JP5369817B2 - Polarimetry SAR device and target identification method using polarimetry SAR data - Google Patents

Description

  The present invention relates to a polarimetry SAR apparatus and a target identification method for identifying a target using horizontal polarization and vertical polarization by a synthetic aperture radar (SAR: Synthetic Aperture Radar), and more specifically, to a polarimetry SAR. The polarimetry SAR apparatus and the target object identification method for identifying the shape of the feature being imaged based on the full polarimetry data acquired by the above method regardless of the direction and the imaging direction of the feature.

  In the remote sensing field, where a satellite (aircraft, aircraft, etc.) is used to observe and identify features (surface state and shape), a microwave is used as a sensor to image the surface state from the flying object. A SAR using a band is generally used. Such a method for interpreting the ground state using the SAR reproduces image data from the amplitude and phase data (complex data) of the received wave obtained by the SAR, and interprets the ground state based on the image data. It is.

  In particular, in polarimetry SAR using polarimetry that observes surface conditions by changing the combination of horizontal (H) and vertical (V) vertical polarizations of transmitted waves and HV polarizations of received waves by SAR. Two types of microwaves whose wavefronts are horizontal (H) and vertical (V) are alternately switched and transmitted, and the H component and V component of each reflected wave are received. Thereby, it is possible to obtain full polarimetry data, that is, data corresponding to a total of four polarizations (HH, HV, VH, and VV polarizations) generated by the combination of the horizontal polarization H and the vertical polarization V. it can. The target identification method (related technology) using such full polarimetry data (that is, each data of HH, HV, VH, and VV) is mainly based on classification by scattering matrix decomposition and eigenvalue analysis. In the technique based on such a related technique, the imaged feature is classified based on decomposition into a typical shape, complexity of a scattering form, and the like.

  In addition, the polarimetric SAR image processing for reproducibly reproducing the image of the target by optimally controlling the angle between the plane of polarization of the transmitted wave to the stationary target and the plane of polarization of the received wave reflected from the target. A technique is also disclosed (see, for example, Patent Document 1). Furthermore, a technique of polarimetric SAR image processing that clearly displays a complex-shaped target on a single SAR image by appropriately selecting the polarization planes of the transmitted and received waves for each part of the stationary target is disclosed. (For example, refer to Patent Document 2).

  In addition, the size and shape of the moving object is measured by ISAR (Inverse Synthetic Aperture Radar), which recognizes the image of the moving object using complex data of amplitude information and phase information based on the movement of the target (flying object). There is also disclosed a technique for identifying (see, for example, Patent Document 3). Furthermore, a technique of ISAR image processing that identifies the movement of a moving object by estimating the rotational motion and posture angle of the moving object using a radar is also disclosed (for example, see Patent Document 4).

JP 2008-014735 A JP 2008-232626 A JP 2001-221857 A JP 09-033649 A

  However, in the above related art, when identifying the shape of a feature by polarimetry SAR, the imaged feature can be decomposed into a typical shape in the decomposition of the scattering matrix, but any shape can be processed. The feature cannot be identified. In addition, when identifying the shape of a feature, the feature may be recognized as a feature having a different shape depending on the direction of the feature or the imaging direction, even though the feature has the same shape.

  In the technique described in Patent Document 1, a feature having a specific shape can be identified, but a feature having an arbitrary shape that has been imaged cannot be identified. Furthermore, in the technique described in Patent Document 2, for a target having an arbitrary shape, the polarization plane of the transmission / reception wave must be appropriately selected for each target. In order to identify these, a considerably complicated process is required. That is, when a target having an arbitrary shape is imaged by the polarimetry SAR image processing method described in Patent Document 2, the image processing using the polarimetry SAR becomes extremely complicated, and the polarimetry SAR device is not easy to use. The techniques described in Patent Documents 3 and 4 are image processing techniques using ISAR (Inverse Synthetic Aperture Radar) that identify and identify the movement of a moving object, and identify ground features from a flying object. Therefore, it cannot be applied to a technique for identifying a feature having an arbitrary shape using a polarimetry SAR apparatus.

