JP7229864B2 - REMOTE SENSING IMAGE ACQUISITION TIME DETERMINATION SYSTEM AND CROPT GROWTH ANALYSIS METHOD - Google Patents

Description

The present invention relates to a technique for analyzing or estimating the growth status of crops based on remote sensing images. In particular, the present invention relates to a technique for determining image acquisition timing necessary for analyzing or estimating the growth status of crops using image information.

It is known to use remote sensing images such as satellite images to calculate a vegetation index by utilizing the characteristics of light reflection by plants and to grasp the state of vegetation. For example, in the background art section of Patent Document 1, using high-resolution satellite data, local data before harvest (before January to February) (harvest amount data, soil sampling data, temperature data, precipitation data, etc.), By using high-resolution satellite images taken at that time together, vegetation indicators (NDVI, EVI, LAI, etc.) measured from satellite images are put into statistical models (linear prediction models, mixed models, etc.) to estimate crop yields. stated to be done.

NDVI (Normalized Difference Vegetation Index), EVI (Enhanced Vegetation Index), and LAI (Estimation of Leaf Area Index) are well-known examples of vegetation indices.

Patent Publication No. 2015-188333

When grasping the growth status of crops from remote sensing images, it is desirable to photograph at the time when there are many fields (hereinafter referred to as the "optimal period") at the time when the expected growth stage is reached.

For example, when estimating the yield of paddy rice, it is ideal that all the fields included in the photographing area reach maturity. Ripening is the gradual growth and enlargement of grain seeds. In general, rice begins to ripen when the ear emerges, blooms, and is pollinated. In the case of rice, the ripening period is about 40 to 50 days after flowering.

However, in reality, since the growing conditions are different for each field, the timing of the ripening stage varies. In order to estimate the yield of paddy rice, it is desirable to acquire remote sensing images at the time when as many fields as possible included in the photographing area reach maturity.

In order to acquire the remote sensing image at the optimum time, it is conceivable to take images periodically and then extract the image at the favorable time after taking the image. However, since high-resolution satellite data suitable for high-precision analysis and estimation are expensive, a method that assumes acquisition of a large number of high-resolution images leads to an increase in cost.

It is also conceivable that a person actually goes to the site to investigate and determine the optimum period, but there is also the problem of human cost.

Therefore, there is a demand for a method that enables the capture of remote sensing images for analysis of the growing conditions of crops at the optimum time while keeping costs down as much as possible.

A preferred aspect of the present invention includes an input device, an output device, a processing device, and a storage device, and determines the acquisition timing of remote sensing images for grasping the growth status of agricultural crops for agricultural fields distributed on the ground surface. It is a system that This system includes a reference date estimating unit that estimates a reference date used for predicting the growth state of crops based on the first remote sensing image, which is time-series data. It also includes a growth prediction unit that predicts the growth state of crops after a reference date based on meteorological data, which is chronological data, and estimates when the crops will reach a specific growth stage. The apparatus also includes an appropriate photographing timing determination unit that determines the timing at which the maximum number of images of a field in which crops have reached a specific growth stage can be photographed as the acquisition timing of the second remote sensing image.

Another preferred aspect of the present invention is to use an input device, an output device, a processing device, and a storage device, and use a remote sensing image to grasp the growth status of agricultural crops in fields distributed on the ground surface. It is a situation analysis method. In this method, a reference date used for predicting the growth state of crops is estimated based on the first remote sensing image, which is time-series data. Also, in this method, based on meteorological data, which is chronological data, the growth status of crops after a reference date is predicted, and the timing at which crops reach a specific growth stage is estimated. Also, in this method, the timing at which the maximum number of fields in which crops have reached the specific growth stage can be photographed is determined as the acquisition timing for the second remote sensing image.

In a more specific form of the present invention, a vegetation index is calculated from the second remote sensing image, and the vegetation index is applied to a predetermined statistical model to analyze the growth status of agricultural crops.

To provide a method for capturing remote sensing images for crop growth analysis during the optimum period.

