Malaria high-risk area identification method based on Sentinel-1 time sequence

Abstract

The invention discloses a malaria high-risk area identification method and device based on a Sentinel-1 time sequence, and the method comprises the steps: obtaining a data set of a research area, carrying out the preprocessing of the data set, obtaining the high-risk time of anopheles breeding, analyzing the Sentinel-1 data through a Google Earth Engine GEE to determine an SWI index, detecting a water body of the research area on an SWI graph corresponding to the SWI index, and carrying out the detection of the malaria high-risk area. According to the method, high-risk time and a water body are classified by adopting majority voting fusion to obtain a time classification chart, high-risk places where anopheles breed are identified according to the time classification chart, the water body is subjected to covering evaluation of an SAM model to determine a malaria high-risk area, and medium-spatial-resolution Sentinel-1 images summarized monthly for a long time are utilized, and temperature and rainfall data are combined. And the process is limited in a high-risk month to reduce the process amount, and time or space prediction is performed on a habitat and a disease high-risk area by using GEE, so that the surface water body detection precision is improved.

Description

Technical Field

[0001] The invention belongs to the field of satellite remote sensing technology, and in particular relates to a method and a device for identifying high-risk malaria areas based on Sentinel-1 time series. Background Art

[0002] Malaria is a disease transmitted by female Anopheles mosquitoes that claims millions of lives every year. In 2022, there were approximately 249 million cases in 80 countries, resulting in 608,000 deaths, with an increase in cases reported in Iran over the past two years. Malaria transmission patterns vary by climate and biogeography, and although malaria has been successfully eradicated in many parts of the world, the risk of transmission remains in areas where Anopheles mosquitoes inhabit. Targeting Anopheles larvae has been shown to be effective in the fight against malaria. Surface water bodies such as lakes and ponds are the main places where malaria-carrying mosquitoes breed. By using maps of surface water bodies along with meteorological data such as precipitation and temperature, the presence of mosquito larvae can be predicted. Identifying the most productive water bodies can help improve targeted Anopheles larval control.

[0003] Traditionally, mapping surface waters involves field observations and recording, which is time-consuming and expensive in terms of human and budgetary resources, especially for large areas. Some researchers have successfully used drone technology to study Anopheles breeding sites. However, their resources and equipment are prohibitive for large-scale water mapping, so satellite imagery is a more practical method that is applicable to a wide range of research areas. Satellite observations have been used for habitat surveys for more than two decades, and mapping water bodies using satellite Earth observation technology is a routine practice, providing spatial information for monitoring and management of aquatic habitats. Advances in remote sensing technology allow for rapid extraction of surface water information from datasets. Optical imagery, while useful, can be hampered by fog, clouds, and inclement weather, limiting data collection to daytime hours. In addition, due to their temporary nature, they are difficult to accurately locate surface water bodies and Anopheles habitats. In contrast, remote sensing methods that use microwave bands, such as synthetic aperture radar (SAR), provide uninterrupted data collection in any weather conditions. For example, Sentinel-1 provides all-weather data every 6 days, making it a valuable resource. Unlike other land covers, water bodies exhibit low backscatter response, making them easier to detect using active sensors. Therefore, there is an urgent need to provide a method to improve the accuracy of detecting surface water and mitigate the risk of malaria occurrence or transmission worldwide. Summary of the invention

[0004] In view of this, the present invention provides a method and device for identifying high-risk areas for malaria based on Sentinel-1 time series, which uses GEE to perform temporal or spatial prediction of habitats and disease high-risk areas, reduce the amount of data and maintain high accuracy, so as to solve the above-mentioned technical problems, and specifically adopts the following technical solutions to achieve this.

[0005] In a first aspect, the present invention provides a method for identifying high-risk areas for malaria based on Sentinel-1 time series, comprising the following steps:

[0006] Acquire a data set of the study area, and pre-process the data set to obtain a high-risk time for Anopheles reproduction, wherein the data set includes Sentinel-1 data, precipitation data, and temperature data;

[0007] Using Google Earth Engine GEE to analyze the Sentinel-1 data to determine the SWI index corresponding to the Sentinel-1 data, and detecting water bodies in the study area on a SWI map corresponding to the SWI index;

[0008] The high-risk time and the water body are classified by majority voting fusion to obtain a time classification map, and the high-risk location for Anopheles reproduction is identified according to the time classification map;

[0009] The water bodies are subjected to a SAM model masking assessment to determine the high-risk areas for malaria corresponding to the high-risk locations.

[0010] As a preferred embodiment of the above technical solution, preprocessing the data set to obtain the high-risk time for Anopheles reproduction includes:

[0011] Select a seven-year time series and use effective environmental parameters to predict the time of malaria outbreak, wherein the seven-year time series includes climate conditions, seasonal patterns and abnormal weather events, and the environmental parameters include precipitation and temperature values;

[0012] Arrange the temperature values ​​from low to high to determine the lower limit LDT and upper limit UDT of the development temperature of Anopheles;

[0013] The rainy season temperature suitable for the growth and stock of Anopheles mosquitoes was extracted based on the seven-year time series, precipitation and temperature values, and the rainy season temperature and the seven-year time series were used in a water extraction model.

[0014] As a preferred embodiment of the above technical solution, the seven-year time series is time aggregated to obtain a Sentinel-1 image, wherein a median operator of the time aggregation is used to obtain the Sentinel-1 image through GEE.

[0015] As a preferred embodiment of the above technical solution, the SWI index is calculated based on the Sentinel-1 image, and after extracting water bodies every month for seven years, the map corresponding to the Sentinel-1 image is fused to identify high-risk areas.

[0016] As a preferred embodiment of the above technical solution, the high-risk time and the water body are classified by majority voting fusion to obtain a time classification map, including:

[0017] Detect and classify the high-risk areas, including:

[0018] The first annual analysis identifies areas that have water four months of the year to determine where water is stored continuously year-round; the second monthly analysis identifies areas that have water stored over the years;

[0019] A four-level regional classification map was used to designate regions with water storage every seven years as high-risk, regions with water storage between four and six years as relatively risky, and regions with water storage less than four years as low-risk, and a fourth regional classification map was assigned to regions with no water in any of the study months;

[0020] The high-risk areas are detected using an abstract hierarchy algorithm and majority voting fusion, wherein the majority voting fusion includes each regional classification map voting on a specific category, all votes have the same weight, and majority voting is used for monthly and annual analysis.

