WO2023116454A1 - 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质 - Google Patents

一种蝗灾潜在高风险区识别方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2023116454A1
WO2023116454A1 PCT/CN2022/137653 CN2022137653W WO2023116454A1 WO 2023116454 A1 WO2023116454 A1 WO 2023116454A1 CN 2022137653 W CN2022137653 W CN 2022137653W WO 2023116454 A1 WO2023116454 A1 WO 2023116454A1
Authority
WO
WIPO (PCT)
Prior art keywords
index
area
image data
remote sensing
pixel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2022/137653
Other languages
English (en)
French (fr)
Inventor
赵龙龙
陈劲松
李晓丽
韩宇
姜小砾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Institute of Advanced Technology of CAS
Original Assignee
Shenzhen Institute of Advanced Technology of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Institute of Advanced Technology of CAS filed Critical Shenzhen Institute of Advanced Technology of CAS
Publication of WO2023116454A1 publication Critical patent/WO2023116454A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/24323—Tree-organised classifiers

Definitions

  • This application relates to the technical fields of pest control and remote sensing, and in particular to a method, device, equipment and storage medium for identifying potentially high-risk areas of locust disasters.
  • Locust disasters are biological disasters that have a devastating blow to agricultural production, and the East Asian migratory locust is the most serious locust species in my country.
  • the monitoring and reporting of migratory locusts in East Asia has been considered one of the important tasks of the plant protection department. Its goal is to detect locust breeding areas in time and take measures to control the population density of locusts and prevent locusts from gathering in swarms and taking off to cause disasters.
  • the locust forecasting system of my country's plant protection department is relatively traditional, and the work is mostly carried out on "points" through manual visual inspection and field sampling, that is, to investigate the locust situation by manually digging eggs and sampling, and pulling nets.
  • Meteorological conditions are based on subjective experience and knowledge to forecast locusts.
  • these methods are labor-intensive and prone to leaks and disasters; on the other hand, they are highly dependent on the professional level of local plant protection personnel, and the actual professional level of relevant personnel varies, which cannot meet the needs of prevention and control.
  • the above methods ignore spatial information, resulting in a lack of accuracy in locust situation forecasting, which is not enough to support the management department to take effective prevention and control measures in a timely manner.
  • RS remote sensing
  • GIS geographic information system
  • locusts are easy to gather to form high-density spots, which become a potential high-risk area for locust disasters.
  • This application provides a method, device, equipment and storage medium for identifying potentially high-risk areas of locust plagues, so as to solve the problem of "monitoring blind areas" that are prone to appear in the current prevention and control system.
  • a technical solution adopted by this application is a method for identifying potential high-risk areas of locust disasters, including: obtaining long-term sequence multi-temporal remote sensing image data of the research area based on a preset method; applying multi-temporal remote sensing image data to Classify the land cover in the study area, and calculate the habitat suitability index of each pixel in the study area based on the first preset rule and multi-temporal remote sensing image data; based on the second preset rule, use the multi-temporal remote sensing image data to calculate the study area
  • the habitat suitability index of each pixel in the research area is calculated based on the first preset rule and the multi-temporal remote sensing image data, including: calculating the normalized difference vegetation in the research area based on the multi-temporal remote sensing image data index, normalized difference water index, normalized difference building index, built-up area index and soil salinity index; normalized difference vegetation index, normalized difference water index, normalized difference building index , built-up area index and soil salinity index are combined with multi-temporal remote sensing image data to obtain the multi-temporal remote sensing synthetic image data of the study area; using the preset classification algorithm, each pixel in the study area is classified according to the multi-temporal remote sensing synthetic image data Classify according to the preset land type to obtain the land cover type data of each pixel; calculate the vegetation coverage of the study area according to the normalized difference vegetation index; calculate the surface temperature and temperature of the study area based on the multi-temporal remote sensing image data Vegetation drought index: Evaluate each
  • the preset classification algorithm is used to classify each pixel in the research area according to the preset land type according to the multi-temporal remote sensing synthetic image data, and the land cover type data of each pixel is obtained, including: obtaining The pre-built and trained random forest classifier is trained according to the pre-prepared samples and preset land types; the multi-temporal remote sensing synthetic image data corresponding to each pixel is input into the random forest classifier, and each The land cover type data of pixels.
  • each pixel in the study area is evaluated according to land cover type data, vegetation coverage, surface temperature, soil salinity index and temperature vegetation drought index, and the habitat suitability index of each pixel is obtained , including: constructing suitability grades based on land cover type data, vegetation coverage, surface temperature, soil salinity index, and temperature vegetation drought index, and assigning values to each grade; based on the principle of patch scale, according to the corresponding Land cover type data, vegetation coverage, surface temperature, soil salinity index and temperature vegetation drought index are calculated to obtain the suitability grade of the central pixel patch scale; Habitat suitability index of each pixel is calculated according to the degree of suitability.
  • the step of calculating the water body change index of the research area by using the multi-temporal remote sensing image data includes: obtaining the water body in the research area corresponding to the multi-temporal remote sensing image data according to the multi-temporal remote sensing image data and the land cover type data.
  • the step of calculating the suitable habitat change index of the research area by using the multi-temporal remote sensing image data includes: obtaining the locust suitable habitat in the research area according to the multi-temporal remote sensing image data and the suitability level of each pixel.
  • the suitable habitat change index of locust suitable habitat in the period corresponding to the multi-temporal remote sensing image data is obtained by calculating with the maximum area.
  • the first preset condition is:
  • tmin is the time corresponding to the minimum water body area
  • tmax is the time corresponding to the maximum suitable habitat area for locusts
  • t1 is the starting time point of multi-temporal remote sensing image data
  • tn is the end time point of multi-temporal remote sensing image data
  • WS is the water body area corresponding to each time point in the time period corresponding to the multi-temporal remote sensing image data
  • LHS is the suitable locust habitat area corresponding to each time point in the time period corresponding to the multi-temporal remote sensing image data
  • ⁇ WS is the water body change index
  • ⁇ LHS is the suitable Habitat change index
  • C 1 is the preset water body change index threshold
  • C 2 is the preset suitable habitat change index threshold
  • the second preset condition is:
  • D water is the distance between the non-water body pixel and the nearest water body pixel
  • HSI is the habitat suitability index of the pixel
  • D is the preset distance threshold
  • C is the preset habitat suitability index threshold.
  • a potential high-risk area identification device for locust disasters including: an acquisition module for acquiring long-term multi-temporal remote sensing images of the research area based on a preset method Data; calculation module, used to apply multi-temporal remote sensing image data to classify the land cover of the research area, and calculate the habitat suitability index of each pixel in the research area based on the first preset rule and multi-temporal remote sensing image data; the first confirmation The module is used to calculate the water body change index and the suitable habitat change index of the research area based on the second preset rule by using the multi-temporal remote sensing image data, and when the water body change index and the suitable habitat change index meet the first preset condition, confirm There are potential high-risk areas of locust disasters in the research area; the second confirmation module is used to confirm the water body pixels of the latest image in the multi-temporal remote sensing image data, and obtain the distance between each non-water body pixel and the nearest
  • the computer device includes a processor, a memory coupled to the processor, and program instructions are stored in the memory, so When the program instructions are executed by the processor, the processor is made to execute the steps of the locust disaster risk prediction method described above.
  • another technical solution adopted by the present application is to provide a storage medium storing program instructions capable of realizing the above-mentioned locust disaster risk prediction method.
  • the method for identifying potential high-risk areas of locust disasters in this application extracts and analyzes the water body and suitable habitat for locusts in the research area to obtain the water body change index of the research area by using the long-term sequence of multi-temporal remote sensing image data in the research area and suitable habitat change index, and when the water body change index and suitable habitat change index meet the first preset condition, it is confirmed that there is a potential high-risk area of locust plague in the study area, and then the habitat suitability index calculated by using the latest remote sensing image data and the distance between each area and the water body in the study area to confirm the area where locust plague risk may occur in the study area, which fully considers the evolution process of locust habitat from unsuitable to suitable, and considers the characteristics of locust cluster outbreaks.
  • the high-precision identification of potential high-risk areas of locust disasters has been realized.
  • Fig. 1 is the schematic flow chart of the locust plague potential high-risk area identification method of the embodiment of the present invention
  • Fig. 2 is a functional module schematic diagram of a locust plague potential high-risk area identification device according to an embodiment of the present invention
  • Fig. 3 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
  • first”, “second”, and “third” in this application are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Thus, features defined as “first”, “second”, and “third” may explicitly or implicitly include at least one of these features.
  • “plurality” means at least two, such as two, three, etc., unless otherwise specifically defined. All directional indications (such as up, down, left, right, front, back%) in the embodiments of the present application are only used to explain the relative positional relationship between the various components in a certain posture (as shown in the drawings) , sports conditions, etc., if the specific posture changes, the directional indication also changes accordingly.
  • FIG. 1 is a schematic flowchart of a method for identifying potential high-risk areas of locust plagues according to an embodiment of the present invention. It should be noted that the method of the present invention is not limited to the flow sequence shown in FIG. 1 if substantially the same result is obtained. As shown in Figure 1, the method includes steps:
  • Step S101 Acquire long-term multi-temporal remote sensing image data of the study area based on a preset method.
  • Step S102 Using the multi-temporal remote sensing image data to classify the land cover of the research area, and calculate the habitat suitability index of each pixel in the research area based on the first preset rule and the multi-temporal remote sensing image data.
  • the land cover classification of the study area is carried out, and then the habitat suitability index of each pixel is calculated.
  • the multi-temporal remote sensing image data in the study area were screened according to the time period of image acquisition, and the remote sensing images of n periods with image cloud content less than 10% were screened out.
  • remote sensing image data usually includes B1 (coastal band), B2 (blue band), B3 (green band), B4 (red band), B5 (near-infrared band), B6 (short-wave infrared band 1), B7 (shortwave infrared 2 band), B8 (panchromatic band), B9 (cirrus cloud band), B10 (thermal infrared 1 band) and B11 (thermal infrared 2 band).
  • step S102 the multi-temporal remote sensing image data is used to classify the land cover of the research area, and the step of calculating the habitat suitability index of each pixel in the research area based on the first preset rule and the multi-temporal remote sensing image data includes:
  • the above screened multi-temporal remote sensing image data are used to calculate the normalized difference vegetation index, normalized difference water index, normalized difference building index, and built-up area index of the study area. and soil salinity index.
  • NDVI represents the normalized difference vegetation index, dimensionless, and the value range is [-1,1];
  • B4 represents the brightness value of the red band of the remote sensing image data;
  • B5 represents the brightness value of the near-infrared band of the remote sensing image data.
  • NDWI represents the normalized difference water index, and the value range is [-1,1];
  • B3 represents the brightness value of the green band of the remote sensing image data;
  • B5 represents the brightness value of the near-infrared band of the remote sensing image data.
  • NDBI represents the normalized difference building index, dimensionless, and the value range is [-1,1]
  • B6 represents the brightness value of the short-wave infrared band of remote sensing image data
  • B5 represents the brightness value of the near-infrared band of remote sensing image data.
  • BUI means built-up area index, dimensionless, and the value range is [-2,2];
  • NDBI means normalized difference building index, and
  • NDVI means normalized difference vegetation index.
  • the formula for calculating the soil salinity index is:
  • SSI represents the soil salinity index
  • B3 represents the brightness value of the green band of the remote sensing image data
  • B4 represents the brightness value of the red band of the remote sensing image data.
  • the calculated normalized difference vegetation index, normalized difference water index, normalized difference building index, built-up area index and remote sensing image data of corresponding time phases are band synthesized.
  • the bands were named NDVI, NDWI, NDBI and BUI in turn, and the multi-temporal remote sensing synthetic image data of the study area were obtained.
  • the object-oriented Random Forest (Random Forest) classification algorithm is used to classify each pixel of the multi-temporal remote sensing synthetic image data in the study area according to the preset land type.
  • the assumed land types include cultivated land, herbaceous wetland, reed and weeds, woodland, water body, artificial surface and other 7 types of land cover type data.
  • the preset classification algorithm is used to classify each pixel in the study area according to the preset land type according to the multi-temporal remote sensing synthetic image data, and the land cover type data of each pixel is obtained, including:
  • random forest classifier needs to be pre-trained, and the specific training process is as follows:
  • Bands involved in classifier training and testing include 12 bands including B2, B3, B4, B5, B6, B7, B10, B11, NDVI, NDWI, NDBI and BUI;
  • m preferably takes a value of 100;
  • the seed points according to the rule of scale d Apply the Simple Non-Iterative Clustering (SNIC) segmentation algorithm on the images synthesized by the B6, B5, and B4 bands to perform superpixel segmentation on the images in the study area, Form the object, and count the average value of the 12 input bands according to the object, and then set the three parameters of the SNIC segmentation algorithm according to the actual needs: compactness, connectivity, and neighborhood size.
  • SNIC Simple Non-Iterative Clustering
  • the preferred value of the scale d for selecting seed points is 5, and the three parameters of the SNIC segmentation algorithm are preferably 0.5, 8 and 256.
  • the random forest classifier is used to perform object-oriented land cover classification on the multi-temporal remote sensing synthetic image data of the study area, and the land cover type data of each pixel is obtained, and the classification
  • land cover type data LC The result is named land cover type data LC.
  • Fv represents the vegetation coverage
  • NDVI represents the normalized difference vegetation index
  • NDVI soil represents the NDVI value of a pixel completely bare soil or no vegetation coverage
  • NDVI veg represents the NDVI value of a pixel completely covered by vegetation .
  • the formula for calculating the surface temperature is:
  • LST represents the surface temperature in degrees Celsius
  • Tb is the TOA brightness temperature in the thermal infrared channel
  • ⁇ is the surface emissivity of the thermal infrared channel
  • the coefficients A i , B i , and C i are calculated by the 10 groups of total column water vapor ( TCWV) (range 0-6 cm, step size 0.6 cm, TCWV values higher than 6 cm for the last group) was determined by linear regression of the radiative transfer model performed.
  • the formula for calculating the temperature vegetation drought index is:
  • LST Smax a*NDVI+b
  • LST Smin c*NDVI+d
  • TVDI represents the temperature vegetation drought index
  • LST S represents the surface temperature when the NDVI value of a certain pixel is S
  • LST Smax represents the maximum value of the surface temperature when the NDVI is equal to a certain value S
  • LST Smin represents the maximum value of the surface temperature when the NDVI is equal to
  • the minimum value of surface temperature at a certain value S, a, b, c, d are undetermined coefficients; the value range of TVDI is [0,1], the larger the value, the lower the soil moisture, and the smaller the value, the higher the soil moisture high.
  • TVDI data is used to characterize the soil moisture SM in the study area.
  • the locust habitat impact factor data is used to evaluate the habitat suitability of East Asian migratory locusts in the study area, so as to obtain the habitat suitability index of each pixel, and confirm the suitability level corresponding to each pixel based on the habitat suitability index.