The block diagram which shows the structural example of a crop growth condition analysis system. FIG. 2 is a schematic diagram showing a configuration example of satellite image data; Schematic diagram showing a configuration example of weather data. FIG. 4 is a schematic diagram showing a configuration example of GIS data; The flow chart of the whole processing in a crop growth condition analysis system. A detailed flow chart of the reference date estimation process. Detailed flowchart of growth prediction processing. The conceptual diagram explaining the weather data change prediction formula determination method in growth prediction processing. FIG. 10 is a conceptual diagram for explaining a method of determining an appropriate shooting time in an appropriate shooting time determination process;

Embodiments will be described in detail with reference to the drawings. However, the present invention should not be construed as being limited to the description of the embodiments shown below. Those skilled in the art will easily understand that the specific configuration can be changed without departing from the idea or gist of the present invention.

In the configuration of the invention described below, the same reference numerals may be used in common for the same parts or parts having similar functions between different drawings, and redundant description may be omitted.

When there are a plurality of elements having the same or similar functions, they may be described with the same reference numerals and different suffixes. However, if there is no need to distinguish between multiple elements, the subscripts may be omitted.

Notations such as “first”, “second”, “third” in this specification etc. are attached to identify the constituent elements, and do not necessarily limit the number, order, or content thereof isn't it. Also, numbers for identifying components are used for each context, and numbers used in one context do not necessarily indicate the same configuration in other contexts. Also, it does not preclude a component identified by a certain number from having the function of a component identified by another number.

The position, size, shape, range, etc. of each configuration shown in the drawings, etc. may not represent the actual position, size, shape, range, etc., in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the positions, sizes, shapes, ranges, etc. disclosed in the drawings and the like.

All publications, patents and patent applications cited herein are hereby incorporated by reference into this description.

Elements presented herein in the singular shall include the plural unless the context clearly dictates otherwise.

<1. Overview of Crop Growth Analysis System>
FIG. 1 is a block diagram showing an overview of a crop growth analysis system according to one embodiment of the invention. The crop growth analysis system according to the present embodiment includes a crop growth analysis device 100 as a component.

The crop growing condition analysis system of this embodiment includes a function of analyzing the growing condition of crops and a function of determining when to acquire images for analysis. It is also possible to configure the portion of the remote sensing image acquisition timing determination system separately.

The crop growth condition analysis device 100 is composed of, for example, a server which is a type of information processing device, and includes a CPU (Central Processing Unit) 101, a memory 102 such as a semiconductor memory, an auxiliary storage device 103 such as a magnetic disk device, and an input device. An output device 104 is included. In addition, the server may include devices generally used.

The memory 102 stores a reference date estimation unit 105, a growth prediction unit 106, an appropriate photographing period determination unit 107, and a crop growth condition analysis unit 108 as programs. The program can cause the information processing device to function as each part of the crop growth situation analysis device 100 . The CPU 101 can, for example, read a program from the auxiliary storage device 103 to the memory 102 and execute it.

In this embodiment, functions such as calculation and control are realized by executing a program stored in the memory 102 by the CPU 101, thereby performing predetermined processing in cooperation with other hardware. A program executed by the CPU 101, its function, or a means for realizing the function may be called a "function," "means," "unit," "unit," "module," or the like.

In the present embodiment, functions equivalent to those configured by software can also be realized by hardware such as FPGA (Field Programmable Gate Array) and ASIC (Application Specific Integrated Circuit).

The crop growth condition analysis device 100 uses satellite image data 110, weather data 120, and GIS (Geographic Information System) data 130 as data to be processed.

The satellite image data 110 is time-series image data, and is, for example, a color still image captured using an earth observation satellite 140 or the like. The photographing range of the satellite image data 110 includes a region (hereinafter referred to as “target region”) in which there is a field for which crop growth status is to be grasped.

The weather data 120 is time-series weather data, and includes data selected from the temperature, amount of rainfall, hours of sunshine, etc. in the target area.

A GIS (Geographic Information System) is data relating to topography in a target area, and includes, for example, elevation and land cover classification map data.

