Learning model, signal processing device, flying object, and program

Abstract

It enables proper observation of the situation of the object being observed or the environment surrounding the object being observed. A learning model, a signal processing device, a flying object, and a program are provided. The learning model is based on the reflected electromagnetic waves that are irradiated onto a first target area. A first reception signal based on the first reception signal is input, and a first meta information having predetermined items corresponding to the first reception signal is generated. The electromagnetic wave irradiated to the second target area is reflected by the training data. A second reception signal based on the reflected electromagnetic wave is input, and a predetermined item is selected in response to the second reception signal. The second meta information is output.

Description

[Technical Field]

[0001] The present invention relates to a learning model, a signal processing device, a flying object, and a program. [Background technology]

[0002] Land and sea areas using flying objects such as satellites, aircraft, or drones Observation of the state of the sphere's surface is widely carried out. One of the observation methods using satellites is to obtain optical images. Synthetic Aperture Radar (SAR) is an observation method that uses The image is obtained by using radar (SAR) technology, which is called SAR image. This is an observation method that uses both optical and SAR images, and combines them. Patent Document 1 describes a method for combining radar images such as SAR images and optical images. A method for generating an image for interpreting geographical features that generates a composite image that makes it easier to distinguish objects is shown. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-047516 Summary of the Invention [Problem to be solved by the invention]

[0004] SAR images are created by irradiating microwaves onto an object from a satellite equipped with a radar device. (electromagnetic waves) are reflected by the object being observed, and a signal corresponding to the reflected electromagnetic waves (hereafter referred to as the received signal) is generated. The SAR image is generated based on the received signal, which is compressed in a predetermined manner. It is generated by

[0005] In the compression process for generating SAR images, the received signal is filtered in the frequency domain. The signal is filtered to remove some of the data from the received signal. This reduces the amount of data in the received signal, and reduces the computational burden in the compression process. The removal of the received signal by filtering may result in the loss of detectable information from the received signal or the loss of information. Missing information and false positives may occur due to the situation of the object being observed or the surroundings of the object being observed. This has an impact on the precision or accuracy of the observation of the environment.

[0006] Therefore, the present invention provides a method for detecting the state of an object to be observed or the environment around the object to be observed with high accuracy or high precision. Learning model, signal processing device, flying object, and program that enable accurate observation The purpose is to provide the following. [Means for solving the problem]

[0007] The learning model according to one aspect of the present invention is a model of a first target region in which an electromagnetic wave is irradiated onto the first target region and reflected therefrom. A first reception signal based on the received electromagnetic wave is input, and the first reception signal corresponds to the first reception signal and has predetermined items. The first meta information is used as an output to learn the training data, and the second target area is subjected to electromagnetic waves irradiated thereto. A second reception signal based on the reflected electromagnetic wave is input, and a predetermined The second meta information having the specified items is output.

[0008] According to this aspect, the learning model, in response to an input of a received signal, generates a signal corresponding to the received signal. It is trained to output meta-information and operates accordingly. By using this learning model, For example, meta information having an item of the total number of moving objects and buildings in the target area is It is possible to obtain the information based on the received signal. For example, the information can be obtained from the received signal via SAR images. This allows the meta information to be acquired without any loss of information contained in the received signal. Therefore, it is possible to observe meta-information that indicates the status of the object to be observed with high accuracy or precision. become.

[0009] In the above aspect, the learning model is generated based on the first received signal and the second received signal. The first generated signal is input, and the first meta information is output, and the training data is used to train the first generated signal. The second received signal and a second generated signal generated based on the second received signal are input, and second meta information is generated. The information may be output.

[0010] According to this aspect, the training data in the learning model includes the first received signal. The first generated signal may be, for example, a signal generated based on the first received signal. It is a signal for generating SAR images (SAR signal). A learning model is used that further includes data generated based on the By doing so, in cases where SAR signals can more appropriately observe meta-information of the target object, etc., It becomes possible to observe the state of the object with high accuracy or precision.