  • the suitability grades are set in advance according to the land cover type data LC, vegetation coverage Fv, surface temperature LST, soil salinity SS, and soil moisture SM, which are divided into the most suitable, second most suitable, generally suitable and unsuitable in turn.
  • Four levels are suitable, and the corresponding level data are assigned as 4, 3, 2, and 1 respectively.
  • table 1 this is the first default rule:
  • the soil salinity SS* and soil moisture SM* in the table are the soil salinity index and temperature vegetation drought index in the study area, and the two indexes are normalized data, not the absolute percentage of soil salinity and soil moisture. Due to the difference in data acquisition time or cultivation and planting status, the suitability grade division of the two factors in different research areas or different time phases will be different. The above table is only an example of the division of a certain time phase in the present invention. In actual use, the two factors The suitability level of each factor should be divided according to the actual situation of the study area.
  • each pixel in the study area is evaluated according to land cover type data, vegetation coverage, surface temperature, soil salinity index and temperature vegetation drought index, and the habitat suitability index of each pixel is obtained, including:
  • the suitability grade of the central pixel patch scale is calculated according to the land cover type data, vegetation coverage, surface temperature, soil salinity index and temperature vegetation drought index corresponding to each pixel.
  • this embodiment fully considers the impact of landscape structure on locust habitat, and at the same time, in order to eliminate the salt and pepper effect of remote sensing data, this embodiment introduces the concept of patch scale suitability level, that is, a window with a given size w is used as a patch block, adopt the moving window method, introduce the level information of adjacent pixels, and realize the determination of the scale suitability level of the central pixel patch, the calculation formula is as follows:
  • M 1,p , M 2,p , M 3,p , M 4,p and M 5,p respectively represent land cover type data LC, vegetation coverage Fv, surface temperature LST, soil salinity SS, soil moisture SM is the grade membership degree of the five habitat factors at the patch scale;
  • w is the size of the moving window (odd);
  • x, y are the number of rows and columns in the study area;
  • j, k are the number of rows and columns in the window;
  • M t (x j ,y k ) represents the grade membership of factor M t located on the (x j ,y k ) pixel;
  • d j,k is the center pixel of the moving window (x (w+1)/2 ,y (w +1)/2 ) to the adjacent pixel (x j
  • the weights of the five habitat factors of land cover type data LC, vegetation coverage Fv, surface temperature LST, soil salinity SS, and soil moisture SM are respectively taken as 0.30 , 0.28, 0.11, 0.13 and 0.18.
  • the suitability level of each pixel is set according to the relationship between the index and the index threshold, and the specific level setting conditions are as follows (this is the second preset rule):
  • Step S103 Based on the second preset rule, the water body change index and the suitable habitat change index of the research area are calculated by using the multi-temporal remote sensing image data, and when the water body change index and the suitable habitat change index meet the first preset condition, the research is confirmed. There are potential high-risk areas for locust plagues in the region.
  • the water body change index represents the change trend and intensity of the water body area in the study area over time
  • the suitable habitat change index represents the change trend and intensity of the locust suitable habitat area in the study area over time
  • the steps of calculating the water body change index in the study area by using multi-temporal remote sensing image data include:
  • the multi-temporal remote sensing image data and land cover type data obtain the minimum area of the water body in the study area in the period corresponding to the multi-temporal remote sensing image data, the first area at the starting time point of the multi-temporal remote sensing image data and the multi-temporal remote sensing image data.
  • the water body change index of the water body in the period corresponding to the multi-temporal remote sensing image data is obtained.
  • ⁇ WS is the water body change index, and its value range is (-1,0).
  • is to 1, the more severe the change in water body area is, and the closer to 0 is to indicate that the change is weaker
  • tmin is the water body in the study area. The time corresponding to the minimum area.
  • the area of the water body is calculated by extracting the water body from the land cover type data LC;
  • WS tmin is the minimum area of the water body in the period corresponding to the multi-temporal remote sensing image data in the study area,
  • WS t1 is the first area of the water body at the starting time point of the multi-temporal remote sensing image data,
  • WS tn is the second area of the water body at the end time point of the multi-temporal remote sensing image data.
  • the steps of using multi-temporal remote sensing image data to calculate the suitable habitat change index of the study area include:
  • the multi-temporal remote sensing image data and the suitability level of each pixel obtain the maximum area of locust suitable habitat in the study area in the period corresponding to the multi-temporal remote sensing image data, and the first time point at the starting time point of the multi-temporal remote sensing image data The third area and the fourth area at the end time point of the multi-temporal remote sensing image data.
  • the suitable habitat change index of the locust suitable habitat in the period corresponding to the multi-temporal remote sensing image data is obtained.
  • the formula for calculating the suitable habitat change index is:
  • ⁇ LHS is the suitable habitat change index, and its value range is (-1,0).
  • is to 1, the more severe the change in the suitable habitat area is, and the closer to 0, the weaker the change is
  • tmax is the suitable habitat for locusts
  • the area of suitable habitat for locusts is the sum of the area of the most suitable habitat and the sub-suitable habitat, and the sum of the areas of the most suitable habitat and the sub-suitable habitat is based on the above-mentioned determination of each pixel After the suitability level is determined, it is obtained by adding the area of the most suitable pixel and the second most suitable pixel;
  • LHS tmax is the maximum area of locust suitable habitat in the study area in the period corresponding to the multi-temporal remote sensing image data, and LHS t1 is The third area at the start time point of the multi-temporal remote sensing image data, LHS tn is the fourth area at the end time point
  • the first preset condition is:
  • tmin is the time corresponding to the minimum water body area
  • tmax is the time corresponding to the maximum suitable habitat area for locusts
  • t1 is the starting time point of multi-temporal remote sensing image data
  • tn is the end time point of multi-temporal remote sensing image data
  • WS is the water body area corresponding to each time point in the time period corresponding to the multi-temporal remote sensing image data
  • LHS is the suitable locust habitat area corresponding to each time point in the time period corresponding to the multi-temporal remote sensing image data
  • ⁇ WS is the water body change index
  • ⁇ LHS is the suitable Habitat change index
  • C 1 is the preset water body change index threshold
  • C 2 is the preset suitable habitat change index threshold.
  • Step S104 Confirm the water body pixels of the latest image in the multi-temporal remote sensing image data, and obtain the distance between each non-water body pixel and the nearest water body pixel, and when there is a suitable habitat corresponding to the target non-water body pixel When the sex index and distance meet the second preset condition, it is confirmed that the area corresponding to the target non-water body pixel is a potential high-risk area of locust plague.
  • the distance and habitat suitability of each non-water body pixel is judged to confirm whether the area corresponding to the non-water body pixel is a potential high-risk area of locust plague.
  • the second preset condition is:
  • D water is the distance between the non-water body pixel and the nearest water body pixel
  • HSI is the habitat suitability index of the pixel
  • D is the preset distance threshold
  • C is the preset habitat suitability index threshold.
  • the preset distance threshold D is preferably 2 km
  • the preset habitat suitability index threshold C is preferably 3.
  • the method for identifying potential high-risk areas of locust plagues in the embodiment of the present invention extracts and analyzes the water body and suitable habitat for locusts in the research area by using the long-term sequence of multi-temporal remote sensing image data in the research area to obtain the water body change index and suitable habitat change index in the research area , and when the water body change index and the suitable habitat change index meet the first preset condition, it is confirmed that there is a potential high-risk area of locust plague in the study area, and then the habitat suitability index calculated by using the latest remote sensing image data and each area and the research area
  • the distance between water bodies in the area is used to confirm the potential high-risk areas of locust plagues in the study area, which fully considers the evolution process of locust habitats from unsuitable to suitable, and also considers the characteristics of locust cluster outbreaks, combined with the dynamic changes of drought and flood, High-precision identification of potential high-risk areas of locust plagues has been achieved.
  • Fig. 2 is a schematic diagram of functional modules of a device for identifying potentially high-risk areas of locust plagues according to an embodiment of the present invention.
  • the device 20 includes an acquisition module 21 , a calculation module 22 , a first confirmation module 23 and a second confirmation module 24 .
  • An acquisition module 21 configured to acquire long-term multi-temporal remote sensing image data of the research area based on a preset method
  • Calculation module 22 used to apply multi-temporal remote sensing image data to classify the land cover of the research area, and calculate the habitat suitability index of each pixel in the research area based on the first preset rule and multi-temporal remote sensing image data; the first confirmation module 23. It is used to calculate the water body change index and the suitable habitat change index of the research area based on the second preset rule by using the multi-temporal remote sensing image data, and when the water body change index and the suitable habitat change index meet the first preset condition, confirm There are potential high-risk areas of locust plague in the study area;
  • the second confirmation module 24 is used to confirm the water body pixel of the latest image in the multi-temporal remote sensing image data, and obtain the distance between each non-water body pixel and the nearest water body pixel, and when there is a target non-water body image
  • the habitat suitability index and distance corresponding to the pixel meet the second preset condition, it is confirmed that the area corresponding to the target non-water body pixel is a potential high-risk area of locust plague.
  • the calculation module 22 executes the operation of calculating the habitat suitability index of each pixel in the research area based on the first preset rule and the multi-temporal remote sensing image data, specifically including: calculating the normalization index of the research area based on the multi-temporal remote sensing image data Normalized difference vegetation index, normalized difference water index, normalized difference building index, built-up area index and soil salinity index; normalized difference vegetation index, normalized difference water index, normalized difference The difference building index, built-up area index and soil salinity index are combined with the multi-temporal remote sensing image data to obtain the multi-temporal remote sensing synthetic image data of the study area; Each pixel is classified according to the preset land type to obtain the land cover type data of each pixel; the vegetation coverage of the study area is calculated according to the normalized difference vegetation index; the vegetation coverage of the study area is calculated according to the multi-temporal remote sensing image data Surface temperature and temperature vegetation drought index; each pixel in the study area is evaluated according to land cover type data, vegetation coverage, vegetation coverage
  • the calculation module 22 executes the operation of using a preset classification algorithm to classify each pixel in the research area according to the preset land type according to the multi-temporal remote sensing synthetic image data, and obtain the land cover type data of each pixel, Specifically include: obtaining a pre-built and trained random forest classifier, which is trained based on pre-prepared samples and preset land types; inputting the multi-temporal remote sensing synthetic image data corresponding to each pixel into the random forest classification device to get the land cover type data of each pixel.
  • the calculation module 22 executes the evaluation of each pixel in the research area according to the land cover type data, vegetation coverage, surface temperature, soil salinity index and temperature vegetation drought index, and obtains the habitat suitability of each pixel
  • the operation of the index includes: constructing suitability grades based on land cover type data, vegetation coverage, surface temperature, soil salinity index and temperature vegetation drought index, and assigning values to each grade; based on the principle of patch scale, according to each The corresponding land cover type data, vegetation coverage, surface temperature, soil salinity index and temperature vegetation drought index are calculated to obtain the suitability level of the central pixel patch scale; The suitability grade at the block scale is calculated to obtain the habitat suitability index of each pixel.
  • the first confirmation module 23 executes the operation of calculating the water body change index of the research area by using the multi-temporal remote sensing image data, which specifically includes: obtaining the multi-temporal remote sensing data of the water body in the research area according to the multi-temporal remote sensing image data and the land cover type data.
  • the water body change index of the water body in the period corresponding to the multi-temporal remote sensing image data is calculated.
  • the first confirmation module 23 executes the operation of calculating the suitable habitat change index of the research area by using the multi-temporal remote sensing image data, which specifically includes: obtaining the habitat change index in the research area according to the multi-temporal remote sensing image data and the suitability level of each pixel.
  • the fourth area and the maximum area are calculated to obtain the suitable habitat change index of the locust suitable habitat in the period corresponding to the multi-temporal remote sensing image data.
  • the first preset condition is:
  • tmin is the time corresponding to the minimum water body area
  • tmax is the time corresponding to the maximum suitable habitat area for locusts
  • t1 is the starting time point of multi-temporal remote sensing image data
  • tn is the end time point of multi-temporal remote sensing image data
  • WS is the water body area corresponding to each time point in the time period corresponding to the multi-temporal remote sensing image data
  • LHS is the suitable locust habitat area corresponding to each time point in the time period corresponding to the multi-temporal remote sensing image data
  • ⁇ WS is the water body change index
  • ⁇ LHS is the suitable Habitat change index
  • C 1 is the preset water body change index threshold
  • C 2 is the preset suitable habitat change index threshold
  • the second preset condition is:
  • D water is the distance between the non-water body pixel and the nearest water body pixel
  • HSI is the habitat suitability index of the pixel
  • D is the preset distance threshold
  • C is the preset habitat suitability index threshold.
  • each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments.
  • the same and similar parts in each embodiment refer to each other, that is, Can.
  • the description is relatively simple, and for related parts, please refer to part of the description of the method embodiments.
  • FIG. 3 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
  • the computer device 60 includes a processor 61 and a memory 62 coupled to the processor 61.
  • Program instructions are stored in the memory 62.
  • the processor 61 performs any of the above-mentioned operations. The steps of the locust plague potential high-risk area identification method described in the embodiment.
  • the processor 61 may also be called a CPU (Central Processing Unit, central processing unit).
  • the processor 61 may be an integrated circuit chip with signal processing capabilities.
  • the processor 61 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components .
  • DSP digital signal processor
  • ASIC application-specific integrated circuit
  • FPGA field programmable gate array
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • FIG. 4 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
  • the storage medium in the embodiment of the present invention stores program instructions 71 capable of realizing all the above-mentioned methods, wherein the program instructions 71 can be stored in the above-mentioned storage medium in the form of software products, including several instructions to make a computer device (which can It is a personal computer, a server, or a network device, etc.) or a processor (processor) that executes all or part of the steps of the methods described in the various embodiments of the present application.
  • a computer device which can It is a personal computer, a server, or a network device, etc.
  • processor processor
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc and other media that can store program codes. , or computer equipment such as computers, servers, mobile phones, and tablets.
  • the disclosed computer equipment, devices and methods may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components can be combined or integrated. to another system, or some features may be ignored, or not implemented.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units. The above is only the implementation mode of this application, and does not limit the scope of patents of this application. Any equivalent structure or equivalent process conversion made by using the contents of this application specification and drawings, or directly or indirectly used in other related technical fields, All are included in the scope of patent protection of the present application in the same way.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Remote Sensing (AREA)
  • Image Processing (AREA)