Using the satellite image data 110, the weather data 120, and the GIS data 130, the crop growth analysis apparatus 100 uses the time when the number of fields in which crops are in a specific growth stage is the largest in the target area as the appropriate time for taking remote sensing images. decide. Hereinafter, the term "appropriate time for imaging" refers to the time when the system of the embodiment estimates that the imaging region is suitable for imaging as the optimal time.

The configuration of the crop growth condition analysis apparatus 100 may be configured as a single server, or alternatively, any portion of the CPU 101, memory 102, auxiliary storage device 103, and input/output device 104 may be connected via a network. computer.

The satellite image data 110, the weather data 120, and the GIS data 130 may be managed by the same administrator as the crop growth analysis device 100, and may be stored as a database in the auxiliary storage device 103, for example. Also, these data may be obtained via the input/output device 104 from an external database managed by a third party connected via a network, for example.

<2. Satellite image data>
FIG. 2 is a schematic diagram showing a configuration example of the satellite image data 110. As shown in FIG. The satellite image data 110 is an image of a target area continuously and periodically taken by an earth observation satellite 140, for example. The satellite image data 110 includes a plurality of still images 201 of the target area captured on different dates. A specific frame in a moving image may be used as the still image.

For continuous, periodic imaging, relatively low-resolution images, which can be obtained at low cost, are preferred over relatively high-resolution images, which are suitable for high-precision analysis and estimation. preferable.

The black square marks 202 indicated by A to F in FIG. 2 represent a representative part of the fields included in the target area to be photographed in order to finally estimate the yield. , and areas A to F having approximately the same area (hereinafter referred to as "target farm area" or simply "area"). The target field district may be determined in advance by the user, or may be automatically determined by the crop growth analysis system as described later. The number of target farm districts is arbitrary.

<3. Meteorological Data>
FIG. 3 is a schematic diagram showing the structure of the weather data 120. As shown in FIG. The meteorological data 120 is basically time-series data, and meteorological data such as time, temperature, amount of precipitation, wind direction, wind speed, hours of sunshine, snow depth, etc. for each observation day for an area including the target area are displayed in time series. It is aggregated in These meteorological data are examples, and only some of the above may be used, or other types of data may be added.

Also, as shown in FIG. 3, weather data corresponding to the target agricultural field areas A to F are prepared. In general, the geographic granularity of weather data provided, for example, by public institutions is different from the granularity of the target field area. Therefore, in this embodiment, data corresponding to the position or coordinates of the target field district is extracted from meteorological data prepared by a known technique, and is used by associating it with each district.

<4. GIS data>
FIG. 4 is a schematic diagram showing the structure of the GIS data 130. As shown in FIG. A digital elevation model 131 and a land cover classification map 132 are included for the area containing the target area.

The digital elevation model 131 is data in which numerical values of elevation (e.g. sea level) are recorded corresponding to coordinates of latitude and longitude, for example.

Also, the land cover classification map 132 classifies the land cover status into objects such as paddy fields, upland fields, bare land, and water areas. This is recorded data. The land cover classification map 132 can be created, for example, by using satellite images and utilizing the fact that the spectral reflection characteristics differ depending on the object.

In the present embodiment, data corresponding to the positions or coordinates of the target agricultural field areas A to F are extracted from GIS data prepared by a known method and used.

<5. Overall Processing Flow of Image Acquisition Timing Estimation>
FIG. 5 is a processing flow chart of the crop growth situation grasping system. First, in the target field area setting process S501, of all the target fields to be photographed for the final estimation of yield, some representative fields are selected from areas A to F having approximately the same area. sample as Although the areas do not have to be exactly the same, it is desirable that they are uniform within about ±10%. Moreover, it is not necessary to include all farm fields to be observed, and the number of farm fields included in areas A to F need not be the same.

In areas A to F, when fields and features other than fields such as buildings coexist, depending on the resolution of the satellite image data 110, the fields and features other than fields may coexist within one pixel. Index values do not represent exact crop growth conditions. Therefore, it is desirable not to include features other than agricultural fields in each district.

