JP2011209780A - Change area specification device and change area specification program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a change area (e.g., a stricken area) where a vegetation area, a water area or the like is excluded from one coherence image.SOLUTION: A coherence image generation part 110 generates the coherence image 193 indicating the coherence of an observation area based on first SAR (Synthetic Aperture Radar) imaging data 191 and second SAR imaging data 192. A mask area exclusion part 120 specifies the vegetation area and the water area included in the observation area as a mask area based on a land covering classification figure 181, a vegetation index figure 182, or an entropy alpha classification figure 183. A change area specification part 130 specifies an area where the coherence is smaller than a coherence threshold in the observation area (except the mask area) as the change area based on the coherence image 193.

Description

  The present invention relates to a change area specifying device and a change area specifying program used for remote sensing, for example.

There is a technique called CCD (Coherence Change Detection) that detects a change area using SAR data (SAR: Synthetic Aperture Radar).
CCD is a technique for calculating coherence (complex correlation) from two SAR data with different observation periods and regarding a low coherence region as a change region.
For example, the CCD is used to detect a disaster area whose state has changed due to a disaster.
However, since the coherence is always low in the vegetation area and the water area, the CCD detects these areas as change areas. For this reason, there was a possibility of accidentally detecting the affected area.

As a technique for solving this problem, Patent Document 1 exists.
The technique disclosed in Patent Document 1 can exclude areas with low coherence (vegetation area, water area, etc.) by using two or more coherence images.
However, since this technique generates two or more coherence images, three or more SAR images are required.

JP 2009-289111 A

Yoshio Yamaguchi, “Basics and Applications of Radar Polarimetry”, IEICE, December 25, 2007, p. 120-p. 124 Motoyuki Sato, “8 Eigendecomposition of Scattering Matrix”, [online], [retrieval on March 16, 2010], Internet <URL: http: // cobalt. cneas. tohoku. ac. jp / users / sato / 8% 20Eigen% 20decomposition. pdf>

An object of the present invention is to make it possible to detect a change area (for example, a disaster area) excluding a vegetation area and a water area from a single coherence image.
Thereby, the convenience and immediacy of the technique which detects a change area | region can be improved.

The change area specifying device of the present invention is:
An image for storing first image data obtained by observing a specific observation area and second image data obtained by observing the observation area at an observation time different from the observation time of the first image data A data storage unit;
A mask region specifying data storage unit for storing mask region specifying data for specifying a region whose state changes with time in the observation region as a mask region;
A coherence calculating unit that calculates, for each pixel, a coherence between the first image data and the second image data based on the first image data and the second image data stored in the image data storage unit; ,
A change area specifying unit that specifies an area other than the mask area in the observation area, which is reflected in a pixel whose coherence is smaller than a predetermined coherence threshold, as a change area;

The image data storage unit stores polarimetric image data obtained by observing the observation area using a plurality of electromagnetic waves having different vibration directions,
The change area specifying device further includes:
An entropy alpha calculator that calculates the entropy of the polarimetry image data and the alpha angle of the polarimetry image data for each pixel;
Based on the entropy and alpha angle calculated by the entropy alpha calculation unit, a portion appearing in a pixel having a predetermined combination of entropy and alpha angle is specified as the mask region, and data indicating the specified mask region is specified as the mask A mask region specifying data generating unit that generates the region specifying data.

  The image data storage unit stores at least one of the first image data and the second image data as the polarimetric image data.

The mask area specifying data storage unit stores area classification data for classifying the observation area for each area as the mask area specifying data;
The change area specifying unit excludes an area classified into a predetermined type based on the area classification data as the mask area from the change area.

The mask area specifying data storage unit stores vegetation index data for specifying a vegetation area in the observation area as the mask area specifying data,
The change area specifying unit excludes the vegetation area from the change area as the mask area based on the vegetation index data.

  The mask area specifying data specifies at least one of a vegetation area and a water area as a mask area.