[0011] In the above aspect, the learned model represents the first received signal and the environment in the first target region. The second received signal is trained using training data that takes the first meta information as input and outputs the first meta information. and information indicating the environment in the second target area may be input, and second meta information may be output.

[0012] According to this aspect, the training data in the learning model includes the environment in the first target domain. The information indicating the environment in the first target area further includes, for example, Meteorological conditions such as weather in the target area, or environmental conditions caused by human factors such as smoke, When outputting the second meta information, an input including information indicating the environment in the second target area is By using the above, it becomes possible to observe the second meta information with high accuracy or high precision.

[0013] In the above aspect, the learning model includes a first model including information indicating an environment in a first target domain. The second target area is trained using training data that outputs the data information, and the second target area is trained using training data that outputs the data information. The second meta information including the above may be output.

[0014] By using the learning model in this manner, the second object area, which is the surrounding environment of the observed object, Therefore, it is possible to obtain information indicating the environment in the area. This makes it possible to observe with high accuracy.

[0015] In another aspect, the signal processing device includes a memory unit in which the learning model of the above aspect is stored. a signal acquisition unit that acquires a second received signal; and inputting the second received signal into a learning model. and an estimation unit that estimates the second meta information.

[0016] According to this aspect, the signal processing device alone can acquire the signal and generate the second meta-analysis signal using the learning model. This allows us to estimate information about external locations, such as the space above the Earth. Even in an environment where communication with other devices is restricted, meta-information can be estimated using a learning model. This allows the signal processing device to observe the state of the object to be observed with high accuracy or precision. It becomes possible.

[0017] In the above aspect, the signal processing device, the estimation unit estimates the second received signal at the first time and The second received signal at the second time is input to the learning model of the above aspect, and the second meta-information at a first time corresponding to a second received signal and the second meta-information at the second time corresponding to the second received signal is estimated, and and the second meta information at the second time, The device may further include a change determination unit that determines a change.

[0018] According to this aspect, the second meta information of the second target area at the first time is When a signal changes, the presence or absence of the change can be detected. The processing device observes the situation of the object to be observed or the environment around the object to be observed with high accuracy or precision. While doing so, changes in the second region of interest can be determined.

[0019] In the above aspect, when the determined change satisfies a predetermined condition, the signal processing device: The device may further include a change information output unit that outputs change information indicating the change.

[0020] According to this aspect, the signal processing device determines whether or not there is a change if the change satisfies a predetermined condition. In addition, the content of the change is output as change information, so that the details of the change can be transmitted to an external device. The signal processing device acquires the change information as needed based on the conditions. By outputting the information, the amount of communication between the signal processing device and the outside and the amount of communication required in the signal processing device can be reduced. This makes it possible to reduce unnecessary power consumption.

[0021] In another aspect, the flying object includes a storage unit in which the learning model of the above aspect is stored, and A signal acquisition unit that acquires a second received signal and inputs a second received signal to a learning model, and an estimation unit that estimates the second meta information; and a signal output unit that outputs an output signal based on the second meta information to an outside. Equipped with.

[0022] According to this aspect, the flying object alone can estimate the second meta information using the learning model. This makes it possible for flying objects placed in environments where there are certain restrictions on communication with the outside world. This makes it possible to estimate the second meta-information without communicating with the outside. The amount of data and the power consumption required for communication can be reduced. Based on the second meta information, for example, output information including the second meta information itself and information indicating a change in the second meta information is output. This allows for the external output of large data such as SAR data. Instead, secondary meta-information equivalent to meta-information observed based on SAR data is sent to the outside. This will reduce the amount of communication between the flying object and the outside world, and the amount of communication by the flying object. The necessary power consumption can be reduced.

[0023] In another aspect, the program is a program for storing the learning model of the above aspect in a computer. a signal acquisition process for acquiring a second received signal input to a storage unit in which the second received signal is stored; The second received signal is input, and an estimation process is executed to estimate the second meta information. To allow a computer to observe meta-information indicating the status of an object with high accuracy or precision. This becomes possible.