[0021] As a preferred embodiment of the above technical solution, a SAM model masking assessment is performed on the water body to determine the high-risk area for malaria corresponding to the high-risk location, including:

[0022] Using multi-point feature hints for water body segmentation, the points in the geographic data frame are converted into a list of coordinates [x, y];

[0023] Use the SamGeo tool as a predictor for a single coordinate, predict the features of a single coordinate in the instance to get the mask, and save the obtained mask to a unique file for each point;

[0024] After generating a mask for each point, all masks are merged into a single mosaic raster file, providing a complete representation of all segmented regions from the multi-point feature cues.

[0025] As a preferred embodiment of the above technical solution, a geo-referenced image is created by downloading tiles and using a tile map service, a satellite image is selected to segment the geo-referenced image, and a path is set to define a SAM model;

[0026] Loop through the coordinate point list to predict, segment and save each point into a unique Geotiff file;

[0027] All masks were resampled to 10m resolution and merged into one file as a mosaic mask containing information of all segmented regions;

[0028] Masks of multiple depressions generated by SAM were used to assess the accuracy of quantities extracted from the SWI index.

[0029] As a preferred embodiment of the above technical solution, a confusion matrix and comprehensive indicators are used to evaluate the performance of the SAM model, and the comprehensive indicators include precision, recall rate, F1 score, overall accuracy and kappa coefficient.

[0030] As a preferred embodiment of the above technical solution, after calculating the surface water distribution map corresponding to the geographic reference image, the flooding frequency IF of the depression is calculated using expression (1):

[0031]

[0032] Among them, nundation Frequency represents the flooding frequency IF of the depression, Water Observation Times represents the water observation time, and Total Observation Times represents the total observation time.

[0033] In a second aspect, the present invention further provides a device for identifying high-risk areas for malaria based on a Sentinel-1 time series, which is applied to the above-mentioned method for identifying high-risk areas for malaria based on a Sentinel-1 time series, comprising:

[0034] A data acquisition unit, used to acquire a data set in a research area, and pre-process the data set to obtain a high-risk time for Anopheles reproduction, wherein the data set includes Sentinel-1 data, precipitation data, and temperature data;

[0035] a water body detection unit, configured to analyze the Sentinel-1 data using Google Earth Engine GEE to determine a SWI index corresponding to the Sentinel-1 data, and detect water bodies in the study area on a SWI map corresponding to the SWI index;

[0036] A classification and identification unit, configured to classify the high-risk time and the water body by majority voting fusion to obtain a time classification map, and identify the high-risk location for Anopheles breeding according to the time classification map;

[0037] The model evaluation unit is used to perform a SAM model masking evaluation on the water body to determine the high-risk area for malaria corresponding to the high-risk location.

[0038] The present invention provides a method and device for identifying high-risk areas for malaria based on Sentinel-1 time series, by acquiring a data set of a study area, pre-processing the data set to obtain a high-risk time for Anopheles reproduction, using Google Earth Engine GEE to analyze the Sentinel-1 data to determine the SWI index corresponding to the Sentinel-1 data, and detecting the water body in the study area on the SWI map corresponding to the SWI index, using majority voting fusion to classify the high-risk time and the water body to obtain a time classification map, and identifying the high-risk location for Anopheles reproduction according to the time classification map, performing a masking assessment of the SAM model on the water body to determine the high-risk area for malaria corresponding to the high-risk location, using a medium spatial resolution Sentinel-1 image aggregated monthly over a long period of time, and combining temperature and precipitation data from a meteorological data set. Using precipitation and temperature parameters and limiting the process to high-risk months can reduce the process volume. By adopting the majority voting algorithm for monthly, annual and overall analysis, temporary water bodies and small depressions with high risk of Anopheles larval habitat formation were effectively detected, and the detection accuracy of surface water bodies was improved by using GEE to make temporal or spatial predictions of habitats and high-risk areas for diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 A flow chart of the method for identifying high-risk areas for malaria based on Sentinel-1 time series provided by the present invention;

[0041] Figure 2 A schematic diagram of the research area provided for the present invention;

[0042] Figure 3 A schematic diagram of the workflow of the method for identifying high-risk areas for malaria based on Sentinel-1 time series provided by the present invention;

[0043] Figure 4 A schematic diagram of precipitation and temperature for a seven-year time series provided by the present invention;

[0044] Figure 5 A surface water map drawn using the SWI method provided by the present invention;

[0045] Figure 6A schematic diagram of monthly analysis of high-risk area detection formed by Anopheles larvae habitats provided by the present invention;

[0046] Figure 7 A schematic diagram of annual analysis of high-risk area detection formed by Anopheles larvae habitats provided by the present invention;

[0047] Figure 8 A map of surface water bodies for the overall analysis provided by the present invention;

[0048] Fig. 9 The images of three samples are processed for the multi-point feature prompt provided by the present invention;

[0049] Fig.10 A schematic diagram of a generated mask and a corresponding detected water body provided by the present invention;

[0050] Fig.11 The flooding frequency IF distribution diagram of the depression provided by the present invention;

[0051] Fig.12 A ground profile of a high-risk area for malaria provided by the present invention;

[0052] Fig.13 This is a structural block diagram of a malaria high-risk area identification device based on Sentinel-1 time series provided by the present invention. DETAILED DESCRIPTION

[0053] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0054] See also Figure 1 The present invention provides a method for identifying high-risk areas for malaria based on Sentinel-1 time series, comprising the following steps:

[0055] S1: Obtain a data set in the study area, and pre-process the data set to obtain the high-risk time for Anopheles reproduction, wherein the data set includes Sentinel-1 data, precipitation data, and temperature data;

[0056] S2: using Google Earth Engine GEE to analyze the Sentinel-1 data to determine the SWI index corresponding to the Sentinel-1 data, and detecting the water body in the study area on the SWI map corresponding to the SWI index;

[0057] S3: using majority voting fusion to classify the high-risk time and the water body to obtain a time classification map, and identifying the high-risk location for Anopheles breeding according to the time classification map;

[0058] S4: Conducting a SAM model masking assessment on the water body to determine the high-risk malaria area corresponding to the high-risk location.