Abstract

一种蝗灾潜在高风险区识别方法、装置、设备及存储介质，其利用研究区内长时间序列的多时相遥感影像数据对研究区水体和蝗虫适宜生境进行提取（S102）并分析得到研究区的水体变化指数和适宜生境变化指数，并当水体变化指数和适宜生境变化指数满足第一预设条件时，确认该研究区存在蝗灾潜在高风险区（S103），再利用多时相遥感影像数据计算得到的生境适宜性指数以及各个区域与研究区中水体之间的距离来确认研究区内的蝗灾潜在高风险区（S104）。通过上述方式，该方法能够充分考虑生境由不适宜到适宜的演变过程，而且考虑了蝗虫聚集性暴发的特性，可实现潜在高风险区的高精度识别。

Description

一种蝗灾潜在高风险区识别方法、装置、设备及存储介质 技术领域
本申请涉及虫害防控及遥感技术领域，特别是涉及一种蝗灾潜在高风险区识别方法、装置、设备及存储介质。
背景技术
蝗虫灾害是对农业生产具有毁灭性打击的生物灾害，东亚飞蝗是对我国危害最严重的蝗虫种类。长期以来，东亚飞蝗虫情的测报工作被认为是植保部门的重要任务之一，其目标是及时发现蝗虫孳生区并采取措施，以控制蝗虫种群密度，防止蝗虫聚集成群、起飞成灾。
目前，我国植保部门的蝗虫测报体系相对传统，多通过人工目测手查和田间取样的方式在“点”上开展工作，即通过人工挖卵取样、拉网普查的方式调查蝗虫虫情，进而结合气象条件基于主观经验知识对蝗虫进行测报。这些方法一方面耗费人力、容易漏防成灾；另一方面，对于地方植保人员的专业水平具有较高依赖，而实际相关人员的专业水平参差不齐，不能满足防控需求。以上方法忽略了空间信息，导致蝗情测报缺乏准确性，不足以支撑管理部门及时采取有效防控措施。
随着遥感(RS)及地理信息系统(GIS)技术的不断发展，其大面积、实时对地观测及强大的空间数据处理能力为蝗虫的遥感监测提供了极大便利。相关学者将RS和GIS技术引入到蝗虫监测预警研究中，进行了诸如蝗虫生境因子遥感监测、生境遥感分类、生境适宜性评价、潜在高风险栖息地识别、蝗区数字化平台构建、蝗灾发生预测等多方面研究，真正实现了大尺度上蝗虫虫情的高精度监测预警，一定程度上提高了蝗虫的测报效率及水平。在以上成果的支撑下，政府和各级植保部门已基本可以实现东亚飞蝗在重点蝗区(蝗虫常发区)的有效控制。
但近年来，在气候变化影响下，极端降水和干旱事件频发，新的蝗 虫适生环境不断形成，且不在植保部门监测范围内，易形成蝗虫“监控盲区”，如遇降雨压缩生存环境，蝗虫极易聚集形成高密度点片，成为蝗灾潜在高风险区。
发明内容
本申请提供一种蝗灾潜在高风险区识别方法、装置、设备及存储介质，以解决当前防控体系易出现“监控盲区”的问题。
为解决上述技术问题，本申请采用的一个技术方案一种蝗灾潜在高风险区识别方法，包括：基于预设方式获取研究区的长时间序列的多时相遥感影像数据；应用多时相遥感影像数据对研究区进行土地覆盖分类，并基于第一预设规则和多时相遥感影像数据计算研究区每个像元的生境适宜性指数；基于第二预设规则，利用多时相遥感影像数据计算得到研究区的水体变化指数和适宜生境变化指数，且当水体变化指数和适宜生境变化指数满足第一预设条件时，确认研究区中存在蝗灾潜在高风险区；确认多时相遥感影像数据中最新一期影像的水体像元，并获取每个非水体像元与最近的水体像元之间的距离，且当存在目标非水体像元对应的生境适宜性指数和距离满足第二预设条件时，确认目标非水体像元对应的区域为蝗灾潜在高风险区。
作为本申请的进一步改进，基于第一预设规则和多时相遥感影像数据计算研究区每个像元的生境适宜性指数，包括：根据多时相遥感影像数据计算研究区的归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数；将归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数与多时相遥感影像数据进行波段合成，得到研究区的多时相遥感合成影像数据；利用预设分类算法，根据多时相遥感合成影像数据对研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据；根据归一化差值植被指数计算得到研究区的植被覆盖度；根据多时相遥感影像数据计算研究区的地表温度和温度植被干旱指数；根据土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱 指数对研究区中的每个像元进行评估，得到每个像元的生境适宜性指数，并基于生境适宜性指数确认每个像元对应的适宜性等级。
作为本申请的进一步改进，利用预设分类算法，根据多时相遥感合成影像数据对研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据，包括：获取预先构建并训练好的随机森林分类器，随机森林分类器根据预先准备的样本和预设土地类型训练得到；将每个像元对应的多时相遥感合成影像数据输入至随机森林分类器，得到每个像元的土地覆盖类型数据。
作为本申请的进一步改进，根据土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数对研究区中的每个像元进行评估，得到每个像元的生境适宜性指数，包括：构建基于土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数的适宜性等级，并为每个等级赋值；基于斑块尺度原理，根据每个像元对应的土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数计算得到中心像元斑块尺度的适宜性等级；利用线性加权求和的方式，根据中心像元斑块尺度的适宜性等级计算得到每个像元的生境适宜性指数。
作为本申请的进一步改进，利用多时相遥感影像数据计算得到研究区的水体变化指数的步骤，包括：根据多时相遥感影像数据和土地覆盖类型数据获取研究区中水体在多时相遥感影像数据对应的时段内的最小面积、在多时相遥感影像数据起始时间点的第一面积和在多时相遥感影像数据结束时间点的第二面积；根据第一面积、第二面积和最小面积计算得到水体在多时相遥感影像数据对应的时段内的水体变化指数。
作为本申请的进一步改进，利用多时相遥感影像数据计算得到研究区的适宜生境变化指数的步骤，包括：根据多时相遥感影像数据和每个像元的适宜性等级获取研究区中蝗虫适宜生境在多时相遥感影像数据对应的时段内的最大面积、在多时相遥感影像数据起始时间点的第三面积和在多时相遥感影像数据结束时间点的第四面积；根据第三面积、第四面积和最大面积计算得到蝗虫适宜生境在多时相遥感影像数据对应 的时段内的适宜生境变化指数。
作为本申请的进一步改进，第一预设条件为：
Figure PCTCN2022137653-appb-000001
其中，tmin为水体面积最小时所对应的时间，tmax为蝗虫适宜生境面积最大时所对应的时间，t1为多时相遥感影像数据起始时间点，tn为多时相遥感影像数据结束时间点，WS为多时相遥感影像数据对应的时段内各个时间点对应的水体面积，LHS为多时相遥感影像数据对应的时段内各个时间点对应的蝗虫适宜生境面积，δ WS为水体变化指数，δ LHS为适宜生境变化指数，C 1为预设水体变化指数阈值，C 2为预设适宜生境变化指数阈值
第二预设条件为：
Figure PCTCN2022137653-appb-000002
其中，D water为非水体像元与最近的水体像元之间的距离，HSI为像元的生境适宜性指数，D为预设距离阈值，C为预设生境适宜性指数阈值。
为解决上述技术问题，本申请采用的另一个技术方案是：提供一种蝗灾潜在高风险区识别装置，包括：获取模块，用于基于预设方式获取研究区的长时间序列的多时相遥感影像数据；计算模块，用于应用多时相遥感影像数据对研究区进行土地覆盖分类，并基于第一预设规则和多时相遥感影像数据计算研究区每个像元的生境适宜性指数；第一确认模块，用于基于第二预设规则，利用多时相遥感影像数据计算得到研究区的水体变化指数和适宜生境变化指数，且当水体变化指数和适宜生境变化指数满足第一预设条件时，确认研究区中存在蝗灾潜在高风险区；第二确认模块，用于确认多时相遥感影像数据中最新一期影像的水体像元，并获取每个非水体像元与最近的水体像元之间的距离，且当存在目标非水体像元对应的生境适宜性指数和距离满足第二预设条件时，确认目标非水体像元对应的区域为蝗灾潜在高风险区。
为解决上述技术问题，本申请采用的再一个技术方案是：提供一种 计算机设备，所述计算机设备包括处理器、与所述处理器耦接的存储器，所述存储器中存储有程序指令，所述程序指令被所述处理器执行时，使得所述处理器执行上述的蝗灾风险预测方法的步骤。
为解决上述技术问题，本申请采用的再一个技术方案是：提供一种存储介质，存储有能够实现上述蝗灾风险预测方法的程序指令。
本申请的有益效果是：本申请的蝗灾潜在高风险区识别方法通过利用研究区内长时间序列的多时相遥感影像数据对研究区水体和蝗虫适宜生境进行提取并分析得到研究区的水体变化指数和适宜生境变化指数，并当水体变化指数和适宜生境变化指数满足第一预设条件时，确认该研究区存在蝗灾潜在高风险区，再利用最新一期遥感影像数据计算得到的生境适宜性指数以及各个区域与研究区中水体之间的距离来确认研究区内可能会出现蝗灾风险的区域，其充分考虑了蝗虫生境由不适宜到适宜的演变过程，而且考虑了蝗虫聚集性暴发的特性，结合旱涝动态变化，实现了对蝗灾潜在高风险区的高精度识别。
附图说明
图1是本发明实施例的蝗灾潜在高风险区识别方法的流程示意图；
图2是本发明实施例的蝗灾潜在高风险区识别装置的功能模块示意图；
图3是本发明实施例的计算机设备的结构示意图；
图4是本发明实施例的存储介质的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅是本申请的一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
本申请中的术语“第一”、“第二”、“第三”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。 由此，限定有“第一”、“第二”、“第三”的特征可以明示或者隐含地包括至少一个该特征。本申请的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。本申请实施例中所有方向性指示(诸如上、下、左、右、前、后……)仅用于解释在某一特定姿态(如附图所示)下各部件之间的相对位置关系、运动情况等，如果该特定姿态发生改变时，则该方向性指示也相应地随之改变。此外，术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。
图1是本发明实施例的蝗灾潜在高风险区识别方法的流程示意图。需注意的是，若有实质上相同的结果，本发明的方法并不以图1所示的流程顺序为限。如图1所示，该方法包括步骤：
步骤S101：基于预设方式获取研究区的长时间序列的多时相遥感影像数据。
具体地，本实施例中，通过获取NASA全球降雨测量(GPM)逐月数据，绘制研究区近5年月均降雨量时间序列曲线，确认研究区降雨发生明显变化的起始年份及结束年份，同时获取研究区一年中雨季开始及结束的月份，从而得到长时间序列的多时相遥感影像数据，以及该多时相遥感影像数据对应的影像获取时间段。
步骤S102：应用多时相遥感影像数据对研究区进行土地覆盖分类，并基于第一预设规则和多时相遥感影像数据计算研究区每个像元的生境适宜性指数。