[0024] In the above aspect, the program causes the computer to generate an output signal based on the second meta information. The processor may further perform a signal output process to output the signal to the outside. In a flying object equipped with a computer in which a program is recorded, the amount of communication between the flying object and the outside and This also reduces the power consumption required for communication by the flying object. [Effects of the Invention]

[0025] According to the present invention, it is possible to appropriately observe the situation of an object to be observed or the environment around the object to be observed. A learning model, a signal processing device, a flying object, and a program can be provided. . [Brief explanation of the drawings]

[0026] [Figure 1] FIG. 1 is a block diagram of an observation system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a diagram illustrating a learning model according to the present embodiment. [Figure 3] FIG. 2 is a diagram illustrating learning of a learning model according to the present embodiment. [Figure 4] FIG. 10 is a diagram illustrating an example of information used for learning a learning model according to the present embodiment. [Figure 5] FIG. 10 is a diagram illustrating an example of information used for learning a learning model according to the present embodiment. [Figure 6] FIG. 10 is a diagram illustrating the correspondence of teacher data used in learning of a learning model according to the present embodiment. [Figure 7] 10 is a flowchart illustrating processing in a flying object according to the present embodiment. [Figure 8] 10A and 10B are diagrams illustrating estimation of meta information by a signal processing device according to the present embodiment. [Figure 9] 10A and 10B are diagrams illustrating an example of meta information estimated by the signal processing device according to the present embodiment. [Figure 10] 10A and 10B are diagrams illustrating how a signal processing device according to the present embodiment determines whether meta information has changed. [Figure 11] 10 is a flowchart illustrating a change determination process performed by the signal processing device according to the present embodiment. [Figure 12] FIG. 10 is a diagram illustrating another aspect of learning and estimation of a learning model. [Figure 13] FIG. 10 is a block diagram showing another aspect of the observation system. DETAILED DESCRIPTION OF THE INVENTION

[0027] A preferred embodiment of the present invention will be described with reference to the accompanying drawings. , components with the same reference numerals have the same or similar configurations.

[0028] FIG. 1 shows a block diagram of an observation system 10 according to this embodiment. 10 includes a flying object 100 and an observation device 200. The flying object 100 is placed in the space above the Earth. In this embodiment, the flying object 100 is placed on the Earth's surface. The target area D is observed by a radar, and an observation signal O processed in the flying object 100 is observed. The observation signal O is, for example, a signal acquired by the flying object 100, which will be described later. The received signal is meta-information corresponding to the received signal.

[0029] The flying object 100 includes a communication antenna 101, a radar device 102, and a signal processing device 103. The flying object 100 is an artificial satellite capable of acquiring and processing received signals, and is located in outer space. The flying object 100 may be a geostationary satellite. The flying object 100 is an aircraft, helicopter, or drone positioned above the Earth. Any device capable of doing this may be used.

[0030] The communication antenna 101 is used to connect the flying object 100 to an external device located on the Earth or in outer space. It is an antenna for communication.

[0031] The radar device 102 transmits electromagnetic waves EM1, such as microwaves, to a target area D on the Earth's surface. The electromagnetic wave EM1 is reflected by the object to be observed in the target area D. The radar device 102 is a device for acquiring electromagnetic waves EM2. The reflected electromagnetic wave EM2 is handled by the flying object 100 by the radar device 102. The received signal (RAW data) based on fluctuations in electromagnetic waves is processed and recorded. The received signal is recorded, for example, as a signal strength for each predetermined coordinate in the target area D. The device 102 transmits an observation signal O to the observation device 200 via the communication antenna 101 .

[0032] The radar device 102 includes a processor for controlling the acquisition process of the received signal and a The device includes a storage device that stores the programs required for the above.

[0033] The signal processing device 103 processes the received signal acquired by the radar device 102. The signal processing device 103 has a storage area such as a memory, and stores A computer that performs a predetermined process by having a processor execute a program that has been written in the be.