Areas A to F to be sampled may be selected by the user while looking at the digital elevation model 131 and the land cover classification map 132, for example. In selecting the areas A to F, it is desirable that the variations in the growing conditions of the crops in the areas A to F sufficiently cover the variations in the growing conditions in all fields to be photographed.

When the crop growth situation analysis system automatically selects districts A to F, the reference date estimating unit 105 calculates the weather and topography of the selected districts A to F based on the weather data 120 and the GIS data 130. The selection should be made so that the variation covers as much as possible the variation in all fields to be photographed. For calculation by the system, for example, it may be selected so that the distribution of the data of districts A to F is large (for example, it is maximum or exceeds a predetermined threshold). Alternatively, other known statistical methods may be used. By selecting the districts A to F in this way, it is possible to grasp the differences in the growing conditions of all the fields to be photographed.

As an example of other processing by the reference date estimating unit 105, for example, all fields to be photographed are classified into three types of "high altitude", "medium altitude", and "low altitude" from the viewpoint of altitude, and furthermore, from the viewpoint of rainfall, It is classified into three types, "high rainfall", "medium rainfall", and "low rainfall". Then, from the results of classification into 3×3=9 types, which covers all combinations of classifications, target field districts are selected so as to cover variations in all fields to be photographed.

Ignoring the cost of the data processing amount, it is possible to divide all the fields to be photographed into equal areas and perform the subsequent processing on all of them.

In the reference date estimating process S502, the reference date estimating unit 105 sets a reference date for ascertaining the growth status of the target crop for each of a plurality of target field districts selected from the target fields for which the growth status of the crop is desired to be ascertained. presume.

In the growth prediction process S503, the growth prediction unit 106 predicts changes in weather data after the reference date estimated in the reference date estimation process S502 in each target field district, and predicts a specific growth stage in each target field district. Predict arrival time.

In the appropriate photographing time determination process S504, the appropriate photographing time determination unit 107 determines the appropriate photographing time of the remote sensing image based on the growth prediction result obtained in the growth prediction process S503.

In the remote sensing image capturing process S505, the earth observation satellite 140 captures a remote sensing image of the target area. An operator of the earth observation satellite 140 (hereinafter referred to as an "observation satellite operator") separately instructs the earth observation satellite 140 to take an image. Instructions to the operator of the observation satellite, such as the location and time of photographing, are given from the crop growth condition analysis device 100 to the operator of the observation satellite online or offline via a person.

Since the (second) remote sensing image captured by the earth observation satellite 140 affects the accuracy required in the growing condition analysis, it is used in the reference date estimation process S502 (first ), it is desirable to have a higher definition than the remote sensing image. The (second) remote sensing image is input to the crop growth condition analysis apparatus 100 via an observation satellite operator, online, or via a person.

Note that the (second) remote sensing image may be acquired by the same equipment as that for acquiring the (first) remote sensing image, or may be acquired by a different equipment. For example, the (first) remote sensing image is an image acquired by a first earth observation satellite 140-1, and the (second) remote sensing image is a second remote sensing image different from the first earth observation satellite 140-1. may be an image acquired by the earth observation satellite 140-2.

In the vegetation index calculation process S506, the crop growth condition analysis unit 108 calculates the vegetation index by a known method based on the photographed remote sensing image. Then, as described in the Background Art section, vegetation indicators such as NDVI, EVI, and LAI are applied to the statistical model to perform analysis such as yield estimation (see Patent Document 1).

<6. Reference date estimation process>
FIG. 6 is a processing flow diagram of reference date estimation processing S502 by the reference date estimation unit 105. As shown in FIG. First, it is determined whether or not there is growth record information relating to a reference date for grasping the growth status of the target crop (S601). The reference date is a date that serves as a starting point for growing rice. For example, in the case of rice, the date of ear emergence, the date of all ears, the date of flowering, the date of rice planting, the date of germination, etc. can be arbitrarily set. When even a little tip of the ear emerges from the stem, it is called heading, and the day when 40 to 50% of the entire paddy field has the heading is defined as the heading date. The time when 80 to 90% of the stalks in the entire paddy field are heading is taken as the heading date.