具体地，在获取到多时相遥感影像数据后，对研究区进行土地覆盖 分类，再计算每个像元的生境适宜性指数。并且，在计算生境适宜性指数之前，对研究区内的多时相遥感影像数据按影像获取时间段进行筛选，筛选出影像含云量低于10％的n个时期的遥感影像。需要说明的是，遥感影像数据通常包括B1(海岸波段)、B2(蓝波段)、B3(绿波段)、B4(红波段)、B5(近红外波段)、B6(短波红外1波段)、B7(短波红外2波段)、B8(全色波段)、B9(卷云波段)、B10(热红外1波段)和B11(热红外2波段)。
其中，步骤S102中应用多时相遥感影像数据对研究区进行土地覆盖分类，并基于第一预设规则和多时相遥感影像数据计算研究区每个像元的生境适宜性指数的步骤，具体包括：
1、根据多时相遥感影像数据计算研究区的归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数。
具体地，本实施例中，利用上述筛选过的多时相遥感影像数据来计算研究区的归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数。
归一化差值植被指数的计算公式为：
NDVI＝(B5-B4)/(B5+B4)；
其中，NDVI表示归一化差值植被指数，无量纲，值域为[-1,1]；B4表示遥感影像数据红波段的亮度值；B5表示遥感影像数据近红外波段的亮度值。
归一化差值水体指数的计算公式为：
NDWI＝(B3-B5)/(B3+B5)；
其中，NDWI表示归一化差值水体指数，值域为[-1,1]；B3表示遥感影像数据绿波段的亮度值；B5表示遥感影像数据近红外波段的亮度值。
归一化差值建筑指数的计算公式为：
NDBI＝(B6-B5)/(B6+B5)；
其中，NDBI表示归一化差值建筑指数，无量纲，值域为[-1,1]；B6表示遥感影像数据短波红外波段的亮度值；B5表示遥感影像数据近红外 波段的亮度值。
建成区指数的计算公式为：
BUI＝NDBI–NDVI；
其中，BUI表示建成区指数，无量纲，值域为[-2,2]；NDBI表示归一化差值建筑指数，NDVI表示归一化差值植被指数。
土壤盐分指数的计算公式为：
Figure PCTCN2022137653-appb-000003
其中，SSI表示土壤盐分指数；B3表示遥感影像数据绿波段的亮度值；B4表示遥感影像数据红波段的亮度值。计算出SSI后对其进行归一化，即可得到研究区土壤盐分数据SS。
2、将归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数与多时相遥感影像数据进行波段合成，得到研究区的多时相遥感合成影像数据。
具体地，将计算得到的各时相的归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数与相应时相的遥感影像数据进行波段合成，并将各波段依次命名为NDVI、NDWI、NDBI和BUI，得到研究区的多时相遥感合成影像数据。
3、利用预设分类算法，根据多时相遥感合成影像数据对研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据。
具体地，在得到多时相遥感合成影像数据后，利用面向对象的随机森林(Random Forest)分类算法，对研究区的多时相遥感合成影像数据对各个像元按预设土地类型进行分类，其中预设土地类型包括耕地、草本湿地、芦苇和杂草、林地、水体、人工表面和其他7类土地覆盖类型数据。
进一步的，该利用预设分类算法，根据多时相遥感合成影像数据对研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据，包括：
3.1、获取预先构建并训练好的随机森林分类器，随机森林分类器根据预先准备的样本和预设土地类型训练得到。
3.2、将每个像元对应的多时相遥感合成影像数据输入至随机森林分类器，得到每个像元的土地覆盖类型数据。
需要说明的是，该随机森林分类器需要预先训练好，具体训练过程如下：
(1)分别选取研究区雨季开始期及结束期的多景遥感影像数据，通过目视解译的方式选取不同土地覆盖类型数据的样本；
(2)参与分类器训练及测试的波段包括B2、B3、B4、B5、B6、B7、B10、B11、NDVI、NDWI、NDBI和BUI共12个波段；
(3)构建包括m棵树的随机森林分类器，并应用选取的样本对其进行训练，本实施例中，m优选取值100；
(4)以尺度d规则选取种子点，在以B6、B5、B4波段合成的影像上应用简单非迭代聚类(Simple Non-Iterative Clustering,SNIC)分割算法对研究区的影像进行超像素分割，形成对象，并根据对象统计12个输入波段的均值，再根据实际需求分别设置SNIC分割算法的3个参数：紧凑度(Compactness)、连通性(Connectivity)和邻域尺寸(Neighborhood Size)。本实施例中，选取种子点的尺度d优选取值为5，SNIC分割算法的3个参数优选为0.5、8和256。
在得到训练好的随机森林分类器后，利用该随机森林分类器对研究区的多时相遥感合成影像数据进行面向对象的土地覆盖分类，得到每个像元的土地覆盖类型数据，并将该分类结果命名为土地覆盖类型数据LC。
4、根据归一化差值植被指数计算得到研究区的植被覆盖度。
具体地，植被覆盖度的计算公式为：
Fv＝(NDVI-NDVI soil)/(NDVI veg-NDVI soil)；
其中，Fv表示植被覆盖度，NDVI表示归一化差值植被指数，NDVI soil为完全是裸土或无植被覆盖像元的NDVI值，NDVI veg则代表完全被植被所覆盖的像元的NDVI值。
5、根据多时相遥感影像数据计算研究区的地表温度和温度植被干旱指数。
具体地，地表温度的计算公式为：
Figure PCTCN2022137653-appb-000004
其中，LST表示地表温度，单位为摄氏度，Tb为热红外通道中的TOA亮度温度，ε为热红外通道的表面发射率，系数A i、B i、C i由对10组总柱水蒸气(TCWV)(范围为0-6cm，步长0.6cm，TCWV值高于6cm为最后一组)执行的辐射传输模型的线性回归确定。
温度植被干旱指数的计算公式为：
Figure PCTCN2022137653-appb-000005
LST Smax＝a*NDVI+b，LST Smin＝c*NDVI+d；
其中，TVDI表示温度植被干旱指数，LST S代表某一像元的NDVI值为S时的地表温度，LST Smax代表当NDVI等于某一特定值S时地表温度的最大值，LST Smin代表当NDVI等于某一特定值S时地表温度的最小值，a、b、c、d为待定系数；TVDI的值域为[0,1]，值越大，土壤湿度越低，值越小，土壤湿度越高。本实施例中使用TVDI数据表征研究区的土壤湿度SM。
6、根据土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数对研究区中的每个像元进行评估，得到每个像元的生境适宜性指数，并基于生境适宜性指数确认每个像元对应的适宜性等级。
具体地，首先根据土壤盐分指数得到土壤盐分，以及根据温度植被干旱指数得到土壤湿度，再依据以上获取的土地覆盖类型数据LC、植被覆盖度Fv、地表温度LST、土壤盐分SS、土壤湿度SM等蝗虫生境影响因子数据，对研究区对东亚飞蝗的生境适宜性进行评估，从而得到每个像元的生境适宜性指数，并基于生境适宜性指数确认每个像元对应的适宜性等级。
其中，本实施例中，预先按照土地覆盖类型数据LC、植被覆盖度Fv、地表温度LST、土壤盐分SS、土壤湿度SM设定适宜性等级，依次划分为最适宜、次适宜、一般适宜和不适宜4个等级，并分别将相应等级数据赋值为4、3、2、1。具体地参见下表1(此为第一预设规则)：
表1
Figure PCTCN2022137653-appb-000006
需要说明的是，表中土壤盐度SS*和土壤湿度SM*为研究区的土壤盐分指数和温度植被干旱指数，两个指数为归一化数据，非土壤盐度及土壤湿度的绝对百分比。因数据获取时间或耕作种植状态的不同，不同研究区或不同时相下两个因子的适宜性等级划分会有不同，上表仅为本发明中某一时相的划分示例，实际使用时，两个因子的适宜性等级需根据研究区实际情况进行划分。
进一步的，根据土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数对研究区中的每个像元进行评估，得到每个像元的生境适宜性指数，包括：
6.1、构建基于土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数的适宜性等级，并为每个等级赋值。
6.2、基于斑块尺度原理，根据每个像元对应的土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数计算得到中心像元斑块尺度的适宜性等级。
具体地，本实施例充分考虑景观结构对蝗虫生境的影响，同时为了消除遥感数据的椒盐效应，本实施例引入了斑块尺度适宜性等级的概念，即以给定大小为w的窗口作为斑块，采用移动窗口法，引入相邻像元的 等级信息，实现中心像元斑块尺度适宜性等级的确定，计算公式如下：
Figure PCTCN2022137653-appb-000007
其中，M t,p(x (w+1)/2,y (w+1)/2)为生境因子t在斑块尺度的适宜性等级隶属度，t＝1、2、3、4、5，M 1,p、M 2,p、M 3,p、M 4,p和M 5,p分别代表土地覆盖类型数据LC、植被覆盖度Fv、地表温度LST、土壤盐度SS、土壤湿度SM 5个生境因子在斑块尺度的等级隶属度；w为移动窗口大小(为奇数)；x,y为研究区的行数及列数；j,k为窗口行数及列数；M t(x j,y k)代表因子M t位于(x j,y k)像元上的等级隶属性；d j,k为移动窗口中心像元(x (w+1)/2,y (w+1)/2)与相邻像元(x j,y k)的距离，其公式为：
Figure PCTCN2022137653-appb-000008
6.3、利用线性加权求和的方式，根据中心像元斑块尺度的适宜性等级计算得到每个像元的生境适宜性指数。
具体地，其计算公式如下：
Figure PCTCN2022137653-appb-000009
其中，HSI(x,y)是生境适宜性指数，值介于1～4；M t,p(x,y)是生境因子t在斑块尺度上的适宜性等级隶属度；n＝5；W t是生境因子t的权重系数，本实施例中，将土地覆盖类型数据LC、植被覆盖度Fv、地表温度LST、土壤盐度SS、土壤湿度SM 5个生境因子的权重分别取值为0.30、0.28、0.11、0.13和0.18。
进一步的，在得到生境适宜性指数后，根据该指数与指数阈值之间的关系设定各个像元的适宜性等级，具体等级设定条件如下(此为第二预设规则)：
Figure PCTCN2022137653-appb-000010
步骤S103：基于第二预设规则，利用多时相遥感影像数据计算得到研究区的水体变化指数和适宜生境变化指数，且当水体变化指数和适宜 生境变化指数满足第一预设条件时，确认研究区中存在蝗灾潜在高风险区。
具体地，水体变化指数表征随时间变化研究区水体面积的变化趋势和强度，适宜生境变化指数表征随时间变化研究区蝗虫适宜生境面积的变化趋势和强度。
其中，利用多时相遥感影像数据计算得到研究区的水体变化指数的步骤，包括：
A1、根据多时相遥感影像数据和土地覆盖类型数据获取研究区中水体在多时相遥感影像数据对应的时段内的最小面积、在多时相遥感影像数据起始时间点的第一面积和在多时相遥感影像数据结束时间点的第二面积。