The change area specifying program of the present invention is:
An image for storing first image data obtained by observing a specific observation area and second image data obtained by observing the observation area at an observation time different from the observation time of the first image data A change area specifying program of a change area specifying device comprising a data storage section and a mask area specifying data storage section for storing mask area specifying data for specifying a mask area specifying an area whose state changes with time in the observation area. And
A coherence calculation process for calculating, for each pixel, a coherence between the first image data and the second image data based on the first image data and the second image data stored in the image data storage unit; ,
The computer is caused to execute a change area specifying process for specifying, as a change area, a part of the observation area other than the mask area that appears in a pixel whose coherence is smaller than a predetermined coherence threshold.

According to the present invention, for example, a change area (for example, a disaster area) excluding a vegetation area and a water area can be detected from one coherence image.
Thereby, the convenience and immediacy of the technique which detects a change area | region can be improved.

FIG. 3 is a functional configuration diagram of a change area detection device 100 according to the first embodiment. FIG. 3 shows an observation method of an observation area in the first embodiment. 4 is a flowchart showing a change region detection method of the change region detection apparatus 100 according to the first embodiment. FIG. 6 is a diagram illustrating an example of a first SAR image 191a and a second SAR image 192a according to Embodiment 1. FIG. 6 shows an example of a coherence image 193 in Embodiment 1. Schematic showing the land cover classification | category figure 181 in Embodiment 1. FIG. Schematic showing the vegetation index figure 182 in Embodiment 1. FIG. FIG. 6 is a schematic diagram of an entropy alpha classification table 184 in the first embodiment. FIG. 6 shows an example of a coherence mask image 194 in the first embodiment. FIG. 6 shows an example of a coherence mask image 194 in which a change area is specified in the first embodiment. FIG. 10 is a diagram showing another example of a flowchart showing a change region detection method of the change region detection apparatus 100 according to the first embodiment. FIG. 3 is a diagram illustrating an example of hardware resources of the change area detection device 100 according to the first embodiment.

Embodiment 1 FIG.
A mode of detecting a change area (for example, a disaster area) excluding a vegetation area and a water area from a single coherence image will be described.

FIG. 1 is a functional configuration diagram of a change region detection apparatus 100 according to the first embodiment.
The functional configuration of the change area detection device 100 according to Embodiment 1 will be described below with reference to FIG.

  The change area detection apparatus 100 (an example of a change area specifying apparatus) includes a coherence image generation unit 110, a mask area exclusion unit 120, a change area specification unit 130, an auxiliary data storage unit 180, and a SAR imaging data storage unit 190.

The SAR imaging data storage unit 190 (an example of an image data storage unit) observes an observation area at an observation time different from the first image data obtained by observing a specific observation area and the observation time of the first image data. The second image data obtained in this way is stored.
The SAR imaging data storage unit 190 stores polarimetric image data (polarimetry observation data 199 described later) obtained by observing an observation area using a plurality of electromagnetic waves having different vibration directions.
For example, the SAR imaging data storage unit 190 stores at least one of first image data and second image data as polarimetric image data.

The auxiliary data storage unit 180 (an example of a mask region specifying data storage unit) stores mask region specifying data for specifying, as a mask region, a region whose state changes with time in the observation region.
For example, the auxiliary data storage unit 180 stores area classification data for classifying the observation area for each area as mask area specifying data.
For example, the auxiliary data storage unit 180 stores vegetation index data for specifying a vegetation area in the observation area as mask area specifying data.
For example, the mask area specifying data specifies at least one of a vegetation area and a water area as a mask area.

  The coherence image generation unit 110 (an example of a coherence calculation unit) is configured to generate first image data and second image data based on the first image data and the second image data stored in the SAR imaging data storage unit 190. Is calculated for each pixel.

The mask area excluding unit 120 (an example of an entropy alpha calculating unit and a mask area specifying data generating unit) specifies a mask area in the observation area based on the mask area specifying data.
For example, the mask area exclusion unit 120 calculates the entropy of the polarimetric image data and the alpha angle of the polarimetric image data for each pixel. Based on the calculated entropy and alpha angle, the mask area exclusion unit 120 identifies an area that appears in a pixel having a predetermined combination of entropy and alpha angle as a mask area. The mask area excluding unit 120 generates data indicating the specified mask area as mask area specifying data.