  The present invention has been made in view of the above circumstances, and polarimetry that can quantitatively identify an imaged feature as having an arbitrary shape while reducing the influence of the direction of the feature and the imaging direction. An object of the present invention is to provide a SAR device and a target identification method.

To achieve the above object, a first configuration of the invention, port Rarimetori SAR and first full polarimetry data is full polarimetric complex image of the feature captured by the (synthetic aperture radar), a known feature A polarimetry SAR device for calculating a similarity with second full polarimetry data that is the full polarimetry complex image data of the image and identifying the shape of the imaged feature based on the calculated similarity , HH, For the first full polarimetry data consisting of four types of HV, VH, and VV, base conversion processing is performed from the HV polarization base to the LR polarization base, and LL, LR, RL, A first basis conversion processing unit that obtains a four-dimensional real vector η by arranging the intensities of RRs, and the second full polarimetry data including four types of HH, HV, VH, and VV A second basis conversion processing unit that performs basis conversion processing from the HV polarization basis to the LR polarization basis and obtains a four-dimensional real vector ξ by arranging the intensities of LL, LR, RL, and RR for each pixel; For each, similarity between the first full polarimetry data subjected to the basis conversion by the first basis conversion processing unit and the second full polarimetry data subjected to the basis conversion by the second basis conversion processing unit A similarity calculation processing unit for calculating the degree, and based on the similarity for each pixel calculated by the similarity calculation processing unit, the shape of the imaged feature is quantitatively identified. It is characterized by having.
Here, HH is reception data observed by a combination of horizontal polarization transmission and horizontal polarization reception, HV is reception data observed by a combination of horizontal polarization transmission and vertical polarization reception, and VH is vertical polarization. Reception data observed by a combination of wave transmission and horizontal polarization reception, VV is reception data observed by a combination of vertical polarization transmission and vertical polarization reception.
LL is reception data converted into a combination of left-handed circular polarization transmission and left-handed circular polarization reception, LR is reception data converted into a combination of left-handed circular polarization transmission and right-handed circular polarization reception, and RL is Received data converted into a combination of right-handed circularly polarized wave transmission and left-handed circularly polarized wave reception, RR is received data converted into a combination of right-handed circularly polarized wave transmission and right-handed circularly polarized wave reception.

The second configuration of the present invention obtains the first full polarimetry data is full polarimetric complex image of the feature captured by the port Rarimetori SAR (Synthetic Aperture Radar), and the acquired first full polarimetry data The degree of similarity with the second full polarimetry data, which is the full polarimetry complex image data of the known feature, is calculated, and polarimetry SAR data for identifying the shape of the imaged feature based on the calculated similarity is obtained. In accordance with the target identification method to be used, base conversion processing is performed from the HV polarization base to the LR polarization base with respect to the first full polarimetry data consisting of four types of HH, HV, VH, and VV. For each pixel, the first step of obtaining the four-dimensional real vector η by arranging the intensities of LL, LR, RL, and RR, and the four steps of HH, HV, VH, and VV A base conversion process is performed on the second full polarimetry data from the HV polarization base to the LR polarization base, and the intensities of LL, LR, RL, and RR are arranged for each pixel to obtain a four-dimensional real vector ξ. A second step of obtaining, for each pixel, the first full polarimetry data subjected to basis conversion by the first step, and the second full polarimetry data subjected to basis conversion by the second step; And calculating the shape of the imaged feature based on the similarity for each pixel calculated in the third step. It is said.
Here, HH is reception data observed by a combination of horizontal polarization transmission and horizontal polarization reception, HV is reception data observed by a combination of horizontal polarization transmission and vertical polarization reception, and VH is vertical polarization transmission.・ Reception data observed with a combination of horizontal polarization reception, VV is reception data observed with a combination of vertical polarization transmission and vertical polarization reception.
LL is reception data converted into a combination of left-handed circularly polarized wave transmission and left-handed circularly polarized wave reception, LR is reception data converted into a combination of left-handed circularly polarized wave transmission and right-handed circularly polarized wave reception, and RL is right-handed rotation Received data converted into a combination of circularly polarized wave transmission and left-handed circularly polarized wave reception, RR is received data converted into a combination of right-handed circularly polarized wave transmission and right-handed circularly polarized wave reception.