A2、根据第一面积、第二面积和最小面积计算得到水体在多时相遥感影像数据对应的时段内的水体变化指数。
具体地，水体变化指数的计算公式如下：
Figure PCTCN2022137653-appb-000011
其中，δ WS为水体变化指数，其值域为(-1,0)，|δ WS|越接近1说明水体面积变化越剧烈，越接近0则说明变化越弱；tmin为研究区中水体的面积最小时所对应的时间，本实施例中，水体的面积由土地覆盖类型数据LC进行水体提取计算得出；t1为研究区多时相遥感影像数据所对应的起始时间，tn为研究区多时相遥感影像数据所对应的结束时间；WS tmin为研究区多时相遥感影像数据对应的时段内水体的最小面积，WS t1为多时相遥感影像数据起始时间点的水体的第一面积，WS tn为多时相遥感影像数据结束时间点的水体的第二面积。
其中，利用多时相遥感影像数据计算得到研究区的适宜生境变化指数的步骤，包括：
B1、根据多时相遥感影像数据和每个像元的适宜性等级获取研究区中蝗虫适宜生境在多时相遥感影像数据对应的时段内的最大面积、在多时相遥感影像数据起始时间点的第三面积和在多时相遥感影像数据结束时间点的第四面积。
B2、根据第三面积、第四面积和最大面积计算得到蝗虫适宜生境在多时相遥感影像数据对应的时段内的适宜生境变化指数。
具体地，适宜生境变化指数的计算公式为：
Figure PCTCN2022137653-appb-000012
其中，δ LHS为适宜生境变化指数，其值域为(-1,0)，|δ LHS|越接近1说明适宜生境面积变化越剧烈，越接近0则说明变化越弱；tmax为蝗虫适宜生境面积最大时所对应的时间，本实施例中，蝗虫适宜生境的面积为最适宜生境和次适宜生境的面积之和，最适宜生境和次适宜生境的面积之和根据上述确定的各像元的适宜性等级确定后，将最适宜的像元和次适宜的像元的面积相加得到；LHS tmax为研究区中蝗虫适宜生境在多时相遥感影像数据对应的时段内的最大面积，LHS t1为多时相遥感影像数据起始时间点的第三面积，LHS tn为多时相遥感影像数据结束时间点的第四面积。
进一步的，本实施例中，第一预设条件为：
Figure PCTCN2022137653-appb-000013
其中，tmin为水体面积最小时所对应的时间，tmax为蝗虫适宜生境面积最大时所对应的时间，t1为多时相遥感影像数据起始时间点，tn为多时相遥感影像数据结束时间点，WS为多时相遥感影像数据对应的时段内各个时间点对应的水体面积，LHS为多时相遥感影像数据对应的时段内各个时间点对应的蝗虫适宜生境面积，δ WS为水体变化指数，δ LHS为适宜生境变化指数，C 1为预设水体变化指数阈值，C 2为预设适宜生境变化指数阈值。
步骤S104：确认多时相遥感影像数据中最新一期影像的水体像元，并获取每个非水体像元与最近的水体像元之间的距离，且当存在目标非水体像元对应的生境适宜性指数和距离满足第二预设条件时，确认目标非水体像元对应的区域为蝗灾潜在高风险区。
具体地，在确定该研究区内存在蝗灾潜在高风险区后，通过获取每 个非水体像元与最近的水体像元之间的距离，在对每个非水体像元的距离和生境适宜性指数进行判断，从而确认该非水体像元对应的区域是否为蝗灾潜在高风险区。
其中，第二预设条件为：
Figure PCTCN2022137653-appb-000014
其中，D water为非水体像元与最近的水体像元之间的距离，HSI为像元的生境适宜性指数，D为预设距离阈值，C为预设生境适宜性指数阈值。本实施例中，将预设距离阈值D优选为2km，预设生境适宜性指数阈值C优选为3。
本发明实施例的蝗灾潜在高风险区识别方法通过利用研究区内长时间序列的多时相遥感影像数据对研究区水体和蝗虫适宜生境进行提取并分析得到研究区的水体变化指数和适宜生境变化指数，并当水体变化指数和适宜生境变化指数满足第一预设条件时，确认该研究区存在蝗灾潜在高风险区，再利用最新一期遥感影像数据计算得到的生境适宜性指数以及各个区域与研究区中水体之间的距离来确认研究区内的蝗灾潜在高风险区，其充分考虑了蝗虫生境由不适宜到适宜的演变过程，而且考虑了蝗虫聚集性暴发的特性，结合旱涝动态变化，实现了对蝗灾潜在高风险区的高精度识别。
图2是本发明实施例的蝗灾潜在高风险区识别装置的功能模块示意图。如图2所示，该装置20包括获取模块21、计算模块22、第一确认模块23和第二确认模块24。
获取模块21，用于基于预设方式获取研究区的长时间序列的多时相遥感影像数据；
计算模块22，用于应用多时相遥感影像数据对研究区进行土地覆盖分类，并基于第一预设规则和多时相遥感影像数据计算研究区每个像元的生境适宜性指数；第一确认模块23，用于基于第二预设规则，利用多时相遥感影像数据计算得到研究区的水体变化指数和适宜生境变化指数，且当水体变化指数和适宜生境变化指数满足第一预设条件时，确认研究区中存在蝗灾潜在高风险区；
第二确认模块24，用于确认多时相遥感影像数据中最新一期影像的水体像元，并获取每个非水体像元与最近的水体像元之间的距离，且当存在目标非水体像元对应的生境适宜性指数和距离满足第二预设条件时，确认目标非水体像元对应的区域为蝗灾潜在高风险区。
可选地，计算模块22执行基于第一预设规则和多时相遥感影像数据计算研究区每个像元的生境适宜性指数的操作，具体包括：根据多时相遥感影像数据计算研究区的归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数；将归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数与多时相遥感影像数据进行波段合成，得到研究区的多时相遥感合成影像数据；利用预设分类算法，根据多时相遥感合成影像数据对研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据；根据归一化差值植被指数计算得到研究区的植被覆盖度；根据多时相遥感影像数据计算研究区的地表温度和温度植被干旱指数；根据土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数对研究区中的每个像元进行评估，得到每个像元的生境适宜性指数，并基于生境适宜性指数确认每个像元对应的适宜性等级。
可选地，计算模块22执行利用预设分类算法，根据多时相遥感合成影像数据对研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据的操作，具体包括：获取预先构建并训练好的随机森林分类器，随机森林分类器根据预先准备的样本和预设土地类型训练得到；将每个像元对应的多时相遥感合成影像数据输入至随机森林分类器，得到每个像元的土地覆盖类型数据。
可选地，计算模块22执行根据土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数对研究区中的每个像元进行评估，得到每个像元的生境适宜性指数的操作，具体包括：构建基于土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分指数和温度植被干旱指数的适宜性等级，并为每个等级赋值；基于斑块尺度原理，根据每个像元对应的土地覆盖类型数据、植被覆盖度、地表温度、土壤盐分 指数和温度植被干旱指数计算得到中心像元斑块尺度的适宜性等级；利用线性加权求和的方式，根据中心像元斑块尺度的适宜性等级计算得到每个像元的生境适宜性指数。
可选地，第一确认模块23执行利用多时相遥感影像数据计算得到研究区的水体变化指数的操作，具体包括：根据多时相遥感影像数据和土地覆盖类型数据获取研究区中水体在多时相遥感影像数据对应的时段内的最小面积、在多时相遥感影像数据起始时间点的第一面积和在多时相遥感影像数据结束时间点的第二面积；根据第一面积、第二面积和最小面积计算得到水体在多时相遥感影像数据对应的时段内的水体变化指数。
可选地，第一确认模块23执行利用多时相遥感影像数据计算得到研究区的适宜生境变化指数的操作，具体包括：根据多时相遥感影像数据和每个像元的适宜性等级获取研究区中蝗虫适宜生境在多时相遥感影像数据对应的时段内的最大面积、在多时相遥感影像数据起始时间点的第三面积和在多时相遥感影像数据结束时间点的第四面积；根据第三面积、第四面积和最大面积计算得到蝗虫适宜生境在多时相遥感影像数据对应的时段内的适宜生境变化指数。
可选地，第一预设条件为：
Figure PCTCN2022137653-appb-000015
其中，tmin为水体面积最小时所对应的时间，tmax为蝗虫适宜生境面积最大时所对应的时间，t1为多时相遥感影像数据起始时间点，tn为多时相遥感影像数据结束时间点，WS为多时相遥感影像数据对应的时段内各个时间点对应的水体面积，LHS为多时相遥感影像数据对应的时段内各个时间点对应的蝗虫适宜生境面积，δ WS为水体变化指数，δ LHS为适宜生境变化指数，C 1为预设水体变化指数阈值，C 2为预设适宜生境变化指数阈值
第二预设条件为：
Figure PCTCN2022137653-appb-000016
其中，D water为非水体像元与最近的水体像元之间的距离，HSI为像元的生境适宜性指数，D为预设距离阈值，C为预设生境适宜性指数阈值。
关于上述实施例蝗灾潜在高风险区识别装置中各模块实现技术方案的其他细节，可参见上述实施例中的蝗灾潜在高风险区识别方法中的描述，此处不再赘述。
需要说明的是，本说明书中的各个实施例均采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似的部分互相参见即可。对于装置类实施例而言，由于其与方法实施例基本相似，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
请参阅图3，图3为本发明实施例的计算机设备的结构示意图。如图3所示，该计算机设备60包括处理器61及和处理器61耦接的存储器62，存储器62中存储有程序指令，程序指令被处理器61执行时，使得处理器61执行上述任一实施例所述的蝗灾潜在高风险区识别方法的步骤。
其中，处理器61还可以称为CPU(Central Processing Unit，中央处理单元)。处理器61可能是一种集成电路芯片，具有信号的处理能力。处理器61还可以是通用处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、现场可编程门阵列(FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
参阅图4，图4为本发明实施例的存储介质的结构示意图。本发明实施例的存储介质存储有能够实现上述所有方法的程序指令71，其中，该程序指令71可以以软件产品的形式存储在上述存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)或处理器(processor)执行本申请各个实施方式所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access  Memory)、磁碟或者光盘等各种可以存储程序代码的介质，或者是计算机、服务器、手机、平板等计算机设备设备。
在本申请所提供的几个实施例中，应该理解到，所揭露的计算机设备，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。以上仅为本申请的实施方式，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (10)