The change area specifying unit 130 (an example of the change area specifying unit) specifies an area other than the mask area in the observation area that is reflected in a pixel whose coherence is smaller than a predetermined coherence threshold as the change area.
For example, the change area specifying unit 130 excludes the area classified into a predetermined type based on the area classification data from the change area as a mask area.
For example, the change area specifying unit 130 excludes the vegetation area from the change area as a mask area based on the vegetation index data.

First SAR image data 191 described later is an example of first image data, and second SAR image data 192 described later is an example of second image data.
The mask area corresponds to an area where coherence is always low (for example, a vegetation area or a water area).
A land cover classification map 181, a vegetation index map 182, and an entropy alpha classification map 183, which will be described later, are examples of mask area specifying data. The land cover classification map 181 is an example of observation area classification data, and the vegetation index map 182 is an example of vegetation index data.

FIG. 2 is a diagram illustrating an observation method of an observation area in the first embodiment.
The observation method of the observation area in Embodiment 1 is demonstrated below based on FIG.

Synthetic Aperture Radar (SAR) is mounted on an aircraft, a satellite, or other moving object (flying object) and flies over the sky.
Synthetic aperture radar 200 includes an antenna and observes an observation area by an active method. In the active method, the synthetic aperture radar 200 transmits an electromagnetic wave having a predetermined waveform as an observation signal from the antenna, and receives the observation signal reflected and backscattered in the observation area by the antenna.
The traveling direction of the synthetic aperture radar 200 is referred to as “azimuth direction”, and the traveling direction of the observation signal is referred to as “range direction”.
Further, an angle θ formed by the observation signal reception direction and the vertical direction is referred to as an “observation angle (or incident angle, off-nadir angle)”.

  SAR image data is generated by performing processing such as range compression, range migration correction, and azimuth compression on the waveform data (observation data) representing the observation signal received by the synthetic aperture radar 200.

It is assumed that the synthetic aperture radar 200 observes the same observation area on different observation dates T1 and T2.
Hereinafter, the SAR image data on the observation date T1 is referred to as “first SAR image data 191”, and the SAR image data on the observation date T2 is referred to as “second SAR image data 192”.
In addition, the distance (baseline length) between the observation position of the synthetic aperture radar 200 on the observation date T1 and the observation position of the synthetic aperture radar 200 on the observation date T2 is sufficiently short. This is because if the base line length is long, the observation direction is greatly different, so that the waveform of the observation signal changes greatly, and a region where the state has not changed may be detected as a change region.

For example, when observing with PALSAR (Phased Array type L-band SAR) mounted on the observation satellite “Daichi”, it is desirable that the baseline length is 1 km or less.
Since the observation satellite “Daichi” is orbitally controlled so that the base line length is about 500 meters, the SAR imaging data obtained by the observation of the observation satellite “Daichi” can be used in this embodiment. .

It is assumed that the first SAR image data 191 and the second SAR image data 192 are stored in advance in the SAR image data storage unit 190 of the change area detection apparatus 100 (see FIG. 1).
However, the first SAR imaging data 191 and the second SAR imaging data 192 may not be data obtained by observation with the synthetic aperture radar 200. For example, it may be data obtained by observation with an inverse synthetic aperture radar (ISAR: Inverse SAR) or other observation device.

FIG. 3 is a flowchart showing a change region detection method of the change region detection apparatus 100 according to the first embodiment.
The change area detection method of change area detection apparatus 100 in the first embodiment will be described below with reference to FIG.

  First, an outline of a change area detection method (an example of a change area specifying method) will be described.

The coherence image generation unit 110 calculates the coherence of the observation area based on the first SAR imaging data 191 and the second SAR imaging data 192, and generates a coherence image 193 (S110).
The mask area exclusion unit 120 identifies the vegetation area and the water area included in the observation area as a mask area based on the land cover classification map 181, the vegetation index map 182 or the entropy alpha classification map 183 (S120).
Based on the coherence image 193, the change area specifying unit 130 specifies an area in the observation area (excluding the mask area) where the coherence is smaller than the coherence threshold as a change flow area (S130).