  According to the configuration of the present invention, since the imaged feature is identified based on the similarity to the known feature, it is possible to quantitatively identify the imaged feature as having an arbitrary shape. It becomes possible. Furthermore, since this identification is performed based on the similarity to a known feature, the influence of the direction of the feature and the imaging direction can be reduced. For example, if you prepare the full polarimetry data of the tail part of the aircraft, the main wing root part, and the main wing part as a library in which a plurality of data are combined into one, each of the tail part, main wing root part, and main wing part An object similar in shape is extracted from the SAR image according to the similarity. That is, if the actually captured image is an aircraft, the arrangement of objects similar to the respective shapes is not dependent on the direction of the aircraft and the imaging direction, and the tail portion, main wing root portion, and main wing portion of the aircraft are arranged. It becomes the same as the arrangement of the parts.

It is a block diagram which shows the structure of the polarimetry SAR apparatus applied to embodiment of this invention, and the outline | summary of data processing.

  A polarimetry SAR device applied to an embodiment of the present invention is a polarimetry SAR device that identifies a feature using a synthetic aperture radar (SAR), and has a plane of polarization that is a horizontal (H) polarization and a vertical (V) polarization. Two types of microwaves are alternately transmitted to the feature. And the polarimetric complex image of the imaged feature is acquired by receiving the H component and the V component of the reflected wave from the feature. Furthermore, the degree of similarity between the acquired polarimetric complex image and the polarimetric complex image of known features stored in a library is calculated. Thereby, when the degree of similarity is high, the imaged feature can be quantitatively identified by the known feature.

  In other words, this polarimetry SAR device includes full polarimetry data obtained by polarimetry SAR that generates H-polarized waves and V-polarized waves of transmission waves and reflected waves (received waves) with respect to a target feature. The shape of the imaged feature is identified by a known feature by comparing with the full polarimetry data of the feature. Thereby, the shape of the feature being imaged can be identified with high accuracy without depending on the direction of the feature and the imaging direction.

  A first basis conversion processing unit that performs a basis conversion process from a linearly polarized wave basis to a circularly polarized wave basis for the first full polarimetric data, and a linear deviation for the second full polarimetric data. A second basis conversion processing unit that performs basis conversion processing from a wave basis to a circularly polarized wave basis; the first full polarimetry data that has undergone basis conversion by the first basis conversion processing unit; and the second basis A similarity calculation processing unit that calculates the similarity to the second full polarimetry data that has undergone the base conversion by the conversion processing unit, so that based on the similarity calculated by the similarity calculation processing unit The configuration for quantitatively identifying the shape of the imaged feature is realized.

  In order to realize the object of the present invention, specifically, from the HV polarization base to the LR polarization base with respect to the first full polarimetry data consisting of four types of HH, HV, VH, and VV. A base conversion process is performed, and LL. For the first basis conversion processing unit for obtaining the four-dimensional real vector η by arranging the intensities of LR, RL, and RR, and for the second full polarimetry data consisting of four types of HH, HV, VH, and VV, HV A base conversion process is performed from the polarization base to the LR polarization base, and LL. A second basis conversion processing unit that obtains a four-dimensional real vector ξ by arranging the intensities of LR, RL, and RR; and the first full polarimetry data that has undergone basis conversion by the first basis conversion processing unit for each pixel. And a similarity calculation processing unit that calculates a similarity with the second full polarimetry data subjected to the basis conversion by the second basis conversion processing unit.

Here, HH is reception data observed by a combination of horizontal polarization transmission and horizontal polarization reception, HV is reception data observed by a combination of horizontal polarization transmission and vertical polarization reception, and VH is vertical polarization. Reception data observed by a combination of wave transmission and horizontal polarization reception, VV is reception data observed by a combination of vertical polarization transmission and vertical polarization reception.
LL is reception data converted into a combination of left-handed circular polarization transmission and left-handed circular polarization reception, LR is reception data converted into a combination of left-handed circular polarization transmission and right-handed circular polarization reception, and RL is Received data converted into a combination of right-handed circularly polarized wave transmission and left-handed circularly polarized wave reception, RR is received data converted into a combination of right-handed circularly polarized wave transmission and right-handed circularly polarized wave reception.