  1. 一种蝗灾潜在高风险区识别方法，其特征在于，包括：
    基于预设方式获取研究区的长时间序列的多时相遥感影像数据；
    应用所述多时相遥感影像数据对研究区进行土地覆盖分类，并基于第一预设规则和所述多时相遥感影像数据计算研究区每个像元的生境适宜性指数；
    基于第二预设规则，利用所述多时相遥感影像数据计算得到所述研究区的水体变化指数和适宜生境变化指数，且当所述水体变化指数和所述适宜生境变化指数满足第一预设条件时，确认所述研究区中存在蝗灾潜在高风险区；
    确认所述多时相遥感影像数据中最新一期影像的水体像元，并获取每个非水体像元与最近的水体像元之间的距离，且当存在目标非水体像元对应的生境适宜性指数和距离满足第二预设条件时，确认所述目标非水体像元对应的区域为蝗灾潜在高风险区。
  2. 根据权利要求1所述的蝗灾潜在高风险区识别方法，其特征在于，所述基于第一预设规则和所述多时相遥感影像数据计算研究区每个像元的生境适宜性指数，包括：
    根据所述多时相遥感影像数据计算所述研究区的归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数；
    将所述归一化差值植被指数、归一化差值水体指数、归一化差值建筑指数、建成区指数和土壤盐分指数与所述多时相遥感影像数据进行波段合成，得到所述研究区的多时相遥感合成影像数据；
    利用预设分类算法，根据所述多时相遥感合成影像数据对所述研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据；
    根据所述归一化差值植被指数计算得到所述研究区的植被覆盖度；
    根据所述多时相遥感影像数据计算所述研究区的地表温度和温度 植被干旱指数；
    根据所述土地覆盖类型数据、所述植被覆盖度、所述地表温度、所述土壤盐分指数和所述温度植被干旱指数对所述研究区中的每个像元进行评估，得到每个像元的所述生境适宜性指数，并基于所述生境适宜性指数确认每个像元对应的适宜性等级。
  3. 根据权利要求2所述的蝗灾潜在高风险区识别方法，其特征在于，所述利用预设分类算法，根据所述多时相遥感合成影像数据对所述研究区中每个像元按预设土地类型进行分类，得到每个像元的土地覆盖类型数据，包括：
    获取预先构建并训练好的随机森林分类器，所述随机森林分类器根据预先准备的样本和所述预设土地类型训练得到；
    将每个像元对应的多时相遥感合成影像数据输入至所述随机森林分类器，得到每个像元的土地覆盖类型数据。
  4. 根据权利要求2所述的蝗灾潜在高风险区识别方法，其特征在于，根据所述土地覆盖类型数据、所述植被覆盖度、所述地表温度、所述土壤盐分指数和所述温度植被干旱指数对所述研究区中的每个像元进行评估，得到每个像元的所述生境适宜性指数，包括：
    构建基于所述土地覆盖类型数据、所述植被覆盖度、所述地表温度、所述土壤盐分指数和所述温度植被干旱指数的适宜性等级，并为每个等级赋值；
    基于斑块尺度原理，根据每个像元对应的所述土地覆盖类型数据、所述植被覆盖度、所述地表温度、所述土壤盐分指数和所述温度植被干旱指数计算得到中心像元斑块尺度的适宜性等级；
    利用线性加权求和的方式，根据所述中心像元斑块尺度的适宜性等级计算得到每个像元的所述生境适宜性指数。
  5. 根据权利要求2所述的蝗灾潜在高风险区识别方法，其特征在于，利用所述多时相遥感影像数据计算得到所述研究区的水体变化指数的步骤，包括：
    根据所述多时相遥感影像数据和所述土地覆盖类型数据获取所述 研究区中水体在所述多时相遥感影像数据对应的时段内的最小面积、在所述多时相遥感影像数据起始时间点的第一面积和在所述多时相遥感影像数据结束时间点的第二面积；
    根据所述第一面积、所述第二面积和所述最小面积计算得到水体在所述多时相遥感影像数据对应的时段内的水体变化指数。
  6. 根据权利要求5所述的蝗灾潜在高风险区识别方法，其特征在于，利用所述多时相遥感影像数据计算得到所述研究区的适宜生境变化指数的步骤，包括：
    根据所述多时相遥感影像数据和每个像元的所述适宜性等级获取所述研究区中蝗虫适宜生境在所述多时相遥感影像数据对应的时段内的最大面积、在所述多时相遥感影像数据起始时间点的第三面积和在所述多时相遥感影像数据结束时间点的第四面积；
    根据所述第三面积、所述第四面积和所述最大面积计算得到所述蝗虫适宜生境在所述多时相遥感影像数据对应的时段内的适宜生境变化指数。
  7. 根据权利要求6所述的蝗灾潜在高风险区识别方法，其特征在于，所述第一预设条件为：
    Figure PCTCN2022137653-appb-100001
    其中，tmin为水体面积最小时所对应的时间，tmax为蝗虫适宜生境面积最大时所对应的时间，t1为所述多时相遥感影像数据起始时间点，tn为所述多时相遥感影像数据结束时间点，WS为所述多时相遥感影像数据对应的时段内各个时间点对应的水体面积，LHS为所述多时相遥感影像数据对应的时段内各个时间点对应的蝗虫适宜生境面积，δ WS为水体变化指数，δ LHS为适宜生境变化指数，C 1为预设水体变化指数阈值，C 2为预设适宜生境变化指数阈值
    所述第二预设条件为：
    Figure PCTCN2022137653-appb-100002
    其中，D water为非水体像元与最近的水体像元之间的距离，HSI为像元的生境适宜性指数，D为预设距离阈值，C为预设生境适宜性指数阈值。
  8. 一种蝗灾潜在高风险区识别装置，其特征在于，包括：
    获取模块，用于基于预设方式获取研究区的长时间序列的多时相遥感影像数据；
    计算模块，用于应用所述多时相遥感影像数据对研究区进行土地覆盖分类，并基于第一预设规则和所述多时相遥感影像数据计算研究区每个像元的生境适宜性指数；
    第一确认模块，用于基于第二预设规则，利用所述多时相遥感影像数据计算得到所述研究区的水体变化指数和适宜生境变化指数，且当所述水体变化指数和所述适宜生境变化指数满足第一预设条件时，确认所述研究区中存在蝗灾潜在高风险区；
    第二确认模块，用于确认所述多时相遥感影像数据中最新一期影像的水体像元，并获取每个非水体像元与最近的水体像元之间的距离，且当存在目标非水体像元对应的生境适宜性指数和距离满足第二预设条件时，确认所述目标非水体像元对应的区域为蝗灾潜在高风险区。
  9. 一种计算机设备，其特征在于，所述计算机设备包括处理器、与所述处理器耦接的存储器，所述存储器中存储有程序指令，所述程序指令被所述处理器执行时，使得所述处理器执行如权利要求1-7中任一项权利要求所述的蝗灾潜在高风险区识别方法的步骤。
  10. 一种存储介质，其特征在于，存储有能够实现如权利要求1-7中任一项所述的蝗灾风险预测方法的程序指令。
PCT/CN2022/137653 2021-12-23 2022-12-08 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质 Ceased WO2023116454A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111589159.8A CN114419431B (zh) 2021-12-23 2021-12-23 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质
CN202111589159.8 2021-12-23