If there is growth record information obtained by investigating the growth status of the target crop locally, refer to the growth survey record 610 that records the information, and based on the reference date recorded in the past, the reference date of each target field area is estimated (S602). The estimation may be performed automatically, or data estimated based on human experience may be input. The growth survey record 610 refers to, for example, a record related to the date of transplantation for each field. Also, if the growth survey record 610 contains a record of the date of heading and the date of full-heading itself, that date may be used as the reference date.

If there is no result of surveying the growth status of target crops by field observation, the reference date is estimated using remote sensing images. From the satellite image data 110 including the target agricultural field districts A to E, time-series image data is read by the satellite image reading process (S603). A vegetation index value in each target field area is calculated based on each read satellite image (S604). A reference date for grasping the growth status of the target crop is estimated from the temporal change in the vegetation index value obtained in each target field area (S605).

In the reference date estimation process S502 of the present embodiment, the vegetation index value is not obtained at the farm field level, but the vegetation index value at the collective (district) level of farm fields is obtained. 110 preferably has better temporal resolution than spatial resolution. For example, high-frequency satellite images taken every day, such as MODIS (moderate resolution imaging spectroradiometer) images, are suitable even if the spatial resolution is low. Also, instead of the satellite image, the vegetation index value may be calculated based on an aerial photograph taken by an aircraft, drone, or the like.

For the vegetation index value calculated in S604, for example, NDVI represented by the following equation is used. Other vegetation index values may be used in consideration of the time-series photographing band configuration of the satellite images used as the input satellite image data 110, the vegetation index specifying the reference date to be grasped, and the like.
NDVI=(IR−R)/(IR+R)
where R is the observed value in the red band and IR is the observed value in the near-infrared band. As described above, NDVI is the difference between the reflectance of the near-infrared band reflected by the plant body and the red band absorbed by chlorophyll divided by the sum, and indicates the amount and activity of the plant body. It can be used as an index.

In S605, the growth status of the target crop is analyzed from the time change of the vegetation index value calculated in S604, and the reference date for grasping the growth status is estimated. For example, regarding the estimation of the growth situation using the chronological change of NDVI, "Evaluation method of paddy rice growth stage by remote sensing" Independent administrative agency National Institute for Agro-Environmental Sciences Ecosystem measurement research area Toshihiro Sakamoto, Innovative 2006 There are documents such as agricultural technology acquisition training (http://www.naro.affrc.go.jp/archive/niaes/techdoc/inovlec2006/7_sakamoto.pdf), and applying this technology to part of this example can be done.

<7. Growth prediction processing>
FIG. 7 is a processing flow diagram of the growth prediction processing S503 by the growth prediction unit 106. As shown in FIG. From the time-series weather data 120A of each target field district for a plurality of years in the past, a weather data change forecasting formula for each district is determined for the weather data that affects the growth status of the target crop (S701).

In general, the growth status and yield of crops grown in fields such as paddy rice, wheat, and soybeans are estimated and predicted based on the date of sowing, planting, or transplanting, or the date of emergence of the head. , the growth status after the reference date can be grasped based on the temperature after the reference date. The most commonly used index for this understanding is an index called the effective cumulative temperature, which is the sum of the daily average temperatures that exceed a certain reference value. Predict to reach a certain growth status. In this way, when the growing conditions of crops are affected by the effective cumulative temperature, the air temperature is extracted as the weather data. Then, in each area, a prediction formula for daily average temperature change is calculated based on a graph of a daily average temperature change curve, which shows the average temperature change with the observation date on the horizontal axis and the average temperature on the vertical axis.

Next, from the current year's time-series weather data 120B for each district and the weather data change prediction formula obtained in S701, changes in weather data after the reference date of the current year are predicted (S702). For example, when determining a change prediction formula from the daily average temperature change curve in S702, the past daily average temperature change curve is classified into patterns according to characteristics, compared with the data of the current year, and the most similar time change is determined. Adopt a change prediction formula for a pattern that shows

In FIG. 8, the horizontal axis represents time (day) and the vertical axis represents average temperature (Celsius). From the top, they are categorized into three categories: “Growth conditions are the same as average year (average temperature)”, “Growth conditions are better than average year (high temperature)”, and “Growth conditions are worse than average year (low temperature)”.