[0034] The signal processing device 103 includes a storage unit 104 and a control unit 105. The storage unit 104 stores, for example, For example, it is a semiconductor memory such as a RAM or an optical disk. It stores various information used in processing in 03.

[0035] The storage unit 104 stores a learning model 1041. The learning model 1041 is It is a program that has been trained to take a signal as input and output meta information corresponding to the received signal. The meta information and learning model 1041 will be described in detail later.

[0036] The control unit 105 performs signal processing in the signal processing device 103. The control unit 105 controls the transmission of the processing results of the received signal through the flying object 100. An acquisition unit 1051, an estimation unit 1052, a signal output unit 1053, a change determination unit 1054, and a change information It has a message output unit 1055.

[0037] The signal acquisition unit 1051 acquires a received signal from the radar device 102 .

[0038] The estimation unit 1052 estimates the received signal acquired by the signal acquisition unit 1051 based on the learning model 10 41 to obtain meta information corresponding to the received signal.

[0039] The signal output unit 1053 outputs the meta data acquired by the estimation unit 1052 via the communication antenna 101. The signal output unit 1053 outputs the information to the observation device 200 as an observation signal O. The received signal corresponding to the meta-information may be output as an observation signal O together with the information.

[0040] The change determination unit 1054 determines whether or not a change has occurred in response to received signals acquired from the target area D at different times. Based on the multiple meta-information, the situation of the observed object in the target area D or the observed object Determine changes in the environment around an object.

[0041] The change information output unit 1055 outputs the change information determined by the change determination unit 1054 under a predetermined condition. If the condition is satisfied, change information indicating the change is output. Among them, the information from which the changed meta information was extracted and the area where the change occurred in the target area D The information indicates the range. More details will be given later.

[0042] The observation device 200 transmits a control signal to the flying object 100 to control the observation of the target area D. 100 and acquires an observation signal O from the flying object 100. 00 has a communication unit 201 including an antenna and a control unit for controlling communication via the antenna. The communication unit 201 is used to transmit and receive information to and from the flying object 100 .

[0043] The signal processing unit 202 processes the observation signal O from the flying object 100. is a result of observation in a target area D based on an observation signal O acquired from the flying object 100. The results are then visualized using images.

[0044] The learning of the learning model 1041 according to this embodiment will be described with reference to FIGS. 2 to 6. do.

[0045] FIG. 2 is a diagram for explaining the learning and estimation of the learning model 1041. 1041 is learned using the learning data LD as training data. A received signal R0 (first received signal) obtained by irradiating an electromagnetic wave onto a target area (first target area) is The received signal R0 is included in the set of meta information MD0 (first meta information) corresponding to the received signal R0. The learning model 1041 is trained using the received signal R0 as input and the meta-information MD0 as output.

[0046] In order to associate the meta information MD0 with the received signal R0, the received signal R0 is treated as SAR data. The received signal R0 is converted into a signal by the observation device 200 unless it undergoes a predetermined conversion process. The conversion process of SAR data is performed by, for example, the observation device 200. Analysis and visualization are performed based on the SAR data, The user can understand the observation results based on the received signal R0.

[0047] The SAR data has multiple levels depending on the content of the transformation of the received signal R0. The AR data includes, for example, range compression and single look adjustment for the received signal R0. There is first-level SAR data that has undergone mass compression. The information on the amplitude and phase of the reflected electromagnetic wave EM2 in the target area D is The first level SAR data is visualized as a SAR image, allowing users to This allows the user to understand the contents of the received signal R0.

[0048] Other SAR data include range compression and multi-look azimuth for the received signal R0. There is also second-level SAR data that has undergone compression. , the received signal R0 can be visualized in a geometrically corrected SAR image.

[0049] Other SAR data include range compression, single look azimuth, and R0. There is also a third level of SAR data that has undergone data compression and orthorectification. This makes it possible to obtain SAR images that can be overlaid on optical images.