[0059] In this embodiment, the data set is preprocessed to obtain the high-risk time for Anopheles reproduction, including: selecting a seven-year time series and using effective environmental parameters to predict the time of malaria outbreak, wherein the seven-year time series includes climate conditions, seasonal patterns and abnormal weather events, and the environmental parameters include precipitation and temperature values; arranging the temperature values ​​from low to high to determine the lower limit LDT and upper limit UDT of the development temperature of Anopheles; extracting the rainy season temperature suitable for the growth and inventory of Anopheles based on the seven-year time series, precipitation and temperature values, and using the rainy season temperature and the seven-year time series for a water extraction model. The seven-year time series is temporally aggregated to obtain a Sentinel-1 image, wherein the median operator of the temporal aggregation is used to obtain the Sentinel-1 image through GEE. The SWI index is calculated based on the Sentinel-1 image, and after extracting water bodies every month within seven years, the map corresponding to the Sentinel-1 image is fused to identify high-risk areas.

[0060] See also Figure 2 In Iran, most malaria cases occur in the southeastern corner of the country, in the southern part of Sistan and Baluchestan, Hormozgan and Kerman provinces. Sistan and Baluchestan Province of Iran covers an area of ​​180,726 square kilometers and has a population of 2.78 million. It is the second largest province in the country. Its center is located at 29°28′39″N and 60°52′12″S. It borders Afghanistan to the northeast, Pakistan to the east, and the coast of the Sea of ​​Oman to the south. Climate classification, the area belongs to the hot desert climate category (BWh). Although the annual rainfall is about 110 mm, the area is prone to heavy rains. An unexpected malaria outbreak in 2021 continued into 2022 with ten times the number of cases compared to 2021, with a total of 3,000 cases estimated by November 2022. Factors contributing to the outbreak include the presence of foreign citizens, heavy rains, and insufficient case detection. The datasets used include Sentinel-1 imagery, PERSIANN-CDR precipitation data, and temperatures extracted from meteorological datasets. Launched by the European Space Agency (ESA) in 2014 and 2016, Sentinel-1 is a polar-orbiting constellation providing dual-polarization data in the C-band, which provides a revisit time of six days via two satellites. 79 Sentinel-1 frames are provided monthly for Sistan and Baluchestan, and after seven years of analysis, this number has increased to 2,212 in four months, making GEE the best choice to handle the processing load for such a large area. The orthogonal corrected backscatter coefficient products of the individual VV and VH channels of Sentinel-1 at 10 m spatial resolution in ascending and descending orbits were utilized. The only remaining preprocessing was the removal of speckle noise using a median filter with a kernel size of 3 × 3. (a) Location of Sistan and Baluchestan, Iran; (b) Example of a depression that is not fully filled with water in the south of Saravan city; (c) Depression in the north of the study area: The left image shows a view of the site during the dry season. On the right, the image shows the depression filled with water during the rainy season.

[0061] Among them, PERSIANN-CDR provides a 40-year precipitation dataset, providing daily near-global data at a resolution of 0.25°, making it valuable for global climate research, especially extreme weather events. Unlike PERSIANN technology, PERSIAN-CDR uses an ANN model trained with preset parameters to estimate historical data. Seven years of meteorological data time series obtained from the Iranian Meteorological Organization (IMO) have been used to consider temperature factors. These data include summary data of daily average temperature measured at weather stations in Sistan and Baluchestan provinces, and then the monthly average temperature is calculated based on the average temperature of each day.

[0062] Specifically, water bodies were mapped over seven years in Sistan and Baluchestan, Iran, to identify high-risk areas for Anopheles breeding, which involved (1) using PERSIAN-CDR and meteorological datasets to identify high-risk months based on precipitation and temperature; (2) acquiring Sentinel-1 data, calculating the SWI index, and detecting water bodies using thresholds on the SWI maps; (3) fusing the seven-year monthly classification maps to identify high-risk sites by majority voting; and (4) evaluating the approach against water suppression masks generated by the SAM model. To predict potential breeding sites, the study period was first extracted, focusing on the high-risk period of the year. The seven-year time series was chosen to balance capturing long-term environmental changes, and seven years was long enough to include different climate conditions, seasonal patterns, and unusual weather events, thereby enhancing the robustness and generalizability of the model. This duration helped identify consistent trends and patterns in parameters that are critical for accurately detecting high-risk Anopheles habitats. Conversely, using shorter time series may not capture the full extent of climate change, resulting in less reliable predictions and potentially overlooking long-term trends. In summary, a seven-year period provides a practical compromise, providing sufficient data to build a reliable model for effective Anopheles habitat detection.

[0063] Specifically, the use of effective environmental parameters can help accurately predict the timing of malaria outbreaks. Precipitation and temperature are key factors that increase the number of Anopheles mosquitoes, affect larval and mosquito populations, and the risk of malaria transmission. Therefore, the monthly PERSIANN-CDR dataset from 2017 to 2023 is used to determine the rainy season for analysis. Temperature is another key environmental factor. From the lowest to the highest, these temperature limits are called the lower and upper developmental temperatures (LDT and UDT). Anopheles eggs require temperatures between 15°C and 35°C and an average of 611 degrees for days to mature into adults. This means that the temperature must be kept within the appropriate range, and the cumulative temperature must reach 611 degrees for several consecutive days. Higher temperatures accelerate the growth rate of Anopheles larvae and shorten their maturation time. To calculate this, the average daily temperature of the seven IMO stations in the study area was used to determine the monthly average temperature. Finally, by summarizing the precipitation and temperature monitoring results from 2017 to 2023, the rainy season with temperatures suitable for the growth and survival of Anopheles mosquitoes, such as 15-35°C, was extracted, and the images corresponding to this period were used in the water extraction model.

