the improved remote sensing ecological index with the human activity intensity to construct the remote sensing ecological index considering human activity intensity and comprehensively evaluate the habitat quality.

[0065] Spatio-temporal database and visualization module: Establish a spatio-temporal database of habitat quality to support query, time series analysis, and spatial distribution research.

[0066] Application and optimization module: Based on the remote sensing ecological index considering human activity intensity, evaluate the risk distribution of habitat quality and provide decision-making support for ecological protection.

[0067] The present invention provides a storage medium on which a computer program is stored: When the computer program is executed by a processor, it implements the above method for extracting the remote sensing ecological index considering human activity intensity.

[0068] Therefore, by adopting the above method, system, and medium for extracting the remote sensing ecological index considering human activity intensity, compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. Based on the Google Earth Engine cloud platform, the present invention establishes a spatio-temporal database of habitat quality, integrates key ecological indicators, and utilizes the efficient cloud parallel computing ability of the Google Earth Engine to achieve dynamic monitoring and refined analysis of habitat quality, improving the integrity of data acquisition and monitoring efficiency.

[0070] 2. The present invention introduces a salinity index and proposes an improved remote sensing ecological index method that better suits the ecological characteristics of cold and arid regions, enhancing the representation ability of the state and spatial heterogeneity of the ecosystem and improving the scientificity and applicability of ecological assessment.

[0071] 3. The present invention combines night-time light, population density, built-up land, and landscape development index, proposes a calculation method for human activity intensity, and couples it into the improved remote sensing ecological index to establish a remote sensing ecological index considering human activity intensity, achieving the comprehensive integration of natural and anthropogenic factors and improving the comprehensiveness and accuracy of habitat quality assessment.

[0072] 4. The present invention designs and develops a remote sensing ecological index extraction system considering human activity intensity, integrating functional modules such as data collection, cloud computing, index analysis, dynamic monitoring, and visualization display, realizing integrated and automated operation, enhancing the practicality and operability of the system, providing high-precision quantitative support for decisions such as the optimization of ecological protection red lines and the layout of land space restoration projects, and significantly improving the scientific decision-making ability and operational work level of ecological management departments.

[0073] The technical solutions of the present invention will be further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 is a schematic flow chart of the method of the present invention;

[0075] Figure 2 is a schematic sub-flow chart of the method of the present invention;

[0076] Figure 3 is a line graph of RSEI values for some years in the Altai Mountains;

[0077] Figure 4 is a schematic diagram of the human activity intensity index in the Altai Mountains in 2000;

[0078] Figure 5 is a schematic diagram of the human activity intensity index in the Altai Mountains in 2021;

[0079] Figure 6 is a classification map of HRSEI in the Altai Mountains in 2000;

[0080] Figure 7It is the HRSEI classification map of the Altay Mountains in 2003;

[0081] Figure 8 It is the HRSEI classification map of the Altay Mountains in 2006;

[0082] Figure 9 It is the HRSEI classification map of the Altay Mountains in 2009;

[0083] Figure 10 It is the HRSEI classification map of the Altay Mountains in 2012;

[0084] Figure 11 It is the HRSEI classification map of the Altay Mountains in 2015;

[0085] Figure 12 It is the HRSEI classification map of the Altay Mountains in 2018;

[0086] Figure 13 It is the HRSEI classification map of the Altay Mountains in 2021;

[0087] Figure 14 It is the table of the proportion of the area of each level of HRSEI in the Altay Mountains from 2000 to 2021;

[0088] Figure 15 It is the line graph of the average value and standard deviation of HRSEI in the Altay Mountains from 2000 to 2021;

[0089] Figure 16 It is the change graph of HRSEI values in the Altay Mountains from 2000 to 2012;

[0090] Figure 17 It is the change graph of HRSEI values in the Altay Mountains from 2012 to 2021;

[0091] Figure 18 It is the change graph of HRSEI values in the Altay Mountains from 2000 to 2021;

[0092] Figure 19 It is the Sankey diagram of the ecological grade change in the Altay Mountains;

[0093] Figure 20 It is the system schematic diagram of the present invention. Detailed implementation manners

[0094] In order to make the objectives, technical solutions, and advantages of the embodiments disclosed in the present invention clearer and more understandable, the following further elaborates on the embodiments of the present invention in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely used to explain the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts fall within the scope of protection of this application. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout.