As shown in FIG. 8, for example, the growth conditions of target crops can be classified into three patterns: average, high temperature, and low temperature. In weather data change prediction (S702), these daily average temperature change curves are compared with the current year's daily average temperature fluctuations, and future daily average temperatures are calculated from a change prediction formula based on the most similar pattern. It is possible to predict the fluctuation situation of

Next, based on the prediction results of future weather data for each district obtained in S702 and the estimation results of the reference date obtained in S602 or S605, the time when the target crop reaches a specific growth stage after the reference date in each district (S703).

In this embodiment, the cumulative temperature from the reference date is calculated using the prediction data obtained by weather data change prediction (S702), and the time when the cumulative temperature from the reference date reaches a certain value is determined as the arrival date of a specific growth stage. , and the period when the accumulated temperature is within a certain range is defined as the period of the growth stage (S703).

For example, when it is desired to estimate the time when paddy rice reaches the harvesting stage as a specific growth stage, the reference date is set as the date of heading or the date of full-heading, and the time when the integrated temperature after heading is within a certain range is set as the harvesting stage. Further, for example, when the heading date of paddy rice is to be estimated, the reference date is the date of transplantation or the young panicle formation period, and the time when the accumulated temperature after the reference date is within a certain range is the heading date.

<8. Appropriate Shooting Timing Decision Processing>
In the appropriate photographing period determination process S504, the appropriate photographing period determination unit 107 sets the period during which each target field district obtained in the growth prediction process S503 is in a specific growth stage as the photographing period in each district, and determines the photographing period for the most target farm field district as the photographing period. Extract a certain time and set it as the appropriate time for shooting.

The photographing period in each district is analytically determined, for example, by taking the date of reaching the growth stage obtained in the growth prediction process S503 as the photographing start date, the photographing end date as the last day at the growth stage, and the photographing period between them. Alternatively, taking into account the target crop, its growth stage, weather conditions, photography conditions, etc., the date of arrival at the growth stage is set as the photography start date, and an arbitrary number of days is set as the photography period.

FIG. 9 shows the concept of a method for determining the appropriate timing for photographing remote sensing images in a target area from the photographing period for each target field area. The vertical axis represents each target agricultural field district to be photographed for the remote sensing image, the horizontal axis represents calendar days, and the photographing period of each district is indicated by hatching. In the period from the earliest shooting start date to the latest shooting end date, the period 901 in which the areas in the shooting period overlap the most is set as the appropriate shooting time. For example, in FIG. 9, the period from (d−2) to (d+2) is determined as the appropriate time for photographing because all six districts from district A to district F are in the photographing period.

In order to estimate the yield of the target area, it is sufficient to take a photograph from the earth observation satellite 140 at least once within the suitable period for photographing. In the case of a high-resolution satellite for estimating crop yields, it is not always possible to take pictures every day due to the satellite orbit, and in general it is possible to take pictures only once every few days. In addition, since it is an optical sensor, there are days when it is not possible to shoot due to weather conditions, etc., so it is desirable to have a certain amount of latitude in the appropriate shooting period.

<9. Remote Sensing Image Acquisition and Vegetation Index Calculation>
The earth observation satellite 140 can acquire a sufficiently high-quality (high-definition) image at a high cost at the appropriate photographing timing determined in the appropriate photographing timing determination processing S504 (S505).

The acquired high-quality remote sensing image 110 is processed by the crop growth condition analysis unit 108 to calculate a vegetation index suitable for yield estimation (S506). Regarding the vegetation index calculation technique, there is a known technique such as Patent Document 1, for example.

Artificial intelligence (AI) may be used as another method for performing the growth prediction processing shown in FIGS. For example, using a DNN (Deep Neural Network) that outputs the time when a specific growth stage is reached with meteorological data as input, AI learns using past meteorological data and predicted times specified by experienced people as teacher data. , growth prediction processing may be performed using learned AI.