[0064] After the study period was determined, it was necessary to acquire Sentinel-1 imagery. Using multi-temporal data allowed for an increased ability to detect temporary water bodies and reduced speckle noise, implementing a temporal aggregation method instead of using time series imagery directly. Therefore, during the study period, the median operator of this method was used to acquire Sentinel-1 imagery through GEE. Areas with the potential for water accumulation during high-risk periods, exceeding a threshold of 611 degrees, can serve as habitat for Anopheles larvae. After calculating the SWI index and extracting water bodies monthly over seven years (2017-2023), the maps were fused at the decision level to identify high-risk areas that provide the necessary surface water for the growth and development of Anopheles. Three different analyses were selected to detect and classify high-risk areas, the first annual analysis identified areas with water in each of the four study months to identify locations with continuous water accumulation throughout the year. The second monthly analysis identified areas with water accumulation over multiple years. Using a four-level classification map, areas with water impoundment every seven years were designated as high-risk, areas with water impoundment between four and six years were marked as relatively dangerous, and areas with water impoundment less than four years were marked as low-risk. The fourth category was assigned to areas with no water in any of the study months. This analysis aimed to identify high-risk areas with continuous water impoundment over the years, indicating its long-term durability, and a third analysis used all the generated maps to map high-risk water bodies within the study area.

[0065] in, Figure 3where est sample collection means test sample collection; study area extent means study area extent; 260water depression samples means 260 depression samples; 260Non-Water depression samples means 260 non-water depression samples; Field observation means field observation; Geo-tagged Field Photo means field photo with geotags; Waterdepressions masking means water depression masking; satellite image acquisition means satellite image acquisition; Water depressions mask generation means water depression mask generation; Tile map service (TMS) framework map service (TMS); segmentanything model (SAM) means segmentation any model (SAM); Pointprompt means point prompt; waterdepression detection means water level drop detection; Meteorological temperature dataset means meteorological temperature dataset; speckle filtering means speckle filtering; SWI index calculation means SWI index calculation; Waterbody maps means waterbody maps; majority voting means majority voting; Monthlyanalysis means monthly analysis; yearly analysis means annual analysis; Extracting high-riskmonths means extracting high-risk months; sentinel-1 time-series monthly images sentinel-1 time-series monthly images; Fusion at decision level means decision-level fusion; weihted majority voting means majority voting; Overall analysis means overall analysis; determining the high risk breeding sites of anopheles mosquitoes means determining the high-risk breeding sites of anopheles mosquitoes; Accuracy assessment means accuracy assessment; visual assessment means visual assessment; Classification measures means classification measures; statistical comparison means statistical comparison;

[0066] It should be understood that by obtaining a data set of the study area and preprocessing the data set to obtain the high-risk time for Anopheles breeding, using Google Earth Engine GEE to analyze the Sentinel-1 data to determine the SWI index corresponding to the Sentinel-1 data, and detecting the water body in the study area on the SWI map corresponding to the SWI index, the high-risk time and the water body are classified and processed by majority voting fusion to obtain a time classification map, and the high-risk sites for Anopheles breeding are identified according to the time classification map, and the water body is subjected to a masking assessment of the SAM model to determine the high-risk area for malaria corresponding to the high-risk site, and the medium spatial resolution Sentinel-1 images aggregated over a long period of time are utilized, and the temperature and precipitation data from the meteorological data set are combined. The use of precipitation and temperature parameters and limiting the process to high-risk months can reduce the process volume. By using the majority voting algorithm for monthly, annual and overall analysis, temporary water bodies and small depressions with high risk of Anopheles larval habitat formation are effectively detected, and GEE is used to perform temporal or spatial predictions of habitats and high-risk areas for diseases, thereby improving the detection accuracy of surface water bodies.

[0067] Optionally, the high-risk time and the water body are classified by majority voting fusion to obtain a time classification map, including:

[0068] Detect and classify the high-risk areas, including:

[0069] The first annual analysis identifies areas that have water four months of the year to determine where water is stored continuously year-round; the second monthly analysis identifies areas that have water stored over the years;

[0070] A four-level regional classification map was used to designate regions with water storage every seven years as high-risk, regions with water storage between four and six years as relatively risky, and regions with water storage less than four years as low-risk, and a fourth regional classification map was assigned to regions with no water in any of the study months;

[0071] The high-risk areas are detected using an abstract hierarchy algorithm and majority voting fusion, wherein the majority voting fusion includes each regional classification map voting on a specific category, all votes have the same weight, and majority voting is used for monthly and annual analysis.

[0072] In this embodiment, an abstract hierarchy algorithm is used to detect high-risk areas using majority voting and weighted majority voting. The most commonly used method is majority voting, in which each classification map votes for a specific category and all votes have the same weight. The final label is the most common label in the classifier output, and majority voting is used for monthly and annual analysis. However, weighted majority voting is affected by the number of votes and the assigned weights. In weighted majority voting, each classification map is assigned a weight based on its perceived reliability or performance, and the final decision is determined by majority votes, and the votes of each map are weighted according to their assigned weights. Votes are weighed by using the precipitation of each month to determine weights. Precipitation directly indicates potential waterlogging. By assigning weights based on precipitation levels, periods with higher rainfall are more important, which is more likely to cause waterlogging. This approach enables the model to adapt to the temporal changes in precipitation, making it more adaptable to real environmental conditions. Focusing on high precipitation periods helps filter out the noise of low or no precipitation months and reduce false positives. Therefore, the precipitation level, which reflects sensitivity and specificity, is a reliable indicator of algorithm performance, enhancing the impact of decisions with high sensitivity and specificity. On the other hand, majority voting may bias detection towards wet seasons, potentially underestimating the risk of dry spells, as small-scale ponding remains problematic. Abnormal weather patterns, such as unseasonal rainfall, may also have a disproportionate impact on the model's output if not managed properly. Furthermore, the effectiveness of the weighted voting system depends on the accuracy of precipitation data, as inaccurate data may lead to incorrect weights and flawed predictions. These contributions are calculated by dividing monthly precipitation by the sum of the corresponding monthly precipitation.

[0073] Optionally, performing a SAM model masking assessment on the water body to determine the high-risk area for malaria corresponding to the high-risk location comprises:

[0074] Using multi-point feature hints for water body segmentation, the points in the geographic data frame are converted into a list of coordinates [x, y];

[0075] Use the SamGeo tool as a predictor for a single coordinate, predict the features of a single coordinate in the instance to get the mask, and save the obtained mask to a unique file for each point;

[0076] After generating a mask for each point, all masks are merged into a single mosaic raster file, providing a complete representation of all segmented regions from the multi-point feature cues.