[0095] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0096] Similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0097] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the present invention is usually placed during use. It is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0098] In the description of the present invention, it should also be noted that unless otherwise clearly specified and limited, the terms "set", "installed", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0099] Embodiment

[0100] As Figures 1 - 20 shown, a method, system, and medium for extracting a remote sensing ecological index considering human activity intensity according to the present invention;

[0101] As Figures 1 - 2, A method for extracting a remote sensing ecological index considering human activity intensity includes the following steps:

[0102] Step S1, based on the Google Earth Engine (GEE) cloud platform, extract cloud, cloud shadow and cirrus information through operations to generate a cloud-free mask, remove pixels affected by atmospheric pollution, and ensure the reliability of surface reflectance data. For the Altai Mountains, crop the data according to the preset summer time window (July - September) to reduce seasonal fluctuation interference and lock in the spatial scope of the study area. Calculate the water body distribution using the improved Normalized Difference Water Index (MNDWI), and create a non-water body mask through a threshold method (MNDWI < 0.2) to exclude the influence of water areas on terrestrial ecological assessment. Integrate the preprocessed images of different summer years in time series to construct a long-term habitat quality spatio-temporal database. Among them, step S1 includes the following steps:

[0103] S101: Based on the GEE cloud platform, through calling the API and functions of GEE, the acquisition, processing and analysis of remote sensing data are realized. The user uploads a vector data set and defines the boundary of the study area. Load the data sets, MOD09A1: MODIS (MOD09A1) data, including Surface Reflectance. MOD13A1: MODIS (MOD13A1) data, including Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI). table: User-defined vector data (study area).

[0104] S102: Cloud removal and image preprocessing Define the cloud mask extraction function bitwiseExtract: Parse the StateQA band to extract cloud state, cloud shadow and cirrus information. Create a cloud-free image function to only retain the pixel data of cloud-free, cloud-shadow-free and cirrus-free pixels, and remove the influence of atmospheric pollution. bitwiseExtract(value, fromBit, toBit) extracts specific bits of the QA band. cloudState.eq(0) && cloudShadowState.eq(0) && cirrusState.eq(0) ensures cloud-free, cloud-shadow-free and cirrus-free. updateMask(mask) applies the mask to retain the cloud-free area.

[0105] S103: Set the research period (2000 - 2021) for the time series at intervals of 3 years. Set the research season (July - September) to reduce seasonal effects. Generate a list of DateRange (each 3 - year period from 2000 to 2021).

[0106] S104: Calculate the Modified Normalized Difference Water Index (MNDWI). Define the cal_mndwi calculation function: Calculate MNDWI = (Green band - Shortwave infrared band) / (Green band + Shortwave infrared band). Set the water threshold: Use MNDWI < 0.2 to identify water bodies. Generate a water mask (WaterMask): Filter out the interference of water areas on ecological indices.

[0107] S105: Image cropping and application of water mask. Define the GetIMG image acquisition function: Filter MODIS images and extract data for the study area according to DateRange. Calculate MNDWI and apply the water mask: Use mndwicol.mean().clip(sa).lt(0.2) to generate WaterMask. Calculate the minimum - maximum value of the image sts_minmax: Calculate the minimum and maximum values of each image for normalization.