As explained above, along with the development of science and technology, techniques using optical remote sensing images obtained from artificial satellites, aircraft, drones, etc. are being researched and used in practical applications to predict crop growth conditions and yields. ing.

In this embodiment, a good quality remote sensing image suitable for analysis can be acquired at low cost. Then, based on the vegetation index and the like calculated from the optical remote sensing image, the yield can be estimated at low cost.

Since the resolution of images and the range that can be captured differ depending on the specifications of the satellites and other equipment used to acquire remote sensing images, when using aerial photographs or high-resolution satellite images obtained by aircraft or drones, When medium to low resolution satellite images are used, they are suitable for grasping the growth situation in each area where multiple fields are grouped together. Regardless of which remote sensing image is used, multiple fields can be photographed in a single shot, so compared to the conventional method of repeating field surveys for each field, the growing conditions are uniform over a wide area. can be efficiently grasped.

Crop growth condition analysis device 100, CPU 101, memory 102, auxiliary storage device 103, input/output device 104, reference date estimation unit 105, growth prediction unit 106, appropriate photographing period determination unit 107, crop growth condition analysis unit 108, satellite image data 110 , weather data 120, GIS data 130, earth observation satellite 140

Claims (10)

comprising an input device, an output device, a processing device, and a storage device;
A system for determining the acquisition timing of remote sensing images for grasping the growth status of agricultural crops targeting agricultural fields distributed on the ground,
a reference date estimation unit for estimating a reference date used for predicting the growth status of agricultural crops based on the first remote sensing image, which is time-series data;
a growth prediction unit that predicts the growth status of crops after a reference date based on meteorological data, which is time-series data, and estimates when the crops will reach a specific growth stage;
an appropriate photographing timing determination unit that determines, as the acquisition timing of the second remote sensing image, the timing at which the largest number of crops having reached the specific growth stage can be photographed;
A remote sensing image acquisition timing determination system comprising: wherein the second remote sensing image has higher definition than the first remote sensing image;
The remote sensing image acquisition timing determination system according to claim 1. The reference date estimation unit
calculating a vegetation index value based on the first remote sensing image, and estimating the reference date from a temporal change in the vegetation index value;
The remote sensing image acquisition timing determination system according to claim 1. The reference date is the date of heading or the date of all headings.
4. The remote sensing image acquisition timing determination system according to claim 3. The growth prediction unit
From past time-series weather data, determine a change prediction formula for weather data that affects the growth status of the target crops,
Predicting changes in weather data after the reference date from the change prediction formula,
Based on changes in the weather data, predicting the growth status of crops after a reference date, and estimating when the crops will reach a specific growth stage.
The remote sensing image acquisition timing determination system according to claim 1. the weather data is temperature data;
Based on the predicted changes in the temperature data, out of the daily average temperatures, the amount exceeding a certain reference value is extracted and totaled to obtain an integrated temperature. predicted to reach a growth state of
6. The remote sensing image acquisition timing determination system according to claim 5. The reference date estimation unit
Sampling the target field area for estimating the reference date from a part of all the target fields for grasping the growth status of agricultural crops,
The remote sensing image acquisition timing determination system according to claim 1. The reference date estimation unit
Using at least one of the weather data and GIS data, sampling so that the data variance of the target field area is large;
8. The remote sensing image acquisition timing determination system according to claim 7. A crop growth condition analysis method for grasping the growth condition of crops in a field distributed on the ground surface using an input device, an output device, a processing device, and a storage device, using a remote sensing image,
Based on the first remote sensing image, which is time-series data, estimating the reference date used to predict the growth status of agricultural crops,
Predicting the growth status of crops after a reference date based on meteorological data, which is time-series data, and estimating when crops will reach a specific growth stage,
Determining a time at which a maximum number of fields in which crops have reached the specific growth stage can be photographed as the acquisition time of the second remote sensing image;
Crop growth analysis method. calculating a vegetation index from the second remote sensing image;
applying the vegetation index to a predetermined statistical model;
Analyze the growth of crops,
The method for analyzing crop growth conditions according to claim 9.

Images