[0050] As mentioned above, the received signal R0 is converted into SAR data and visualized as a SAR image. This allows the user to understand the observation results in the target area D. can associate meta-information MD0 with received signal R0 as information indicating the meaning of the observation result. Alternatively, a learning model that associates SAR images with meta-information can be used to After associating meta information with the SAR image using the data, the received signal and meta information are matched. You may also respond.

[0051] For example, in the case of observations of ships at sea, the meta information M0 includes information about the ship and sea conditions. Specifically, ship information includes information on location, length, and type. It is obtained from SAR images. Phase information based on SAR data with complex components is also available. It is possible to obtain the ship's speed using the above. The wind direction and speed above the sea can be obtained based on SAR data. is based on the backscatter coefficient of SAR data and the actually measured wind direction and wind speed data. The backscatter coefficient is the amount of light scattered by the surface of the target area. This is a coefficient based on the strength of the electromagnetic waves that return in the direction of irradiation.

[0052] The correspondence between the received signal R0 and the meta information MD0 includes additional information other than the SAR data. For example, the meta information MD0 about ships can be Automatic Identification System (AIS) This includes AIS information, which is information such as location, ship name, flag, and overall length obtained from the system. Meta information MD0 is generated based on the AIS information when the received signal R0 is acquired. Regarding sea conditions, wind direction and wind speed obtained from buoys installed on the ocean may be used. , and weather may be used as meta information MD0.

[0053] Meta information MD0 has various items depending on the object of observation. The meta information for the moving object includes information on the moving trajectory of the moving object. This is obtained from the change in SAR data due to the change in the interference of reflected electromagnetic waves caused by the In this case, the meta information includes a movement trajectory based on a GPS device attached to the moving object. That's fine.

[0054] When flooded areas are the subject of observation during a disaster, the extent of the flooded areas is recorded in the SAR image as meta information. In this case, the meta information may be acquired as a low-brightness area in the aircraft or satellite image. The extent of flooded areas based on optical images acquired by stars may also be included.

[0055] The meta-information used for managing agricultural crops may include the growth rate of the crops. The growth rate of the crops is calculated based on the backscatter coefficient of the SAR data and the actual observed growth rate of the crops. In this case, the meta-information includes the optical The information may include information on the spectral reflectance characteristics measured by the field survey and the height of the crops actually measured.

[0056] In the detection of buildings, information about new buildings may be used as meta information. Information on new buildings is obtained from the backscatter coefficients of the SAR data and the actual observed new buildings. It is estimated based on correlations calculated based on information about the building. It is also possible to obtain information about buildings detected from academic images and information about new buildings from map information. In addition, information about the type of building may be acquired based on maps, etc., and used as meta-information. good.

[0057] In associating meta-information with received signals, electromagnetic wave simulation is used. In the simulation model, the irradiation and reflection of electromagnetic waves are simulated. In this case, the received signal and SAR data can be generated. The conditions in the model, such as the number of ships, the shape of the ships, and the trajectory of the moving object's position, are generated. This becomes meta information associated with the generated received signal and SAR data.

[0058] Based on the electromagnetic waves irradiated by the flying object 100 to another target area (second target area), The received signal R1 (second received signal) acquired by the object 100 is input to the trained learning model 1041 When the received signal R1 is input, the learning model 1041 generates meta information MD1 (Second meta information) is output.

[0059] The learning model 1041 is configured to acquire, in addition to the received signal R1, additional information when the received signal R1 is acquired. For example, in the example of the ship mentioned above, AIS information and information from buoys on the ocean can be additional information. The Model 1041 combines SAR data, AIS information, and information from buoys on the ocean. The learning model is assumed to be trained using meta-information MD0 based on the additional information including the When estimating meta information MD1 using Rule 1041, AIS information or buoys installed on the ocean are used. Any of the information from is input to the learning model 1041 as additional information. By including additional information in addition to SAR data in the input of 1041, the meta information can be improved. It may be possible to observe with higher accuracy or precision.

[0060] The relationship between the received signal R0 and the meta information MD0 will be described with reference to FIGS. In this embodiment, it is possible to estimate the sea conditions, which are the ship at sea and the environment around the ship. This section explains the learning model as an example.