[0108] In step S2, calculate four ecological parameters, namely the Normalized Difference Vegetation Index (NDVI), Wetness Index (WET), Salinity Index (S6), and Normalized Difference Built - up and Soil Index (NDBSI) through the bands of the remote sensing dataset. Perform range standardization processing on each of them to eliminate the dimension difference. Subsequently, extract the first principal component (Principal Component1, PC1) of the covariance matrix of the four indicators through principal component analysis (PCA) to capture the maximum variation information of multi - dimensional ecological characteristics. Finally, perform secondary normalization on PC1 to generate a remote sensing ecological index (RSEI) in the range of 0 - 1. Among them, step S2 includes the following steps:

[0109] S201: Conduct index calculation and normalization calculation. The calculation formula is:

[0110] Salinity index S6 The calculation formula is: ;

[0111] In the formula,NIR : Near-infrared band, Red : Red light band, Green : Green light band;

[0112] Normalized Difference Vegetation Index NDVI The calculation formula is: ;

[0113] In the formula, NIR : Near-infrared band, Red : Red light band;

[0114] Moisture index WET The calculation formula is:

[0115] ;

[0116] In the formula, Red : Red light band, NIR : Near-infrared band, Blue : Blue light band, Green : Green light band, MIR : Medium-infrared band, SWIR1 : Short-wave infrared 1, SWIR2 : Short-wave infrared 2;

[0117] Normalized Difference Built-up and Soil Index NDBSI The calculation formula is: ;

[0118] In the formula, BI : Bare soil index, IBI : Improved built-up index, Red : Red light band, NIR : Near-infrared band, Blue : Blue light band, Green : Green light band, MIR : Medium-infrared band, SWIR1 : Short-wave infrared 1, SWIR2 : Short-wave infrared 2;

[0119] S202: Perform range normalization (Min-Max normalization). Since the value ranges of different indices are different, normalization is required to make them all between 0 and 1, which can eliminate the dimension difference and make different ecological parameters comparable.

[0120] S203: Perform Principal Component Analysis (PCA), calculate the covariance matrix of the four standardized ecological parameters, and find the main direction of data variation. Extract the first principal component (Principal Component 1, PC1), which captures the most important variation information of the four indicators and can represent the ecological environment characteristics to the greatest extent. Calculation method: Let the input standardized data matrix be X , and obtain the first principal component through PCA PC1 = XW1 , where W1 is PCA the weight vector of the first principal component calculated.

[0121] S204. Perform Min - Max normalization on PC1 , adjust its value to the range between 0 and 1 to obtain RSEI value: ; This can ensure that the finally calculated Remote Sensing Ecological Index (RSEI) has a consistent scale, facilitating comparison of different regions.

[0122] S205: Display the RSEI values for each three - year cycle in Arcgis10.7, and create a line chart of the RSEI values. As shown in Figure 3 , the RSEI values in the Altay Mountains vary between 0.575 and 0.810. The highest value is 0.81 (in 2003), and the lowest value is 0.575 (in 2021). During 2000 - 2021, the RSEI in the Altay Mountains showed irregular volatility characteristics. However, there are also certain phased characteristics. There was a short growth period and the RSEI reached its peak between 2000 and 2003, and the RSEI decreased between 2003 and 2021.

[0123] Step S3: Integrate four types of indicators, namely night - time light (representing the intensity of economic activities), percentage of built - up land (directly reflecting land development), population density (human distribution density), and landscape development index (quantifying landscape fragmentation). Use the Analytic Hierarchy Process (AHP) to calculate the weights of the four indicators of human activity intensity, and generate a human activity intensity value in the range of 0 - 1 through weighted normalization. Among them, step S3 includes the following steps:

[0124] S301: Obtain data such as night - time light, percentage of built - up land, population density, and landscape development index, ensuring that the data covers the study area. Perform projection transformation on data from different sources to unify them to the same geographic coordinate system or projection coordinate system to ensure spatial matching. Based on the study area boundary, crop all data to ensure that the data coverage is consistent. ​

[0125] S302: Since the dimensions and numerical ranges of each indicator are different, normalization processing is required to eliminate the dimensional differences, using Min-Max normalization (0-1 standardization);