Embodiment

  Hereinafter, embodiments of a polarimetry SAR apparatus of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an outline of a configuration and data processing of a polarimetry SAR apparatus applied to an embodiment of the present invention.

  First, the configuration of the polarimetry SAR apparatus shown in FIG. 1 will be described. The polarimetry SAR device includes a transmitter 1, a transmission wave changeover switch 2, circulators 3 and 4, antennas 5 and 6, receivers 7 and 8, data processing units 9 and 10, separation processing units 11 and 12, reproduction processing unit 13, 14, 15, 16, storage devices 17, 18, 21, 22, 25, 26, 29, 30, a first basis conversion processing unit 19, a second basis conversion processing unit 20, an intensity conversion processing unit 23, 24, The filter processing units 27 and 28 and the similarity calculation processing unit 31 are provided at least.

  Next, basic functions of each component will be described. The transmitter 1 generates a transmission signal (chirp signal) subjected to linear frequency modulation in synchronization with a pulse repetition frequency serving as a transmission trigger signal, and outputs this transmission signal to the transmission wave changeover switch 2. The transmission wave changeover switch 2 switches the transmission signal received from the transmitter 1 alternately for each pulse repetition frequency, and sends it to one of the antennas 5 and 6 via one of the circulators 3 and 4.

  The antenna 5 is an antenna for horizontal polarization (H polarization), transmits a transmission wave of H polarization to a target, and receives an HH reception wave and a VH reception wave as reflected waves from the target. The antenna 6 is an antenna for vertical polarization (V polarization), transmits a transmission wave of V polarization to a target, and receives an HV reception wave and a VV reception wave as a reflection wave from the target. That is, the antenna 5 and the antenna 6 are alternately switched in synchronization with the pulse repetition frequency, respectively, and transmit the H polarization and the V polarization as transmission waves toward the target, respectively, and receive the HH and VH reception waves, respectively. HV and VV received waves are received.

  More specifically, the reflected waves from the target (for example, the ground surface) irradiated with the transmission waves of the H-polarized wave and the V-polarized light of the antenna 5 and the antenna 6 are formed in the shape of the target (ground surface). Accordingly, both horizontal polarization (H polarization) and vertical polarization (V polarization) transmission waves include a horizontal polarization (H polarization) component and a vertical polarization (V polarization) component. . Therefore, at the time corresponding to the transmission timing of each transmission wave, the H polarization component and the V polarization component are converted into the HH reception wave (H transmission-H reception) and the HV reception wave in each of the antenna 5 and the antenna 6. (H transmission-V reception), VH reception wave (V transmission-H reception), or VV reception wave (V transmission-V reception).

  The receiver 7 receives the HH reception wave (H transmission-H reception) and the VH reception wave (V transmission-H reception) received by the antenna 5 through the circulator 3, performs a reception process, and performs a data processing unit. Output to 9. On the other hand, the receiver 8 receives the HV reception wave (H transmission-V reception) and the VV reception wave (V transmission-V reception) received by the antenna 6 through the circulator 4, performs reception processing, and performs data processing. Output to the processing unit 10.

  The data processing unit 9 stores HH reception wave (H transmission-H reception) data and VH reception wave (V transmission-H reception) data from the receiver 7 as SAR data. The HV reception wave (H transmission-V reception) data and VV reception wave (V transmission-V reception) data from the machine 8 are stored as SAR data, and output to the separation processing units 11 and 12, respectively.

  The separation processing unit 11 separates the SAR data from the data processing unit 9 into the SAR data of the HH reception wave and the SAR data of the VH reception wave, and outputs them to the reproduction processing units 13 and 14. On the other hand, the separation processing unit 12 separates the SAR data from the data processing unit 10 into SAR data of the HV reception wave and SAR data of the VV reception wave, and outputs them to the reproduction processing units 15 and 16.

  Based on the respective SAR data (respective received waves of HH, VH, HV, and VV) output from the separation processing units 11 and 12, the reproduction processing units 13, 14, 15, and 16 generate each as a complex image. Then, it is stored in the storage device 17 as a fully polarimetric complex image (HH, VH, HV, VV).