FIG. 4 is a diagram illustrating an example of the first SAR image 191a and the second SAR image 192a according to the first embodiment.
An image obtained by reproducing the first SAR imaging data 191 is referred to as a first SAR image 191a, and an image obtained by reproducing the second SAR imaging data 192 is referred to as a second SAR image 192a.

The first SAR image 191a shows an observation area before inundation, and the second SAR image 192a shows an observation area after inundation where a part of the city area has been inundated.
For the sake of explanation, the first SAR image 191a and the second SAR image 192a clearly show the change area (flooded area), but analyzing the signal intensity of the observation signal clearly extracts the change area. It is known to be difficult.

  The details of the change area detection method (see FIG. 3) will be described below by taking as an example the case of detecting the flooded area shown in the second SAR image 192a.

  In S110, the coherence image generation unit 110 obtains the coherence between the first SAR imaging data 191 and the second SAR imaging data 192 from the first SAR imaging data 191 (or the second SAR imaging data 192). Calculated for each pixel of the image.

The coherence is a value obtained by correlating two pieces of waveform data (first SAR imaging data 191 and second SAR imaging data 192) expressed as a complex function.
The coherence has a value from “0” to “1”, and the closer the two waveform data are, the closer the value is to “1”, and the closer the two waveform data are, the closer the value is to “0”.
In other words, if the coherence is a value close to “1”, the waveform data obtained in the first observation is similar to the waveform data obtained in the second observation, and therefore the first observation and the second observation. It is thought that the ground surface of the observation area has not changed between observations.
Conversely, if the coherence is close to “0”, the waveform data obtained in the first observation is different from the waveform data obtained in the second observation, so the first observation and the second observation. It is thought that the surface of the observation area has changed.

For example, vegetation (such as tree branches and leaves) sways and grows in the wind, and the water surface sways in the wind. For this reason, areas that constantly change like vegetation areas and water areas have coherence values close to “0”.
Moreover, since the state has changed before and after the flooded area and the area where a new building has been built, the coherence is a value close to “0”.

  A method for calculating the coherence will be described later.

  Hereinafter, the image represented using the coherence calculated in S110 is referred to as a “coherence image 193”. The coherence image 193 represents the observation area in terms of coherence instead of luminance (signal intensity).

  It progresses to S120 after S110.

FIG. 5 is a diagram illustrating an example of the coherence image 193 in the first embodiment.
A coherence image 193 as shown in FIG. 5 is obtained for the first SAR image 191a (FIG. 4, before flooding) and the second SAR image 192a (FIG. 4, after flooding).
The coherence image 193 represents mountains, lakes, and flooded areas with values close to “0” (coherence). For this reason, the flooded area cannot be identified from the coherence image 193.

  Returning to FIG. 3, the description of the change region detection method will be continued.

In S120, the mask area exclusion unit 120 specifies the vegetation area and the water area included in the observation area as a mask area based on the land cover classification map 181, the vegetation index map 182 or the entropy alpha classification map 183.
The land cover classification map 181, the vegetation index map 182 and the entropy alpha classification map 183 will be described below.

FIG. 6 is a schematic diagram showing the land cover classification map 181 in the first embodiment.
The land cover classification map 181 is map data showing areas divided into areas such as urban areas, paddy fields, upland fields, lawns, broad-leaved forests, coniferous forests, sandy land, bare land, and water areas.
Based on the land cover classification map 181, the mask area exclusion unit 120 identifies a paddy field, a field, a lawn, a broadleaf forest, a coniferous forest, a sand, a bare land, a water area, and the like in the observation area as a mask area.

FIG. 7 is a schematic diagram showing the vegetation index diagram 182 in the first embodiment.
The vegetation index diagram 182 is map data representing the distribution of vegetation, and is generated using the near infrared rays emitted from the plants as a vegetation index (NDVI: Normalized Difference Vegetation Index).
In FIG. 7, an area where vegetation is active (vegetation area) is indicated by dark shading.
The mask area exclusion unit 120 identifies a vegetation area in the observation area as a mask area based on the vegetation index map 182.

FIG. 8 is a schematic diagram of the entropy alpha classification table 184 in the first embodiment.
The entropy alpha classification diagram 183 is generated based on the entropy alpha classification table 184 as shown in FIG.