[0126] S303: To determine the indicator weights for the evaluation of human activity intensity, the Analytic Hierarchy Process (AHP) is used to construct a hierarchical structure model. This model includes three levels: the target level is human activity intensity, the criterion level consists of four factors: nighttime light, population density, percentage of built-up land, and landscape development index, and the indicator level is supported by specific data sources. On this basis, a pairwise comparison judgment matrix is constructed based on literature analysis and scored according to the Saaty 1-9 scale principle. Subsequently, by calculating the eigenvector of the judgment matrix, the normalized weights of each indicator are obtained to quantify the relative importance of each factor in the evaluation system, thus providing a scientific basis for the quantitative analysis of human activity intensity.

[0127] S304: Calculate the human activity intensity index by weighted summation of the four normalized indicators according to their weights: ;

[0128] The indicator weights are 46.9% for nighttime light XNL 27.9% for population density XPD 16.1% for built-up land XLU 9.1% for landscape development intensity XLDI and finally obtain the human activity intensity value within the range of 0-1. For example Figure 4 、 Figure 5 。

[0129] Step S4: Use the entropy method to obtain the weights of human activity intensity and the Remote Sensing Ecological Index (RSEI), calculate the Human Activity-integrated Remote Sensing Ecological Index (HRSEI) model considering human activity intensity, and classify it into five levels (poor, bad, medium, good, excellent) according to the equal value method. In Arcgis 10.7, the difference method is used to calculate the differences in HRSEI within different time periods, and the overall ecological environment quality grade change maps of the study area from 2000 to 2012, from 2012 to 2021, and the long time series from 2000 to 2021 are obtained. Among them, step S4 includes the following steps:

[0130] S401: Use the entropy method to evaluate the weights of human activity intensity and RSEI, considering their relative importance and information contribution in ecosystem assessment. Calculate the weights of human activity intensity and RSEI. The weight of human activity intensity is 38.6%, and the weight of RSEI is 61.4%. Couple the human activity intensity value with RSEI to construct a Remote Sensing Ecological Index integrating human activity intensity (HRSEI) model, and calculate the final HRSEI value.

[0131] S402: When using Arcgis 10.7, the equal value method was adopted to classify the HRSEI results into five levels: "poor, bad, medium, good, excellent", such as Figure 6 、 Figure 7 、 Figure 8 、 Figure 9 、 Figure 10 、 Figure 11 、 Figure 12 、 Figure 13 。 By spatial statistical methods, calculate the area and proportion of different levels of ecological environment quality in the Altai Mountains from 2000 to 2021, such as Figure 14 。 From Figures 6 - 14 it can be concluded that there are differences in the area and proportion of the RSEI values in different levels in the study area for each year, indicating that the ecological environment quality status is different in different periods. The HRSEI values in the Altai Mountains vary between 0.35 and 0.55. Generally, the area with the medium level accounts for the largest proportion, followed by the area with the good level.

[0132] S403: Statistically calculate the average value and standard deviation of the HRSEI values, such as Figure 15 。 The standard deviation is relatively stable, and the average value of HRSEI reaches an extreme value in 2015, first decreasing and then increasing.

[0133] S404: In Arcgis 10.7, the difference method is used to calculate the differences in HRSEI within different time periods, and obtain the overall ecological environment quality grade change maps of the study area in 2000, 2012, and the long time series from 2000 to 2021. They are classified into ≤ -0.5, (-0.5, -0.2], (-0.2, 0.2), [0.2, 0.5), ≥ 0.5, corresponding to "significantly deteriorated, slightly deteriorated, unchanged, slightly improved, significantly improved", such as Figure 16 、 Figure 17 、 Figure 18Define slight improvement and obvious improvement as ecological improvement, and obvious deterioration and slight deterioration as ecological degradation. From 2000 to 2012, the area of ecological improvement was larger than that of ecological degradation, accounting for 8.9% and 5.4% of the total area respectively, and the ecological environment improved most significantly. From 2012 to 2021, the ecological environment deteriorated most significantly, with about 8.3% of the area deteriorating and only 3.8% of the area improving. Generally, from 2000 to 2021, the ecological environment in Alxa Left Banner changed slightly for the worse.