[0077] In this embodiment, a georeferenced image is created by downloading tiles and using a tile map service, a satellite image is selected to segment the georeferenced image, and a SAM model is defined by setting a path; a list of coordinate points is looped to predict, segment, and save each point into a unique Geotiff file; all masks are resampled to 10m resolution and merged into one file as a mosaic mask containing information about all segmented regions; and the masks of multiple depressions generated by SAM are used to evaluate the accuracy of the quantities extracted from the SWI index. The performance of the SAM model is evaluated using a confusion matrix and comprehensive indicators, including precision, recall, F1 score, overall accuracy, and kappa coefficient.

[0078] It should be noted that after calculating the surface water distribution map corresponding to the geographic reference image, the flooding frequency IF of the depression is calculated using expression (1):

[0079]

[0080] Among them, nundation Frequency represents the flooding frequency IF of the depression, Water Observation Times represents the water observation time, and Total Observation Times represents the total observation time.

[0081] The geospatial analysis is performed with the help of a custom tool “SamGeo”, which is an integral part of the original module. SAM has several models available that have different computational requirements and are different in the underlying architecture. In this example, the ViT-H SAM model is used, which is the most advanced and complex model currently available. SAM provides various hints to perform water body segmentation using the multi-point feature hint. This hint converts the points in the geodata frame into a list of coordinates [x, y]. SamGeo acts as a predictor for the individual coordinates, predicting their features in the instance and saving the resulting masks into unique files for each point. After generating the mask for each point, all masks are merged into a mosaic raster file, providing a complete representation of all segmented areas from the multi-point hint.

[0082] The centroids of the 260 water bodies were initially located as points in Google Earth Pro and then exported as shapefiles to mask the water bodies. These points were chosen based on visual interpretation because during the dry season these depressions exhibit unique characteristics due to the structure of the silty clay layer caused by soil erosion caused by runoff, as observed during the field assessment (e.g. Figure 2c). On the other hand, these places are prone to water accumulation during the rainy season. Of the 260 depressions, 31 were observed in the field. The satellite image was selected for segmentation by downloading tiles and creating a large georeferenced image using a tile map service (TMS2GeoTIFF). The model of SAM was then defined by setting the path. SamGeo was initialized with the specified parameters and the shapefile was converted from a geodataframe to a list of point coordinates. In the prediction method, the coordinate reference system was defined and set to “EPSG:4326”. Subsequently, by looping through the list of coordinate points, each point was predicted, segmented and saved into a unique Geotiff file. This process generated a unique mask for each point. Finally, all masks were resampled to 10m resolution and merged into a single file as a mosaic mask containing information on all segmented areas. All parameters defined in the SAM model are listed in Table 1.

[0083] Table 1 SAM model parameters

[0084] Parameter Value Model_type ViT_H checkpoint sam_vit_h_4b8939.pth Automatic False sam_kwargs None

[0085] Among them, Parameter represents parameter, Value represents value, Model_type represents model; checkpoint represents checkpoint, Automatic represents automatic, and all SAM-related processes are completed using the code available on GitHub: AI Remote Sensing Repository, which has been modified according to the case. The masks of 260 depressions generated by SAM are used to evaluate the accuracy of water volume extracted from the SWI index. In addition, 260 samples representing non-water areas are randomly selected in Google Earth Pro. To evaluate the performance of the model, confusion matrices (as shown in Table 2) and comprehensive indicators such as precision, recall, F1 score, overall accuracy (OA) and Kappa coefficient are used. Pixel-level accuracy criteria, especially the union intersection (IoU) commonly used in object detection, are considered for depression detection verification. Among them, Actual represents actual and Predicted represents prediction.

[0086] Table 2 Confusion matrix for binary classification

[0087]

[0088] In another assessment, in order to better understand and appreciate the spatiotemporal distribution characteristics of water bodies, after calculating the surface water distribution maps of all 28 images, the inundation frequency (IF) of depressions will be calculated using expression (1). The IF also reflects the topography, and lower areas are more prone to flooding.

[0089] Specifically, to evaluate the performance of the proposed method, the results were compared with four classification algorithms: Random Forest (RF), Support Vector Machine (SVM), Smiling Gradient Tree Boosting (SGTB) and Smiling Car (SC). To this end, 4 land cover categories of vegetation, urban areas, wasteland and water bodies were classified by using each of the above-mentioned machine learning models, and the training samples associated with the water body category were those located in Google Earth Pro. Four high-risk periods were identified by combining the average monthly precipitation and temperature data. During these months, the areas with the potential for water accumulation were identified using the SWI index on Sentinel-1 images. The overall analysis was evaluated using an object-based approach using masks generated by the SAM model, and the proposed method outperformed other classifiers, including machine learning models, in terms of accuracy.

[0090] See also Fig.13 The present invention also provides a device for identifying high-risk areas for malaria based on Sentinel-1 time series, which is applied to the above-mentioned method for identifying high-risk areas for malaria based on Sentinel-1 time series, comprising:

[0091] A data acquisition unit, used to acquire a data set in a research area, and pre-process the data set to obtain a high-risk time for Anopheles reproduction, wherein the data set includes Sentinel-1 data, precipitation data, and temperature data;

[0092] a water body detection unit, configured to analyze the Sentinel-1 data using Google Earth Engine GEE to determine a SWI index corresponding to the Sentinel-1 data, and detect water bodies in the study area on a SWI map corresponding to the SWI index;

[0093] A classification and identification unit, configured to classify the high-risk time and the water body by majority voting fusion to obtain a time classification map, and identify the high-risk location for Anopheles breeding according to the time classification map;

[0094] The model evaluation unit is used to perform a SAM model masking evaluation on the water body to determine the high-risk area for malaria corresponding to the high-risk location.