  As shown in FIG. 1, the first basis conversion processing unit 19 uses the first full polarimetric complex image (HH, VH, HV, VV) of the SAR data stored in the storage device 17 to perform linear polarization. A base conversion process is performed from a certain HV (Horizontal-Vertical) polarization basis to a circularly polarized LR (Lefr-right: left turn-right turn) polarization base, and the first LR basis converted first is obtained. The four types of complex data (LL, LR, RL, RR) of the full polarimetric complex image are stored in the storage device 21. Here, LL is reception data converted into a combination of L transmission and L reception, LR is reception data converted into a combination of L transmission and R reception, and RL is converted into a combination of R transmission and L reception. The received data RR means received data converted into a combination of R transmission and R reception.

  The intensity change processing unit 23 converts the first full polarimetric complex image (LL, LR, RL, RR) converted to the LR polarization base into an intensity image (LL, LR, RL, RR), respectively, and a storage device. 25. Further, the filter processing unit 27 performs speckle filter processing on these intensity images (LL, LR, RL, RR) and stores them in the storage device 29.

  On the other hand, the storage device 18 stores the SAR data (HH, VH, HV, VV) of the second full polarimetric complex image of the known feature, and the basis conversion processing unit 20 stores the second SAR data stored in the storage device 18. For the SAR data (HH, VH, HV, VV) of the full polarimetric complex image of, the base conversion processing from the HV polarization base to the LR polarization base is performed, and the second full polarimetry complex image subjected to the LR polarization base conversion is converted. Four types of complex data (LL, LR, RL, RR) are stored in the storage device 22.

  The intensity change processing unit 24 converts the second full polarimetric complex image (LL, LR, RL, RR) converted into the LR polarization base into an intensity image (LL, LR, RL, RR), respectively, and a storage device. 26 is stored. Then, the filter processing unit 28 applies a speckle filter to these intensity images (LL, LR, RL, RR) and stores them in the storage device 30.

  The similarity calculation processing unit 31 is a feature data (first full polarimetry complex image) captured by the polarimetry SAR extracted from the storage device 29 and a known feature data (second image) extracted from the storage device 30. A similarity degree with a full polarimetric complex image) is calculated and a similarity degree image is output. At this time, the similarity calculation processing unit 31 calculates the similarity for each pixel, and outputs an image representing the similarity of each pixel by shading.

  Next, the operation of the polarimetry SAR apparatus shown in FIG. 1 will be described. Two transmission waves from one transmitter 1 are switched for each pulse repetition frequency by a transmission wave changeover switch 2, and alternately polarized from two antennas 5 and 6 via circulators 3 and 4. The transmission wave and the V polarization transmission wave are transmitted toward a target not shown. The transmission waves and reception waves transmitted and received by the antennas 5 and 6 are distributed to the single transmitter 1 and the two receivers 7 and 8 by the circulators 3 and 4.

  The H polarization component and the V polarization component reflected from the target are received by the two antennas 5 and 6 as reception waves, respectively, and subjected to reception processing by the receivers 7 and 8 to be two data processing units. 9 and 10 are recorded as SAR data. The SAR data recorded in the two data processing units 9 and 10 are respectively (HH) SAR data, (VH) SAR data, (HV) SAR data, and (VV) in the separation processing units 11 and 12. The data is separated into SAR data and input to the four reproduction processing units 13, 14, 15, and 16.

  In addition, (HH) SAR data, (VH) SAR data, (HV) SAR data, and (VV) SAR data are generated as respective complex images by the four reproduction processing units 13, 14, 15, and 16. The SAR data (HH, VH, HV, VV) of the first full polarimetric complex image is stored in the storage device 17. Through the processing as described above, the SAR data of the first full polarimetry complex image acquired by the polarimetry SAR is stored in the storage device 17.