[0134] S405: The HRSEI levels in the Altai Mountains show a dynamic change process, and this change process can be effectively visualized through a Sankey diagram, such as Figure 19 . Most of the areas with medium HRSEI levels remain stable.

[0135] Such as Figure 20 , A remote sensing ecological index extraction system considering human activity intensity described in the present invention applies the above-mentioned method for extracting a remote sensing ecological index considering human activity intensity, and includes the following modules:

[0136] Data acquisition and preprocessing module: Integrate data such as landscape development index, night lights, population density, and built-up land, perform cloud removal, radiometric correction, atmospheric correction, and resampling processing, and store them in the GEE platform to achieve efficient management and calculation.

[0137] Remote sensing ecological index calculation module: Calculate the RSEI index on the GEE platform, including the normalized difference vegetation index, humidity, and dryness index, and introduce a salinity index optimization algorithm to adapt to the ecological environment in cold and arid regions and improve the evaluation accuracy.

[0138] Human activity intensity calculation module: Process night light data, combine the percentage of built-up land, landscape development index, population density, etc., and use a multi-factor coupling method to quantify human activity intensity.

[0139] HRSEI calculation and analysis module: Couple the improved RSEI and human activity intensity, construct the HRSEI index, comprehensively evaluate the habitat quality, and reveal the spatio-temporal evolution of the ecosystem under natural and human interventions.

[0140] Spatio-temporal database and visualization module: Establish a spatio-temporal database of habitat quality in the Altai Mountains, support query, time series analysis, and spatial distribution research, and display the change trend through visualization methods such as GIS and GEE maps.

[0141] Application and optimization module: Based on the HRSEI, evaluate the risk distribution of habitat quality, provide decision support for ecological protection, and contribute to the application and promotion in ecological management and land planning by governments and research institutions.

[0142] A storage medium according to the present invention stores a computer program thereon: when the computer program is executed by a processor, the above-mentioned method for extracting a remote sensing ecological index considering human activity intensity is implemented.

[0143] Therefore, the method, system and medium for extracting a remote sensing ecological index considering human activity intensity according to the present invention propose a method for calculating human activity intensity based on night lights, built-up land, population density and landscape development index, and couple it with an improved remote sensing ecological index to form a remote sensing ecological index considering human activity intensity, realizing a more comprehensive evaluation of the ecological environment quality, and can provide strong data support for ecological management and scientific decision-making in alpine mountainous areas.

[0144] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that: they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for extracting a remote sensing ecological index considering human activity intensity, characterized in that: It includes the following steps: Step S1: Based on the Google Earth Engine cloud platform, extract cloud, cloud shadow, and cirrus cloud information through operations to generate a cloud-free mask and remove pixels affected by atmospheric pollution; Crop the data according to the preset summer time window to lock the spatial range of the study area; Calculate the water body distribution using the improved Normalized Difference Water Index, and create a non-water body mask through the threshold method to exclude the influence of water areas on land ecological assessment; Integrate the preprocessed images of different summer years in time series to construct a long-time series spatio-temporal database of habitat quality; Step S2: Calculate four ecological parameters, namely the Normalized Vegetation Index, Humidity Index, Salinity Index, and Dryness Index, through the bands of the remote sensing dataset. Perform range normalization processing on each of them to eliminate the dimension difference. Extract the first principal component of the covariance matrix of the four indicators through principal component analysis to capture the maximum variation information of multi-dimensional ecological characteristics. Quadratically normalize the first principal component to generate a remote sensing ecological index in the range of 0-1; Step S3: Integrate four indicators, namely night-time light, percentage of built-up land, population density, and landscape development index. Use the Analytic Hierarchy Process to calculate the weights of the four indicators of human activity intensity, and generate a human activity intensity value in the range of 0-1 through weighted normalization; Step S4: Use the entropy method to obtain the weights of human activity intensity and the remote sensing ecological index, calculate the remote sensing ecological index model considering human activity intensity, and divide it into five grades: poor, bad, medium, good, and excellent according to the equal value method; Adopt the difference method to calculate the difference of the remote sensing ecological index considering human activity intensity in different time periods, and obtain the change map of the overall ecological environment quality grade in different periods of the study area.