[0095] In this embodiment, Figure 4 (a) and Figure 4 (b) shows the monthly average precipitation and monthly minimum, average and maximum temperatures of the seven weather stations. Figure 4 (c) in Figure 2 shows the integration of temperature and precipitation results to extract the study period. Sentinel-1 imagery for the selected study period from 2017 to 2023, Figure 5The final classification results at a spatial resolution of 10 meters for a small area in Saravan during the high-risk months of 2017 are shown. During the high-risk periods of the year, depressions with the possibility of accumulating water pass the threshold of 611 degrees and can become habitats for Anopheles larvae. Three different analyses were performed to detect high-risk areas for the formation of larval habitats. Monthly analysis: The analysis was performed to identify areas with water in four months from 2017 to 2023. Therefore, it was necessary to extract the areas that were usually identified as water in the classification output maps for the months of March, April, May and July. For this purpose, the logical operator of intersection (AND operator) was used to perform this analysis, and the results of the analysis for a small area of ​​the city of Saravan are shown below. Figure 6 Annual analysis: The areas with water were classified according to their frequency during the study period. The classification results for some areas of Saravan city are shown in Figure 6 as shown. The process of this classification is that areas that had water in all seven years of the study are classified as risk areas, if they had water in four to six years out of seven years, they are classified as medium risk areas, and areas that had water for less than four years are classified as low risk areas. The gray area refers to the part that has no water in any of the monthly classification maps. Overall analysis: In addition to the monthly and annual analysis, the third analysis also took into account all 28 classification maps for four months in the seven-year study. For this, a weighted majority vote was used, with weights calculated based on precipitation, and the output maps of this analysis were used to evaluate the accuracy. The results of the analysis for a small part of the Saravan municipality are shown below. Figure 8 The blue areas indicate that there is water in the depressions, which is based on the overall analysis of the 28-month research results. Figure 4 (a) Time series of monthly average precipitation from 2017 to 2023; (b) Time series of monthly average minimum, average and maximum temperatures from 2017 to 2023; (c) Time prediction of the increase in the number of Anopheles mosquitoes in seven years based on precipitation and temperature parameters. The blue and red cells represent the precipitation and temperature suitable for the breeding of Anopheles mosquito larvae in the province.

[0096] It should be noted that 260 shelters were generated using a point-based approach by performing SAM on marked points representing high potential areas of Anopheles breeding habitat. By focusing on a singular point, SAM was able to provide accurate segmentation results. These points served as precise spatial cues sufficient for the model to accurately identify and segment the depressions. This approach may be particularly relevant to applications that require accurate identification and segmentation of individual objects in an image. Fig. 9The results of masked depressions for some samples in the northern part of Saravan city are shown. After generating the masks, they were resampled to 10 meters, which is the same as the mapping results of Sentinel-1 images. In order to accurately verify the proposed method and make meaningful and reliable quantitative analysis of the final map, the masks generated by the SAM model and the sampling points in the non-water body area are superimposed on the output map of the weighted majority voting method. The samples of depression detected are shown in Figure 2. Fig.10 As shown, Fig.10 The masks generated in Figure 3 and the corresponding detected water bodies: (a) Google Earth Pro image used to create the mask; (b) SAM masked depression; (c) water body detection using the proposed method. Then, the confusion matrix is ​​obtained by comparing the prediction results with the reference mask, and six indicators are calculated based on the confusion matrix. The results are shown in Table 3. Among them, Evaluationmeasure represents the evaluation measure and Proposed method represents the proposed method.

[0097] Table 3 Detection accuracy of the proposed method

[0098] Evaluation measure Proposed method Precision 98.7％ Recall 93.4％ F-1score 96％ OA 96.1％ Kappa 92.2％ IoU 34.1％

[0099] Fig.11 The intensity-frequency (IF) distribution map corresponding to the monthly SAR data of a small area in the study area is shown. The water body detection performance of the proposed method is compared with that of other four classification algorithms RF, SVM, SGTB and SC by implementing land cover classification. The classification results are evaluated using 520 collected samples. Table 4 summarizes the detection results of all classifiers. Wherein, Evaluation measure represents the evaluation measure.

[0100] Table 4 Detection accuracy of classification algorithm

[0101]

[0102] By comparing the results shown in Table 3 and Table 4, the Kappa of the proposed method is 17% higher on average compared with RF, SVM, GTB and SC.

[0103] like Figure 4As shown in a, January, March, April, May, July and October had rainfall in at least four out of seven years, which are significantly wetter than other months. Although some months have rainfall for several consecutive years, these months have the highest rainfall. As mentioned earlier, temperatures between 15°C and 35°C are most suitable for the growth of Anopheles larvae, and most Anopheles species cannot complete their life cycle and survive below 15°C, while the growth from the immature stage to the adult stage stops at temperatures above 35°C. High temperatures can also reduce the stability of pits and depressions after heavy rains, leading to surface evaporation and drying of the soil, which is not conducive to the growth of Anopheles larvae. According to Figure 4 b. Average temperatures from March to November were within the ideal range in all seven years.

[0104] To select the study period, the trends of monthly contributions to annual precipitation and temperature were analyzed. Figure 4 c shows that in most years, March, April, May and July are favorable for the formation of Anopheles larval habitats in terms of precipitation and temperature. Under suitable precipitation and temperature conditions, aquatic habitats become the main place for the growth and development of Anopheles larvae after reaching 611 degree days, eventually leading to the disease peak. In addition, when it was noted that there were two separate epidemic peaks in the study area, the first peak occurred in late winter to late spring, and the second peak occurred in midsummer to mid-autumn, which coincided with the selected months. Therefore, the study period was limited to four months (March, April, May and July) from 2017 to 2023, which are necessary conditions for the formation of larval habitats. When the maximum temperature reaches 35°C or higher, the growth and larvalization process stops.

[0105] Figure 5 Visually, it was confirmed that the method effectively delineated water bodies, highlighting depressions with high water retention potential. The ability to identify these areas depends on factors such as the water storage capacity of the depression, the amount of water, the size of the depression, the slope of the area, and the connectivity of the waterways. Due to its fine spatial resolution, the index also displays detailed information. Comparing the water body maps extracted in different months, it can be found that changes in surface water cover are affected by precipitation, evaporation, and infiltration. Among them, a small part of the surface water body map in the southeast of Saravan city drawn using the SWI method during the study period in 2017 (blue): (a) March 2017; (b) April 2017; (c) May 2017; and (d) July 2017,

[0106] Figure 6Shown are depressions that retained water in all four study months of the year, with blue areas indicating high-risk areas for that year. Larger and deeper depressions retained water in successive months and years of the study, which could be due to runoff filling the depressions after rainfall or ponding of water from previous rainfalls. The lifespan of water in a depression depends on factors such as depression area and volume, evaporation rate, and infiltration rate. Longer water retention provides suitable conditions for the formation and growth of Anopheles mosquito habitats, and after reaching the 611 degree threshold, the comparison Figure 6 Maps from different years show that smaller depressions have greater variations, probably due to different rainfall amounts and the subsequent flow and accumulation of water in surrounding depressions. Therefore, these specific areas have the highest risk for Anopheles habitat formation in the annual survey. Among them, high-risk areas for Anopheles larval habitat formation were detected in the monthly analysis. Areas depicted in blue are common areas that had water in all four months of the year studied. High-risk areas in a small area in the southeast of Saravan city: (a) 2017; (b) 2018; (c) 2019; (d) 2020; (e) 2021; (f) 2022; (g) 2023.