Publications (1)

Publication Number Publication Date
WO2023116454A1 true WO2023116454A1 (zh) 2023-06-29

Family

ID=81267404

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/137653 Ceased WO2023116454A1 (zh) 2021-12-23 2022-12-08 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN114419431B (zh)
WO (1) WO2023116454A1 (zh)

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116823576A (zh) * 2023-06-30 2023-09-29 北京极嗅科技有限公司 一种毒品原植物适生区的评估方法及系统
CN116994133A (zh) * 2023-07-22 2023-11-03 福州大学 一种基于多源遥感数据的陆地植被响应干旱的评估方法
CN117115672A (zh) * 2023-07-22 2023-11-24 农业农村部大数据发展中心 一种基于卫星遥感的涝灾玉米提取方法
CN117275208A (zh) * 2023-11-13 2023-12-22 广东天顺为信息科技有限公司 农业生物灾害监测预警信息化应用系统
CN117451968A (zh) * 2023-09-12 2024-01-26 中国农业大学 一种基于遥感数据融合的农田根层土壤含盐量反演方法及应用
CN117576572A (zh) * 2024-01-16 2024-02-20 杭州稻道农业科技有限公司 一种综合种养稻田水稻种植覆盖度提取方法、装置及介质
CN118155001A (zh) * 2024-05-11 2024-06-07 中国电建集团西北勘测设计研究院有限公司 评估风电场对土地覆盖影响的方法、装置、设备及介质
CN118226434A (zh) * 2024-03-19 2024-06-21 中国科学院地理科学与资源研究所 基于物候学的整合光学和雷达数据的全球水稻制图方法
CN118447398A (zh) * 2024-05-07 2024-08-06 中国科学院地理科学与资源研究所 盐田分类方法、装置、电子设备和计算机可读存储介质
CN118797317A (zh) * 2024-09-13 2024-10-18 江西省生态环境科学研究与规划院 一种确定区域干旱影响最佳时间尺度的方法
CN118799726A (zh) * 2024-06-17 2024-10-18 航天宏图信息技术股份有限公司 基于气象卫星遥感数据的冰川提取方法、装置及设备
CN119091297A (zh) * 2024-08-19 2024-12-06 福建师范大学 基于多时相数据的土地覆盖变化与热环境影响研究方法
CN119251706A (zh) * 2024-12-06 2025-01-03 绍兴文理学院 一种基于分段任意模型的疟疾高风险区域识别方法及装置
CN119580094A (zh) * 2024-11-19 2025-03-07 兰州大学 一种基于时序优化的大尺度动态不透水面覆盖度估算方法
CN119600460A (zh) * 2024-11-05 2025-03-11 潍坊市海洋发展研究院 一种基于大数据的海岸信息评估方法、装置、设备及介质
CN119939523A (zh) * 2025-04-03 2025-05-06 厦门城市职业学院(厦门开放大学) 一种多源数据融合的植被覆盖率预测方法及系统
CN120032256A (zh) * 2025-04-24 2025-05-23 中国科学院生态环境研究中心 一种梯田格局演变量化及其生态系统服务权衡方法
CN120408467A (zh) * 2025-07-02 2025-08-01 易维集控(北京)园林科技有限公司 基于多源遥感数据融合的植被环境耦合预测方法及系统
CN120526184A (zh) * 2025-07-23 2025-08-22 内蒙古医科大学 基于遥感与gis的布鲁氏菌病风险监测及预警方法