  Next, using the SAR data (HH, VH, HV, VV) of the first full polarimetric complex image, the first basis conversion processing unit 19 performs basis conversion processing from the HV polarization basis to the LR polarization basis. Then, the first full polarimetric complex image data (LL, LR, RL, RR) subjected to LR basis conversion is stored in the storage device 21. Further, the SAR data (LL, LR, RL, RR) of the first full polarimetric complex image converted into the LR polarization base is converted into intensity images (LL, LR, RL, RR) by the intensity change processing unit 23, respectively. And stored in the storage device 25. Then, a speckle filter is applied to these intensity images (LL, LR, RL, RR) by the filter processing unit 27 and stored in the storage device 29.

  On the other hand, for the second full polarimetric complex image (HH, VH, HV, VV) of the known feature stored in the storage device 18, the second basis conversion processing unit 20 converts the LR polarization from the HV polarization base. Base conversion processing is performed on the base, and LR base conversion data (LL, LR, RL, RR) subjected to LR polarization base is stored in the storage device 22. Then, the second full polarimetric complex image data (LL, LR, RL, RR) subjected to the LR basis conversion is converted into intensity images (LL, LR, RL, RR) by the intensity change processing unit 24, respectively, and stored therein. 26. Further, a speckle filter is applied to these intensity images (LL, LR, RL, RR) by the filter processing unit 28 and stored in the storage device 30.

  Then, by the similarity calculation processing unit 31, the feature data (first full polarimetric complex image) captured by the polarimetry SAR stored in the storage device 29 and the known feature stored in the storage device 30. Similarity with the data (second full polarimetric complex image) is calculated. At this time, the degree of similarity is calculated for each pixel, and an image representing the degree of similarity of each pixel is output by shading.

Next, a process of obtaining a full polarimetric complex image by polarimetry SAR and calculating the similarity will be described using mathematical expressions. Equation (1) is an equation representing the basis conversion from the HV polarization basis to the LR polarization basis. That is, for the four types of complex data HH, HV, VH, and VV, which are SAR data of known features stored in the storage device 18, the HV polarization base reception data ( S _HH , S _VH , S _HV , S _VV ) are converted into LR polarization base reception data (S _LL , S _LR , S _RL , S _RR ).

・・・（１）

... (1)

However, S _HH , S _VH , S _HV , and S _VV are received data (complex data) for SAR data HH, HV, VH, and VV, respectively, and i represents an imaginary unit.
S _LL , S _LR , S _RL , and S _RR are base data (complex data) converted into LR polarization bases, respectively.

  In addition, a library is prepared by arranging the intensities of LL, LR, RL, and RR of one pixel of the second full polarimetric complex image converted to the LR polarization base to form a four-dimensional real vector of known features. deep. The four-dimensional real vector in this known feature is denoted as ξ. The four-dimensional real vector ξ is expressed by equation (2), and the absolute value | ξ | is expressed by equation (3).

・・・（２）

... (2)

・・・（３）

... (3)

On the other hand, a SAR data of the first full polarimetric complex image stored in the storage device 17 is observed from the feature HH, HV, VH, 4 kinds of VV complex data _{(S HH, S VH, S} HV, S _VV ), the base data of S _LL , S _LR , S _RL , S _RR is calculated by the above-described equation (1), and the polarization is changed from the HV polarization base reception data to the LR polarization base reception data. Convert the base.
Here, the intensities of LL, LR, RL, and RR of one pixel are arranged to form a four-dimensional real vector of the imaging feature. This is expressed as a four-dimensional real vector η in the imaged feature. The four-dimensional real vector η is expressed by equation (2), and the absolute value | η | is expressed by equation (3).

  Based on the known feature vector ξ and the imaged feature vector η obtained as described above, the similarity calculation processing unit 31 calculates the similarity r [%] by the following equation (4). Is calculated.

・・・（４）
ただし、（ξ/|ξ|，η/|η|）はベクトルの内積を表わす。

... (4)
However, (ξ / | ξ |, η / | η |) represents an inner product of vectors.

  Through the processing as described above, the imaged feature is quantitatively identified by the known feature. This similarity r [%] reduces the influence of the direction of the feature and the imaging direction. In this way, the first full polarimetry data obtained by the polarimetry SAR of the target feature is compared with the second full polarimetry data of the known feature, and the imaged feature is obtained. The shape is identified. Thereby, the shape of the feature being imaged can be identified with high accuracy without depending on the direction of the feature and the imaging direction.