The entropy alpha classification table 184 is table data in which entropy, alpha angle, and observation target classification are associated with each other. Entropy represents the scattering complexity of the observed signal, and alpha angle represents the scattering mechanism.
It is known that observation objects can be classified based on entropy and alpha angle (see Non-Patent Document 1).

The entropy and the alpha angle are obtained from observation data obtained by polarimetric observation from the synthetic aperture radar 200. Polarimetry observation is an observation method using a plurality of electromagnetic waves (multi-polarized waves) having different vibration directions as observation signals.
Hereinafter, observation data obtained by polarimetry observation is referred to as “polarimetry observation data 199”. Polarimetry observation data 199 is a type of SAR imaging data.

The mask area excluding unit 120 calculates entropy and alpha angle for each pixel of the SAR image using polarimetry observation data 199 obtained by observing the observation area. Description of the calculation formulas for entropy and alpha angle is omitted (see, for example, Non-Patent Document 2).
The polarimetry observation data 199 is stored in the SAR imaging data storage unit 190.
However, when at least one of the first SAR imaging data 191 and the second SAR imaging data 192 is polarimetry observation data, it is not necessary to prepare another polarimetry observation data 199. The mask area exclusion unit 120 calculates entropy and alpha angle using the first SAR image data 191 or the second SAR image data 192.

The mask area exclusion unit 120 refers to the entropy alpha classification table 184, identifies the classification corresponding to entropy and alpha angle for each pixel, and identifies the range in the SAR image classified as vegetation as the vegetation area.
The mask area exclusion unit 120 generates an entropy alpha classification chart 183 as map data indicating the identified vegetation area, and stores the generated entropy alpha classification chart 183 in the auxiliary data storage section 180. The entropy alpha classification diagram 183 represents the distribution of the vegetation area in the same manner as the land cover classification diagram 181 and the vegetation index diagram 182.

  Returning to FIG. 3, the description of the change region detection method will be continued.

In S120, the mask area exclusion unit 120 identifies the vegetation area and the water area in the observation area shown in the land cover classification map 181, the vegetation index map 182 or the entropy alpha classification map 183 as a mask area.
The mask area exclusion unit 120 masks the mask area in the coherence image 193 in order to exclude the mask area from the change area to be detected (for example, the flooded area). For example, the mask area exclusion unit 120 sets a predetermined mask value indicating a mask area in each pixel of the mask area in the coherence image 193. Further, for example, the mask area excluding unit 120 generates an image corresponding to the coherence image 193 and having a mask value set in the mask area. By comparing the coherence image 193 and the mask image, it is possible to distinguish the mask area and the area other than the mask area from the coherence image 193.
Hereinafter, the coherence image 193 masked in S120 is referred to as a “coherence mask image 194”.
It progresses to S130 after S120.

FIG. 9 is a diagram showing an example of the coherence mask image 194 in the first embodiment.
When the mountains and lakes in the coherence image 193 (see FIG. 5) are masked as mask regions, a coherence mask image 194 as shown in FIG. 9 is obtained.

  Returning to FIG. 3, the description of the change region detection method will be continued.

In S <b> 130, the change area specifying unit 130 compares the coherence shown in the coherence mask image 194 with a predetermined coherence threshold (a value close to “0”) for each pixel. The coherence shown in the coherence mask image 194 is the coherence of pixels other than the mask area.
The change area specifying unit 130 specifies a part in the coherence mask image 194 where the coherence is smaller than the coherence threshold as a change pixel area.
The change area specifying unit 130 specifies an area shown in the change pixel area as the change area. For example, the change area specifying unit 130 corresponds to the pixels in the change pixel area based on information such as the observation position of the synthetic aperture radar 200, the observation angle of the antenna, and the propagation time (time from transmission to reception) of the observation signal. The latitude and longitude are calculated, and the latitude and longitude range of the change area is specified.

For example, the change area specifying unit 130 can specify only the flooded area as the change area by performing threshold processing on the coherence mask image 194 (see FIG. 9).
By S130, the change basin detection method ends.