[0107] Figure 7 It shows that small depressions (shown in blue) have less frequent water storage (less than 4 years), while larger depressions are more frequently observed as water bodies. The frequent presence of water in larger depressions indicates a high risk because they have a greater water retention capacity and can fill up with the first rainfall in the risk month. In addition, red areas, representing high-risk areas, appear in deeper boreholes where runoff is directed when it rains. These deep areas are the first to fill up even when rainfall is low, keeping them consistently wet and conducive to the growth of Anopheles mosquitoes. Surrounding the red areas (high risk level) are green areas (medium risk level), representing shallower parts of the depressions that can store water for 4 to 6 years. These areas may also be covered with water due to relatively high rainfall, and these findings can help health officials prioritize high-risk areas for monitoring and control. Among them, the high-risk areas for the formation of Anopheles mosquito larval habitats are detected in annual analysis. Red areas are high-risk areas with 7 years of water storage, green areas are medium-risk areas with 4 to 6 years of water storage, and blue areas are areas with less than 4 years of water storage. Gray indicates areas where habitat formation is unlikely. Risk classification map for a small area in the southeast of Saravan city: (a) March; (b) April; (c) May; (d) July.

[0108] Figure 8The consistent detection of larger, high-risk depressions as water bodies is highlighted, demonstrating the effectiveness of the approach in identifying important water features associated with potential risk. However, a refinement was observed in this analysis where smaller depressions (usually caused by low precipitation and widespread across the region) were systematically excluded and not classified as water bodies. This subtle improvement improves the accuracy of water detection by filtering out insignificant depressions, resulting in a more accurate representation of water bodies in the study area. The maps analyzed as a whole will be used for accuracy assessment. Image illustrations of three samples in the northern part of Saravan city processed using point hinting. The first column shows a Google Earth image with the selected point as the centroid of the depression. The second column shows the prediction results of SAM.

[0109] See also Fig.11 , the spatial distribution of IF highlights the water storage capacity of depressions based on the time and location of water presence in the hydrological system. The results of this index confirm the analysis we conducted previously. Areas with IF values ​​close to 100 (shown in red) indicate consistent water cover throughout the study period and are highly suitable habitats for Anopheles larvae due to their water retention capacity. In contrast, non-water areas (blue areas with IF values ​​close to 0) do not store water and therefore cannot support Anopheles habitat. Some areas (bright blue, white, or bright red) were flooded for several months during the study period and, although not as high risk as the red dots, these areas may become potential habitats as precipitation rates increase and the hydrological system changes. The low IoU accuracy of 34.1% can be attributed to several factors in the classification and evaluation process. First, the complexity of the classified objects can significantly affect the accuracy of the classification and the sensitivity of the IoT metric to the boundaries of objects should be carefully considered as it plays a crucial role in evaluating the accuracy of boundary delineation. The amount of water in a depression depends on precipitation and the depression structure. Each depression may contain several concave points, which can be visually compared with pits (such as Fig.12 ). After rainfall, runoff fills these points first due to the slope. Increased rainfall can cause more flooding in other areas of the depression. Evaporation and permeability also affect the amount of water. In the SAM model, the entire depression is considered as a water body and the depression is fully extracted, but since some holes may not be filled with water, it is difficult to align the ground truth mask generated by SAM with the predicted water body, which reduces the intersection area and IoU value (such as Figure 2 b). Fig.12 In the figure, the profile shows a row of depressions that are located along the road, indicating that these depressions contain multiple cavities and are more likely to accumulate runoff than other areas, which leads to a reduction in IoT indicators: (a) Overview; (b) Ground profile.

[0110] Furthermore, the model achieved 98.7% accuracy and 93.4% recall, demonstrating its accuracy in effectively identifying water bodies, a balance that ensures accurate extraction while minimizing false positives, which is critical for reliable water resource monitoring. The overall performance of the model is strong, as confirmed by the high F1 score, which achieved an overall accuracy of 96.1% and a kappa coefficient of 92.2%. Key factors contributing to this success include the suitability of Sentinel-1 imagery for distinguishing water bodies from other land cover and the robustness of the classification model. Object-based assessment and the model’s ability to adapt to temporal changes in hydrodynamics further improve its performance. Limitations in image resolution, especially for smaller depressions, can lead to misclassification. Addressing this issue may involve higher-resolution data sources or post-processing techniques. However, the synergy of suitable satellite data and a well-optimized model allows for extremely high accuracy in delineating water bodies using Sentinel-1 imagery.

[0111] Freely available radar images and environmental parameters from meteorological datasets were utilized, and all processing was done in cloud-based services such as GEE to efficiently manage extensive time series analysis, eliminating the need for physical data storage and processing resources. The proposed method has the potential to significantly reduce the need for extensive field visits and manual operations. By utilizing medium-resolution data, the proposed method can identify high-risk areas over large, extensive areas. By using long-time series of Sentinel-1 data combined with environmental parameters such as precipitation and temperature, the kappa coefficient is consistent with Ovakoglou's study on non-vegetated wetlands and rice fields. Compared to Hardy's findings, a higher kappa coefficient was shown. As an alternative to machine learning models that have not shown good accuracy, the proposed method is a flexible solution that demonstrates its efficiency in detecting water in a wide range of research areas and contributes to the overall progress in identifying Anopheles larval habitats. Water maps should never be relied upon alone to portray malaria risks and develop disease control strategies. Instead, the relationship between the location of identified habitats and human population density should be carefully considered.