2. The method for extracting the remote sensing ecological index considering the intensity of human activities according to claim 1, wherein: Step S1 includes the following steps: S101: Based on the Google Earth Engine cloud platform, realize the acquisition, processing, and analysis of remote sensing data by calling the API and functions of Google Earth Engine; S102: Cloud removal and image preprocessing Define the cloud mask extraction function bitwiseExtract: Parse the StateQA band to extract cloud status, cloud shadow, and cirrus cloud information; Create a cloud-free image function to only retain the pixel data of cloud-free, cloud-shadow-free, and cirrus-free pixels, and remove the influence of atmospheric pollution; S103: Set the study period at intervals of 3 years; Set the study season from July to September; Generate a DateRange list with each 3-year period as a time segment; S104: Calculate the improved Normalized Difference Water Index; Set the water body threshold: Use the Normalized Difference Water Index <0.2 to identify water bodies; Generate a water body mask to filter out the interference of water areas on the ecological index; S105: Image cropping and water body mask application Define the GetIMG image acquisition function: Filter the MODIS images and extract the data of the study area according to DateRange; Calculate the Normalized Difference Water Index and apply the water body mask: Use mndwicol.mean().clip(sa).lt(0.2) to generate the water body mask; Calculate the minimum-maximum value sts_minmax of the image: Calculate the minimum and maximum values of each image for normalization processing.

3. The method for extracting the remote sensing ecological index considering the intensity of human activities according to claim 1, characterized in that: Step S2 includes the following steps: S201: Perform exponential calculation and normalization calculation. The calculation formula is as follows: Salinity index S6 The calculation formula is as follows: ; In the formula, NIR : near-infrared band, Red : red light band, Green : green light band; Normalized Difference Vegetation Index NDVI The calculation formula is as follows: ; In the formula, NIR : near-infrared band, Red : red light band; Humidity index WET The calculation formula is as follows: ； wherein, Red : red light band, NIR : near-infrared band, Blue : blue light band, Green : green light band, MIR : mid-infrared band, SWIR1 : short-wave infrared 1, SWIR2 : short-wave infrared 2; Normalized Building Soil Index NDBSI The calculation formula is as follows: ； In the formula, BI : bare soil index, IBI : improved building index, Red : red light band, NIR : near-infrared band, Blue : blue light band, Green : green light band, MIR : mid-infrared band, SWIR1 : short-wave infrared 1, SWIR2 : short-wave infrared 2; S202: Perform range normalization on different indices X so that each index is between 0 and 1: ; S203: Perform principal component analysis, calculate the covariance matrix of the four standardized ecological parameters, and extract the first principal component PC1 , calculate the weight of the first principal component, and the calculation method is: set the input standardized data matrix as X , through principal component analysis, obtain the first principal component PC1 = XW1 , where W1 is the weight vector of the first principal component obtained by principal component analysis calculation; S204, perform Min-Max normalization on the first principal component PC1 to adjust its value to between 0 and 1 to obtain the remote sensing ecological index RSEI value, and the calculation formula is as follows: ; S205: Record the remote sensing ecological index values on a three-year cycle and create a line graph of the remote sensing ecological index values.