[0061] In the example of FIG. 3, based on the received signal R0 corresponding to the target area D, S In the SAR image IG1, ships S31, S32, and S3 3 is detected. In this case, as shown in FIG. 4, the first vessel information and Meta information MD01 containing first oceanographic information is obtained from SAR image IG1. The meta information MD01 is associated with the received signal R0. The "Ship ID" item in the second ship information identifies the ship in the meta information MD01. It is any information used to

[0062] In addition, when the received signal R0 corresponding to the target area D is acquired, the AIS Based on the information, meta information MD02 as shown in FIG. 5 is obtained. 2 includes second vessel information regarding the vessel and second sea state information regarding sea conditions. The information and second sea condition information have different items from the first ship information and second sea condition information. The data information MD02 is associated with the received signal R0.

[0063] Based on the AIS information, as shown in the correct image IG2 in Figure 3, the vessel S31 In addition to S33, the meta information MD02 indicates that ship S34 is also present in the target area D. Therefore, it is shown.

[0064] The meta information MD0 associated with the received signal R0 as the learning data LD is It is prepared based on MD1 and meta information MD2. Meta information MD0 contains the following information about the ship: The third vessel information is based on the first vessel information and has the same items as the first vessel information. In addition, the meta information MD0 will have the same items as the second sea condition information regarding sea conditions. In this way, the third sea state information is included based on the second sea state information. The meta information MD0 is based on SAR images and meta information MD01 is based on information from other devices. The meta information MD02 based on the above can be generated by combining the meta information MD02. Either the meta information MD01 or the meta information MD02 is used as the meta information MD0. Good too.

[0065] The learning model 1041 uses the learning data LD prepared as described above to, for example, It is trained by common machine learning methods such as using neural networks. The learning model 1041 may be configured as a single learning model, or may be configured as a plurality of learning models. The learning model may be configured as a combination of models.

[0066] The processing by the flying object 100 will be described with reference to FIGS.

[0067] In step S701 of FIG. 7, the radar device 102 emits electromagnetic waves E The timing of irradiation is controlled by the observation device 200. Alternatively, the timing may be a timing designated in advance in the flying object 100.

[0068] In step S702, the signal acquisition unit 1051 acquires the reflected wave detected by the radar device 102. A received signal R1 based on the electromagnetic wave EM2 is acquired from the radar device 102.

[0069] In step S703, the estimation unit 1052 estimates the received signal R1 as At this time, the estimation unit 1052 receives the received signal R1 as well as the received signal R2. The additional information acquired may also be input to the learning model 1041 .

[0070] In step S704, the estimation unit 1052 estimates the meta information MD corresponding to the received signal R1. 1 is obtained from the learning model 1041.

[0071] In step S705, the signal output unit 1053 outputs an output signal based on the meta information MD1. to the observation device 200. The output signal based on the meta information MD1 is Alternatively, the output signal may be a signal that conveys a part of the meta information MD1. It may also be a signal that transmits information resulting from information processing.

[0072] The estimation of meta-information by the learning model 1041 will be described with reference to FIG. A learning model 104 trained using the learning data LD having the correspondence shown in FIG. 6 We will explain the case where 1 is used.

[0073] The received signal R1 can be converted into SAR data indicative of the conditions shown in the SAR image IG3. The meta information generated based on the SAR image IG3 includes ships S81 to S8 The actual situation is the situation shown in the correct image IG4. In other words, information about the ship S84 is obtained by converting the received signal R1 into SAR data. It's missing.

[0074] In this case, the signal processing device 103 inputs the received signal R1 to the learning model 1041, and The meta information MD1 shown in the figure below can be obtained. The meta information MD1 contains four ships. That is, the signal processing device 103 detects the This makes it possible to detect meta-information that would otherwise be impossible to find.

[0075] By using the received signal and learning model1041, the influence of the transformation of SAR data is eliminated. This allows for estimation of meta-information without the need for a dedicated server. When false positives occur in the R image, such as when an object that should not be detected is detected as a false image. Even if the metadata is not available, the meta information can be estimated appropriately.