[0112] Remote sensing has been widely used in disease control and prevention, including malaria. Unlike earlier studies, this method utilizes medium spatial resolution Sentinel-1 imagery aggregated monthly over a long period of time and combines it with temperature and precipitation data from a meteorological dataset. The process volume can be reduced using precipitation and temperature parameters and limiting the process to high-risk months. By adopting a majority voting method for monthly, annual, and overall analysis, the method effectively detects temporary water bodies and small depressions with a high risk of Anopheles larval habitat formation. To overcome the challenges associated with limited training data, the proposed method can effectively identify and monitor high-risk Anopheles habitats over a large study area. It is able to overcome the challenges associated with obtaining sufficient training data and defining robust features for machine learning models. In addition, the proposed method is more accurate than classification models. The design of the method ensures generalizability and scalability across different geographical regions. The method can be effectively localized by adjusting parameters according to local environmental conditions, such as different precipitation patterns and temperature ranges. This flexibility makes the method applicable to a variety of environments, enhancing its practicality in global disease surveillance and control efforts. It is a valuable tool for detecting potential breeding sites of Anopheles mosquitoes, especially in areas prone to vector-borne diseases where field work is expensive, time-consuming and challenging. This research not only promotes the application of remote sensing technology in public health, but also provides a practical means for disease monitoring and control.

[0113] In all examples shown and described herein, any specific values ​​should be interpreted as merely exemplary and not as limiting, and thus other examples of the exemplary embodiments may have different values.

[0114] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0115] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A method for identifying high-risk areas for malaria based on Sentinel-1 time series, characterized in that: The following steps are involved: Acquire a data set of the study area, and pre-process the data set to obtain a high-risk time for Anopheles reproduction, wherein the data set includes Sentinel-1 data, precipitation data, and temperature data; Using Google Earth Engine GEE to analyze the Sentinel-1 data to determine the SWI index corresponding to the Sentinel-1 data, and detecting water bodies in the study area on a SWI map corresponding to the SWI index; The high-risk time and the water body are classified by majority voting fusion to obtain a time classification map, and the high-risk location for Anopheles reproduction is identified according to the time classification map; The water bodies are subjected to a SAM model masking assessment to determine the high-risk areas for malaria corresponding to the high-risk locations.

2. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 1, characterized in that: The data set is preprocessed to obtain the high-risk time for Anopheles reproduction, including: Select a seven-year time series and use effective environmental parameters to predict the time of malaria outbreak, wherein the seven-year time series includes climate conditions, seasonal patterns and abnormal weather events, and the environmental parameters include precipitation and temperature values; Arrange the temperature values ​​from low to high to determine the lower limit LDT and upper limit UDT of the development temperature of Anopheles; The rainy season temperature suitable for the growth and stock of Anopheles mosquitoes was extracted based on the seven-year time series, precipitation and temperature values, and the rainy season temperature and the seven-year time series were used in a water extraction model.

3. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 2, characterized in that: Also includes: The seven-year time series is temporally aggregated to obtain a Sentinel-1 image, wherein a median operator of the temporal aggregation is used to obtain the Sentinel-1 image through GEE.

4. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 3, characterized in that: Detecting the water body in the study area on the SWI map corresponding to the SWI index includes: The SWI index was calculated based on the Sentinel-1 images, and after extracting water bodies every month for seven years, the maps corresponding to the Sentinel-1 images were fused to identify high-risk areas.

5. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 4, characterized in that: The high-risk time and the water body are classified by majority voting fusion to obtain a time classification map, including: Detect and classify the high-risk areas, including: The first annual analysis identifies areas that have water four months of the year to determine where water is stored continuously year-round; the second monthly analysis identifies areas that have water stored over the years; A four-level regional classification map was used to designate regions with water storage every seven years as high-risk, regions with water storage between four and six years as relatively risky, and regions with water storage less than four years as low-risk, and a fourth regional classification map was assigned to regions with no water in any of the study months; The high-risk areas are detected using an abstract hierarchy algorithm and majority voting fusion, wherein the majority voting fusion includes each regional classification map voting on a specific category, all votes have the same weight, and majority voting is used for monthly and annual analysis.

6. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 1, characterized in that: Conducting a SAM model masking assessment on the water body to determine the high-risk areas for malaria corresponding to the high-risk locations, including: Using multi-point feature hints for water body segmentation, the points in the geographic data frame are converted into a list of coordinates [x, y]; Use the SamGeo tool as a predictor for a single coordinate, predict the features of a single coordinate in the instance to get the mask, and save the obtained mask to a unique file for each point; After generating a mask for each point, all masks are merged into a single mosaic raster file, providing a complete representation of all segmented regions from the multi-point feature cues.

7. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 6, characterized in that: Also includes: Create a geo-referenced image by downloading tiles and using a tile map service, segment the geo-referenced image by selecting a satellite image, and define a SAM model by setting a path; Loop through the coordinate point list to predict, segment and save each point into a unique Geotiff file; All masks were resampled to 10m resolution and merged into one file as a mosaic mask containing information of all segmented regions; Masks of multiple depressions generated by SAM were used to assess the accuracy of quantities extracted from the SWI index.

8. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 7, characterized in that: The performance of the SAM model was evaluated using confusion matrix and comprehensive indicators including precision, recall, F1 score, overall accuracy and kappa coefficient.

9. The method for identifying high-risk areas for malaria based on Sentinel-1 time series according to claim 7, characterized in that: After calculating the surface water distribution map corresponding to the georeferenced image, the flooding frequency IF of the depression is calculated using expression (1): Among them, nundation Frequency represents the flooding frequency IF of the depression, Water Observation Times represents the water observation time, and Total Observation Times represents the total observation time.

10. A device for identifying high-risk areas for malaria based on Sentinel-1 time series, applied to the method for identifying high-risk areas for malaria based on Sentinel-1 time series as claimed in any one of claims 1 to 9, characterized in that: include: A data acquisition unit, used to acquire a data set in a research area, and pre-process the data set to obtain a high-risk time for Anopheles reproduction, wherein the data set includes Sentinel-1 data, precipitation data, and temperature data; a water body detection unit, configured to analyze the Sentinel-1 data using Google Earth Engine GEE to determine a SWI index corresponding to the Sentinel-1 data, and detect water bodies in the study area on a SWI map corresponding to the SWI index; A classification and identification unit, configured to classify the high-risk time and the water body by majority voting fusion to obtain a time classification map, and identify the high-risk location for Anopheles breeding according to the time classification map; The model evaluation unit is used to perform a SAM model masking evaluation on the water body to determine the high-risk area for malaria corresponding to the high-risk location.

Images