FIG. 10 is a diagram showing another example of a flowchart showing the change region detection method of the change region detection apparatus 100 according to the first embodiment.
In the change area detection method (see FIG. 3), after generating the coherence image 193 (S110), the mask area is specified (S120), and the change area is specified from areas other than the mask area based on the coherence mask image 194. (S130).
However, as shown in FIG. 10, the order of generating the coherence image 193 (S220) and specifying the mask area (S210) may be changed.

In S210, the mask area excluding unit 120 specifies the vegetation area and the water area included in the observation area as a mask area based on the land cover classification map 181, the vegetation index map 182 or the entropy alpha classification map 183.
In S220, the coherence image generation unit 110 calculates the coherence of the observation area (excluding the mask area) based on the first SAR image data 191 and the second SAR image data 192, and generates a coherence image 193. This coherence image 193 corresponds to the coherence mask image 194 described above.
In S230, based on the coherence image 193, the change area specifying unit 130 specifies an area in the observation area (excluding the mask area) where the coherence is smaller than the coherence threshold as a change flow area.

  In the processing order of FIG. 10, since the coherence is not calculated for the pixels in the mask area in S220, the processing time can be shortened.

  Next, a method for calculating coherence will be described.

Coherence γ (0 ≦ γ ≦ 1) is expressed by the following equation (1) using SAR complex data S1 (first SAR image data 191) and SAR complex data S2 (second SAR image data 192). Calculated.
“<>” Represents an average value, and “*” represents a complex conjugate.

However, the coherence γ _{(i, j)} of the (i, j) pixel is calculated based on the SAR complex data S1 and S2 of the pixel region including the (i, j) pixel. The (i, j) pixel means a pixel located at coordinates (i, j) in the SAR image.
The SAR complex data S of (i, j) pixels is expressed as in the following equation (2). “A” indicates an amplitude (real part), and “e” indicates a complex number (imaginary part).

For example, the coherence γ _{(i, j)} is a SAR of a pixel region of “2 × 2” pixels including (i, j) pixels or a pixel region of “3 × 3” pixels centered on (i, j) pixels. It is calculated based on the complex data S1 and S2.

Expression (3) obtained by developing the above expression (1) is shown below.
(I ′, j ′) indicates the coordinates of the pixel in the target pixel region.

As the target pixel area is increased, the amount of sampling increases and the waveform of the observation signal can be expressed in detail, so that the accuracy of the coherence γ _{(i, j)} is increased. However, the larger the pixel area, the greater the amount of calculation required to calculate the coherence γ _{(i, j)} .
The size of the pixel area may be set in consideration of the accuracy of the coherence γ _{(i, j)} and the calculation amount.

FIG. 11 is a diagram illustrating an example of hardware resources of the change area detection device 100 according to the first embodiment.
In FIG. 11, the change area detection apparatus 100 includes a CPU 911 (Central Processing Unit) (also referred to as a microprocessor or a microcomputer). The CPU 911 is connected to the ROM 913, the RAM 914, the communication board 915, the display device 901, the keyboard 902, the mouse 903, the drive device 904, and the magnetic disk device 920 via the bus 912, and controls these hardware devices. The drive device 904 is a device that reads and writes a storage medium such as an FD (Flexible Disk Drive), a CD (Compact Disc), and a DVD (Digital Versatile Disc).

  The communication board 915 is wired or wirelessly connected to a communication network such as a LAN (Local Area Network), the Internet, or a telephone line.

  The magnetic disk device 920 stores an OS 921 (operating system), a window system 922, a program group 923, and a file group 924.

  The program group 923 includes programs that execute the functions described as “units” in the embodiment. The program is read and executed by the CPU 911. That is, the program causes the computer to function as “to part”, and causes the computer to execute the procedures and methods of “to part”.

  The file group 924 includes various data (input, output, determination result, calculation result, processing result, etc.) used in “˜part” described in the embodiment.

  In the embodiment, arrows included in the configuration diagrams and flowcharts mainly indicate input and output of data and signals.

  In the embodiment, what is described as “to part” may be “to circuit”, “to apparatus”, and “to device”, and “to step”, “to procedure”, and “to processing”. May be. That is, what is described as “to part” may be implemented by any of firmware, software, hardware, or a combination thereof.