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114419431B (zh) * 2021-12-23 2024-07-19 深圳先进技术研究院 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质
CN115983500B (zh) * 2023-03-06 2023-05-30 中国科学院空天信息创新研究院 一种预测沙漠蝗的方法及装置
CN116665046A (zh) * 2023-05-19 2023-08-29 中国科学院城市环境研究所 一种植被物候变化趋势评价方法、系统、设备及介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103955606A (zh) * 2014-04-23 2014-07-30 北京大学 一种基于遥感技术的草原蝗灾渐进式预测方法
CN105389559A (zh) * 2015-11-12 2016-03-09 中国科学院遥感与数字地球研究所 基于高分辨率遥感影像的农业灾害范围识别系统及方法
CN114419431A (zh) * 2021-12-23 2022-04-29 深圳先进技术研究院 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104504279B (zh) * 2014-12-31 2017-09-15 中国科学院深圳先进技术研究院 检测蝗灾的方法
CN106682624B (zh) * 2016-12-29 2019-08-02 中国科学院深圳先进技术研究院 基于时间序列遥感信息的建成区提取方法及装置
CN110687049B (zh) * 2018-07-04 2022-12-20 中国科学院深圳先进技术研究院 一种用材林识别方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103955606A (zh) * 2014-04-23 2014-07-30 北京大学 一种基于遥感技术的草原蝗灾渐进式预测方法
CN105389559A (zh) * 2015-11-12 2016-03-09 中国科学院遥感与数字地球研究所 基于高分辨率遥感影像的农业灾害范围识别系统及方法
CN114419431A (zh) * 2021-12-23 2022-04-29 深圳先进技术研究院 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
GENG YUN, DONG YINGYING;HUANG WENJIANG;ZHAO LONGLONG;TU XIONGBING;LI HONGMEI: "Dynamic remote sensing monitoring of oriental migratory locust habitats in Dagang reservoir, Tianjin", JOURNAL OF PLANT PROTECTION, vol. 45, no. 1, 7 February 2021 (2021-02-07), pages 122 - 128, XP093075525, DOI: 10.13802/j.cnki.zwbhxb.2021.2021815 *
HUANG JIANXI;ZHUO WEN;YANG CHUNXI;LI LIN;ZHANG CHAO;LIU JIA: "Locust Remote Sensing Monitoring Methods Based on Landsat8 Satellite Data", TRANSACTIONS OF THE CHINESE SOCIETY FOR AGRICULTURAL MACHINERY, vol. 46, no. 5, 25 December 2014 (2014-12-25), pages 258 - 264, XP093075522, ISSN: 1000-1298, DOI: 10.6041/j.issn.1000-1298.2015.05.037 *
MENG QINGHUI, CHEN JIAN,SHENG SHIJIE,LIU JIANQI: "Oriental Migratory Locust Habitat Classification based on Multi-temporal Remote Sensing Data", REMOTE SENSING TECHNOLOGY AND APPLICATION, vol. 28, no. 1, 15 February 2013 (2013-02-15), pages 116 - 121, XP093075519, ISSN: 0004-0323 *

Cited By (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116823576A (zh) * 2023-06-30 2023-09-29 北京极嗅科技有限公司 一种毒品原植物适生区的评估方法及系统
CN116823576B (zh) * 2023-06-30 2023-12-26 北京极嗅科技有限公司 一种毒品原植物适生区的评估方法及系统
CN116994133A (zh) * 2023-07-22 2023-11-03 福州大学 一种基于多源遥感数据的陆地植被响应干旱的评估方法
CN117115672A (zh) * 2023-07-22 2023-11-24 农业农村部大数据发展中心 一种基于卫星遥感的涝灾玉米提取方法
CN117451968A (zh) * 2023-09-12 2024-01-26 中国农业大学 一种基于遥感数据融合的农田根层土壤含盐量反演方法及应用
CN117451968B (zh) * 2023-09-12 2026-04-17 中国农业大学 一种基于遥感数据融合的农田根层土壤含盐量反演方法及应用
CN117275208A (zh) * 2023-11-13 2023-12-22 广东天顺为信息科技有限公司 农业生物灾害监测预警信息化应用系统
CN117275208B (zh) * 2023-11-13 2024-02-27 广东天顺为信息科技有限公司 农业生物灾害监测预警信息化应用系统
CN117576572A (zh) * 2024-01-16 2024-02-20 杭州稻道农业科技有限公司 一种综合种养稻田水稻种植覆盖度提取方法、装置及介质
CN118226434A (zh) * 2024-03-19 2024-06-21 中国科学院地理科学与资源研究所 基于物候学的整合光学和雷达数据的全球水稻制图方法
CN118447398A (zh) * 2024-05-07 2024-08-06 中国科学院地理科学与资源研究所 盐田分类方法、装置、电子设备和计算机可读存储介质
CN118155001A (zh) * 2024-05-11 2024-06-07 中国电建集团西北勘测设计研究院有限公司 评估风电场对土地覆盖影响的方法、装置、设备及介质
CN118799726A (zh) * 2024-06-17 2024-10-18 航天宏图信息技术股份有限公司 基于气象卫星遥感数据的冰川提取方法、装置及设备
CN119091297A (zh) * 2024-08-19 2024-12-06 福建师范大学 基于多时相数据的土地覆盖变化与热环境影响研究方法
CN118797317A (zh) * 2024-09-13 2024-10-18 江西省生态环境科学研究与规划院 一种确定区域干旱影响最佳时间尺度的方法
CN119600460A (zh) * 2024-11-05 2025-03-11 潍坊市海洋发展研究院 一种基于大数据的海岸信息评估方法、装置、设备及介质
CN119600460B (zh) * 2024-11-05 2025-12-02 潍坊市海洋发展研究院 一种基于大数据的海岸信息评估方法、装置、设备及介质
CN119580094A (zh) * 2024-11-19 2025-03-07 兰州大学 一种基于时序优化的大尺度动态不透水面覆盖度估算方法
CN119251706A (zh) * 2024-12-06 2025-01-03 绍兴文理学院 一种基于分段任意模型的疟疾高风险区域识别方法及装置
CN119939523A (zh) * 2025-04-03 2025-05-06 厦门城市职业学院(厦门开放大学) 一种多源数据融合的植被覆盖率预测方法及系统
CN120032256A (zh) * 2025-04-24 2025-05-23 中国科学院生态环境研究中心 一种梯田格局演变量化及其生态系统服务权衡方法
CN120408467A (zh) * 2025-07-02 2025-08-01 易维集控(北京)园林科技有限公司 基于多源遥感数据融合的植被环境耦合预测方法及系统
CN120526184A (zh) * 2025-07-23 2025-08-22 内蒙古医科大学 基于遥感与gis的布鲁氏菌病风险监测及预警方法

Also Published As

Publication number Publication date
CN114419431B (zh) 2024-07-19
CN114419431A (zh) 2022-04-29

Similar Documents

Publication Publication Date Title
WO2023116454A1 (zh) 一种蝗灾潜在高风险区识别方法、装置、设备及存储介质
US12444189B2 (en) Methods and systems for classifying and benchmarking irrigation performance
US10909368B2 (en) Crop type classification in images
US9489576B2 (en) Crop stand analysis
CN104089647B (zh) 一种作物病害发生范围监测方法及系统
US20220012385A1 (en) Computer-implemented methods, computer readable medium and systems for a precision agriculture platform that detects healthy conditions
Beyer et al. Rainfall characteristics and their implications for rain-fed agriculture: a case study in the Upper Zambezi River Basin
CN111738066A (zh) 综合多源遥感信息的网格化水稻纹枯病生境评价方法
Shahrin et al. Agricultural analysis and crop yield prediction of habiganj using multispectral bands of satellite imagery with machine learning
CN110991714A (zh) 一种农作物倒伏灾情监测方法及系统
CN107437262B (zh) 作物种植面积预警方法和系统
CN116842351B (zh) 滨海湿地碳汇评估模型构建方法、评估方法和电子设备
CN113570273A (zh) 一种灌溉耕地统计数据的空间化方法及系统
Rouze et al. Evaluating unoccupied aerial systems (UAS) imagery as an alternative tool towards cotton-based management zones
Rulinda et al. Image mining for drought monitoring in eastern Africa using Meteosat SEVIRI data
Mcfarland et al. Evaluation of NDVI to assess avian abundance and richness along the upper San Pedro River
CN115420688A (zh) 一种基于物联网的农业灾害信息遥感提取损失评估方法
CN118521969B (zh) 一种水稻退种风险的监测方法
Albano et al. Status and Trends of Groundwater Dependent Vegetation in Relation to Climate and Shallow Groundwater in the Harney Basin, Oregon
Duan et al. Monitoring apple flowering date at 10 m spatial resolution based on crop reference curves
CN119514893A (zh) 一种耕地撂荒风险综合评估和预测方法、设备和介质
Wilkinson et al. Change detection techniques for use in a statewide forest inventory program
Mongkolnithithada et al. Rice yield estimation based on machine learning approaches using MODIS 250 m data
Fenetahun et al. Mapping and Modeling Aboveground Biomass Dynamics in Savanna Grasslands Southern Ethiopia
Nabizada et al. Assessing spatiotemporal dynamics of meteorological and agricultural drought in the North Basin of Afghanistan using multiple remote sensing-based drought indices

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22909769

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22909769

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 22909769

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 17/01/2025)