[0076] The detection of a change in meta information by the flying object 100 will be described with reference to FIGS. 10 and 11. do.

[0077] In FIG. 10, the detection of a change in meta information about a building is explained. At time 1, the buildings are arranged as shown in the layout image IG5. At T2 (second time), buildings are placed in the same area as shown in placement image IG6. The layout images IG5 and IG6 are, for example, images of a building being installed from the air or space. The images IG5 and IG6 are optical or SAR images observed from a specified As shown in the layout image IG6, at time T2 A new building NB is being constructed in area A03.

[0078] An example will be described in which the above change is detected using the learning model 1041. In response to the input of the received signal, the module 1041 outputs building information, which is the number of buildings in each area, to the memory. It is assumed that the system has been trained to output the data as input data.

[0079] FIG. 11 shows a flowchart of the processing by the flying object 100.

[0080] In step S1101, the radar device 102 detects an electromagnetic wave in a target area at time T1. Emits waves.

[0081] In step S1102, the signal acquisition unit 1051 receives the reflected signal detected by the radar device 102. The received signal R3 based on the emitted electromagnetic waves is acquired from the radar device 102. The received signal R3 is stored in a memory unit 104.

[0082] In step S1103, the estimation unit 1052 estimates the received signal R3 as a learning model 1041. Enter.

[0083] In step S1104, the estimation unit 1052 estimates the meta information M corresponding to the received signal R3. D3 is acquired from the learning model 1041. The meta information MD3 is stored in the storage unit 104. Good too.

[0084] In step S1105, the radar device 102 detects an electromagnetic wave in the target area at time T2. Emits waves.

[0085] In step S1106, the signal acquisition unit 1051 acquires the reflected signal detected by the radar device 102. A received signal R4 based on the emitted electromagnetic waves is acquired from the radar device 102.

[0086] In step S1107, the estimation unit 1052 estimates the received signal R4 as a learning model 1041. Enter.

[0087] In step S1108, the estimation unit 1052 estimates the meta information M corresponding to the received signal R4. D4 is obtained from the learning model 1041.

[0088] In step S1109, the change determination unit 1054 compares the meta information MD3 and the meta information MD 4, the change in the target area is determined. In the case of FIG. 10, the meta information MD04 is The number of buildings in the area A03 has increased to four. It is determined that there is a change in the number of buildings.

[0089] In step S1110, the change information output unit 1055 outputs the change information if the change satisfies a predetermined condition. Here, the predetermined conditions include the result of determining whether or not there is a change, and the specific details of the change. For example, in the example in Figure 10, the specific content of the change is Conditions include the existence of a change or an increase or decrease in the number of buildings in one area. It could be.

[0090] Here, the condition is that a change in the number of buildings occurs in any area. At this time, the change information output unit 1055 determines that the change satisfies a predetermined condition.

[0091] If it is determined that the change satisfies the predetermined condition, in step S1111, the change information The output unit 1055 outputs information indicating the change as an output signal to the observation device 200. If a certain condition is not met, the process ends. The output signal may contain, for example, meta-information MD3 and Alternatively, the output signal may be a signal conveying MD4, or a signal relating to the area A03 where a change has occurred. It is a signal that extracts meta-information and transmits the extracted meta-information. Among the determined area A03, information on the coordinates showing the part where the change occurred in more detail is It may also be a signal to be transmitted.

[0092] The flying object 100 determines the target area based on the accurate or precise meta-information based on the received signal. Therefore, the accuracy or precision of detecting the change is improved.

[0093] The flying object 100 determines the change and outputs only the meta information itself and the part where the change occurred. By transmitting information indicating the range of the change to the observation device 200, This allows the flying object 100 to reduce power consumption. This is because the space environment above the Earth limits the amount of power available. This is advantageous for the flying object 100.