The change area detection apparatus 100 can detect a change area (for example, a flooded area) except for a basin that is constantly changing, such as a vegetation area and a water area. Further, the change area that can be detected is not limited to the flooded area, and it is also possible to detect a building collapsed by a fire or an earthquake, or a road with a crack. Furthermore, the change area detection apparatus 100 can be used not only when a disaster occurs but also when a newly constructed building is detected.
The change area detection apparatus 100 can detect a change area using one coherence image generated from two SAR images. For this reason, the change area detection apparatus 100 can be used when three SAR images having different observation periods cannot be obtained or when it is desired to quickly identify a disaster area without waiting for observation of the third SAR image. it can.

  DESCRIPTION OF SYMBOLS 100 Change area detection apparatus, 110 Coherence image generation part, 120 Mask area exclusion part, 130 Change area specification part, 180 Auxiliary data storage part, 181 Land cover classification map, 182 Vegetation index map, 183 Entropy alpha classification map, 184 Entropy alpha Classification table, 190 SAR imaging data storage unit, 191 first SAR imaging data, 191a first SAR image, 192 second SAR imaging data, 192a second SAR image, 193 coherence image, 194 coherence mask image, 199 Polarimetry observation data, 200 synthetic aperture radar, 901 display device, 902 keyboard, 903 mouse, 904 drive device, 911 CPU, 912 bus, 913 ROM, 914 RAM, 915 communication board, 920 magnetism Disk drive, 921 OS, 922 Window system, 923 Program group, 924 File group.

Claims (7)

An image for storing first image data obtained by observing a specific observation area and second image data obtained by observing the observation area at an observation time different from the observation time of the first image data A data storage unit;
A mask region specifying data storage unit for storing mask region specifying data for specifying a region whose state changes with time in the observation region as a mask region;
A coherence calculating unit that calculates, for each pixel, a coherence between the first image data and the second image data based on the first image data and the second image data stored in the image data storage unit; ,
A change area specifying apparatus comprising: a change area specifying unit that specifies an area other than the mask area in the observation area, which is reflected in a pixel whose coherence is smaller than a predetermined coherence threshold, as a change area. The image data storage unit stores polarimetric image data obtained by observing the observation area using a plurality of electromagnetic waves having different vibration directions,
The change area specifying device further includes:
An entropy alpha calculator that calculates the entropy of the polarimetry image data and the alpha angle of the polarimetry image data for each pixel;
Based on the entropy and alpha angle calculated by the entropy alpha calculation unit, a portion appearing in a pixel having a predetermined combination of entropy and alpha angle is specified as the mask region, and data indicating the specified mask region is specified as the mask The change area specifying device according to claim 1, further comprising: a mask area specifying data generating unit that generates the area specifying data.   The change area specifying device according to claim 2, wherein the image data storage unit stores at least one of the first image data and the second image data as the polarimetric image data. The mask area specifying data storage unit stores area classification data for classifying the observation area for each area as the mask area specifying data;
The change area specifying device according to claim 1, wherein the change area specifying unit excludes an area classified into a predetermined type based on the area classification data as the mask area from the change area. The mask area specifying data storage unit stores vegetation index data for specifying a vegetation area in the observation area as the mask area specifying data,
The change area specifying device according to claim 1, wherein the change area specifying unit excludes a vegetation area from the change area as the mask area based on the vegetation index data.   The change area specifying device according to any one of claims 1 to 5, wherein the mask area specifying data specifies at least one of a vegetation area and a water area as a mask area. An image for storing first image data obtained by observing a specific observation area and second image data obtained by observing the observation area at an observation time different from the observation time of the first image data In a change area specifying program of a change area specifying device comprising a data storage section and a mask area specifying data storage section for storing mask area specifying data for specifying a mask area specifying an area whose state changes with time in the observation area,
A coherence calculation process for calculating, for each pixel, a coherence between the first image data and the second image data based on the first image data and the second image data stored in the image data storage unit; ,
A change area specifying program that causes a computer to execute a change area specifying process that specifies an area other than the mask area in the observation area and having a coherence reflected in a pixel smaller than a predetermined coherence threshold as a change area.

Images