4. The method for extracting the remote sensing ecological index considering the intensity of human activities according to claim 1, characterized in that: Step S3 includes the following steps: S301: Obtain data on night lights, percentage of built-up land, population density, and landscape development index, so that the obtained data covers the study area; Perform projection transformation on data from different sources, unify them to the same geographic coordinate system or projection coordinate system, and perform spatial matching; Based on the study area boundary, crop all the data; S302: Normalize the extreme values of each indicator X so that each indicator is between 0 and 1: ; S303: To determine the index weights for evaluating the intensity of human activities, use the analytic hierarchy process to construct a hierarchical structure model; This hierarchical structure model includes three levels: the target level, the criterion level, and the index level. The target level is the intensity of human activities. The criterion level consists of four factors: night lights, population density, percentage of built-up land, and landscape development index. The index level is supported by night lights, percentage of built-up land, population density, and landscape development index as data sources; On this basis, construct a pairwise comparison judgment matrix based on literature analysis and score it according to the Saaty 1-9 scale principle; By calculating the eigenvector of the judgment matrix, obtain the normalized weights of each index to quantify the relative importance of each factor in the evaluation system; S304: Calculate the human activity intensity index by weighted calculation, and perform weighted summation on the four normalized indicators according to the weights: ； XNL : Nighttime lighting, XPD : Population density, XLU : Built-up land, XLDI : Landscape development intensity, H : Human activity intensity value within the range of 0 - 1.

5. The method for extracting a remote sensing ecological index considering human activity intensity according to claim 1, wherein: Step S4 includes the following steps: S401: Use the entropy method to evaluate the weights of human activity intensity and remote sensing ecological index, couple the human activity intensity value with the remote sensing ecological index, and construct a coupling model; S402: Use the equal value method to classify the results of the remote sensing ecological index considering human activity intensity into five grades: poor, bad, medium, good, and excellent; Through spatial statistical methods, calculate the area and proportion of ecological environment quality at different time periods and different grades; S403: Statistically calculate the average value and standard deviation of the remote sensing ecological index values considering human activity intensity; S404: Use the difference method to calculate the difference of the remote sensing ecological index considering human activity intensity in different time periods, and divide the remote sensing ecological index considering human activity intensity into five grades: significantly deteriorated, slightly deteriorated, unchanged, slightly improved, and significantly improved; S405: Statistically show that the grades of the remote sensing ecological index considering human activity intensity present a dynamic change process and perform effective visualization through a Sankey diagram.

6. A remote sensing ecological index extraction system considering human activity intensity, which applies a remote sensing ecological index extraction method according to any one of claims 1-5, is characterized in that: Include the following modules: Data acquisition and preprocessing module: Integrate data on landscape development index, night lights, population density, and built-up land, and after cloud removal, radiometric correction, atmospheric correction, and resampling processing, store them in the Google Earth Engine platform; Remote Sensing Ecological Index Calculation Module: Calculate the remote sensing ecological index on the Google Earth Engine platform, including the Normalized Difference Vegetation Index, humidity, dryness index, and introduce a salinity index optimization algorithm; Human Activity Intensity Calculation Module: Process night-time light data, combine the percentage of built-up land, landscape development index, and population density, and use a multi-factor coupling method to quantify human activity intensity; Remote Sensing Ecological Index Calculation and Analysis Module Considering Human Activity Intensity: Couple the improved remote sensing ecological index with human activity intensity, construct a remote sensing ecological index considering human activity intensity, and comprehensively evaluate the habitat quality; Spatio-Temporal Database and Visualization Module: Establish a spatio-temporal database of habitat quality, supporting query, time series analysis, and spatial distribution research; Application and Optimization Module: Evaluate the risk distribution of habitat quality based on the remote sensing ecological index considering human activity intensity, and provide decision-making support for ecological protection.

7. A storage medium, on which a computer program is stored, characterized in that: When the computer program is executed by a processor, it implements the method for extracting the remote sensing ecological index considering human activity intensity as described in any one of claims 1-5.

Images