[0094] FIG. 12 is a diagram for explaining the learning and estimation of the learning model 1041A. In addition to the received signal R0 and meta information MD0, the Dell 1041A also uses the following as training data LD: It includes SAR data (first generated signal) generated based on the received signal R0 and additional information. By preparing the learning data LD in this way, the received signal R1 , SAR data (second generated signal) generated based on the received signal R1 and It is possible to estimate the meta information MD1 using the corresponding additional information as input.

[0095] For example, the additional information may include the environment in the target area when the received signal R0 was acquired. The weather in the target area can be used as information. When estimating the meta-information MD1, the environment in the target area when the received signal R1 was acquired is indicated. This information can be acquired from other devices and included in the input of the learning model 1041A. This allows the meta information MD1 to be estimated with higher accuracy or precision.

[0096] FIG. 13 shows a block diagram of an observation system 10A according to another embodiment. As shown in FIG. 1, the signal processing device 103 is provided so as to be included in the observation device 200A. The control unit 1301 of the flying object 100A can also The received signal is transmitted to the observation device 200A. The above processing can also be performed by

[0097] The above-described embodiments are provided to facilitate understanding of the present invention and are not intended to limit the present invention. The elements and conditions of the embodiments are not intended to be construed as examples. The above is not limited to the above, and can be changed as appropriate. They can be partially substituted or combined. [Explanation of symbols]

[0098] 10...observation system, 100...flying object, 101...communication antenna, 102...radar device, 103...signal processing device, 104...storage unit, 1041...learning model, 105...control unit, 10 51... signal acquisition unit, 1052... estimation unit, 1053... signal output unit, 1054... change determination unit, 1055...change information output unit, 200...observation device, 201...communication unit, 202...signal processing unit

Claims

1. learning is performed using teacher data in which a first synthetic aperture radar reception signal based on an electromagnetic wave reflected from an electromagnetic wave irradiated onto a first target area and a first generated signal generated based on the first synthetic aperture radar reception signal are input, and first meta information corresponding to the first synthetic aperture radar reception signal and having predetermined items is output; A learning model that causes a computer to function such that, in response to input of a second synthetic aperture radar received signal based on a reflected electromagnetic wave that is irradiated onto a second target area and a second generated signal generated based on the second synthetic aperture radar received signal, the computer outputs second meta information that corresponds to the second synthetic aperture radar received signal and has the specified items.

2. learning is performed using teacher data in which a first synthetic aperture radar reception signal based on a reflected electromagnetic wave that is irradiated onto a first target area and information indicating an environment in the first target area are input, and first meta information corresponding to the first synthetic aperture radar reception signal and having predetermined items is output; A learning model that causes a computer to function such that, in response to input of a second synthetic aperture radar received signal based on a reflected electromagnetic wave that is irradiated onto a second target area and information indicating the environment in the second target area, it outputs second meta information that corresponds to the second synthetic aperture radar received signal and has the specified items.

3. The learning model according to claim 1 or 2, The learning model is trained using the training data that outputs the first meta information, which includes information indicating the environment in the first target area, and causes the computer to function to output the second meta information, which includes information indicating the environment in the second target area.

4. A storage unit in which the learning model according to any one of claims 1 to 3 is stored; a signal acquisition unit that acquires the second synthetic aperture radar received signal; an estimation unit that inputs the second synthetic aperture radar received signal to the learning model and estimates the second meta information.

5. A storage unit in which the learning model according to any one of claims 1 to 3 is stored; a signal acquisition unit that acquires the second synthetic aperture radar received signal; an estimation unit that inputs the second synthetic aperture radar received signal to the learning model and estimates the second meta information; a signal output unit that outputs an output signal based on the second meta information to the outside.

6. On the computer, a signal acquisition process for acquiring the second synthetic aperture radar received signal input to a storage unit in which the learning model according to any one of claims 1 to 3 is stored; a program for executing an estimation process in which the second synthetic aperture radar received signal is input to the learning model and the second meta information is estimated.

7. 7. The program according to claim 6, The computer, a signal output process for outputting an output signal based on the second meta information to the outside.

Images