  Note that the target identification method using the polarimetry SAR data described above is realized by a computer reading a program. Therefore, each process of the target identification method using the polarimetry SAR data is stored in a computer-readable recording medium in the form of a program, and the program is read and executed by the computer. Each process is performed. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM (Compact Disk-Read Only Memory), a DVD-ROM (Digital Versatile Disk-Read Only Memory), a semiconductor memory, or the like.

  Although the embodiments of the present invention have been described in detail with reference to the drawings, the specific configuration of the present invention is not limited to these embodiments, and the design does not depart from the spirit of the present invention. These changes are included in the present invention. For example, the program may be distributed to an external computer via a communication line, and the computer that has received the distribution may execute the program. In addition, the features are not limited to objects on the ground, but also include flying objects flying in a low sky when viewed from an artificial satellite or the like.

  Since the polarimetry SAR apparatus of the present invention can identify and identify ground features with high accuracy, it can be effectively used for creating maps based on aerial photographs.

DESCRIPTION OF SYMBOLS 1 Transmitter 2 Transmission wave changeover switch 3, 4 Circulator 5, 6 Antenna 7, 8 Receiver 9, 10 Data processing part 11, 12 Separation processing part 13, 14, 15, 16 Reproduction processing part 17, 18, 21, 22 , 25, 26, 29, 30 Storage device 19 First basis conversion processing unit 20 Second basis conversion processing unit 23, 24 Intensity conversion processing unit 27, 28 Filter processing unit 31 Similarity calculation processing unit

Claims (7)

Similarity between first full polarimetry data that is a full polarimetry complex image of a feature imaged by polarimetry SAR (synthetic aperture radar) and second full polarimetry data that is a full polarimetry complex image data of a known feature And a polarimetry SAR device for identifying the shape of the imaged feature based on the calculated similarity ,
The first full polarimetry data consisting of four types of HH, HV, VH, and VV is subjected to base conversion processing from the HV polarization base to the LR polarization base, and LL, LR, A first basis conversion processing unit that obtains a four-dimensional real vector η by arranging the intensities of RL and RR;
The second full polarimetry data consisting of four types of HH, HV, VH, and VV is subjected to basis conversion processing from the HV polarization base to the LR polarization base, and LL, LR, A second basis conversion processing unit that arranges the strengths of RL and RR to obtain a four-dimensional real vector ξ;
For each pixel, the first full polarimetry data base-transformed by the first base transform processing unit and the second full polarimetry data base-transformed by the second base conversion processing unit A similarity calculation processing unit for calculating the similarity,
A polarimetric SAR device configured to quantitatively identify the shape of the imaged feature based on the similarity for each pixel calculated by the similarity calculation processing unit .
Here, HH is reception data observed by a combination of horizontal polarization transmission and horizontal polarization reception, HV is reception data observed by a combination of horizontal polarization transmission and vertical polarization reception, and VH is vertical polarization. Reception data observed by a combination of wave transmission and horizontal polarization reception, VV is reception data observed by a combination of vertical polarization transmission and vertical polarization reception.
LL is reception data converted into a combination of left-handed circular polarization transmission and left-handed circular polarization reception, LR is reception data converted into a combination of left-handed circular polarization transmission and right-handed circular polarization reception, and RL is Received data converted into a combination of right-handed circularly polarized wave transmission and left-handed circularly polarized wave reception, RR is received data converted into a combination of right-handed circularly polarized wave transmission and right-handed circularly polarized wave reception. First full polarimetry data that is a full polarimetry complex image of a feature imaged by polarimetry SAR (synthetic aperture radar), and full polarimetry complex image data of a known feature that is acquired in advance and stored in a library A polarimetry SAR device that calculates the degree of similarity with second full polarimetry data and quantitatively identifies the shape of the imaged feature based on the calculated degree of similarity ,
The first full polarimetry data consisting of four types of HH, HV, VH, and VV is subjected to base conversion processing from the HV polarization base to the LR polarization base, and LL, LR, A first basis conversion processing unit that obtains a four-dimensional real vector η by arranging the intensities of RL and RR;
The second full polarimetry data consisting of four types of HH, HV, VH, and VV is subjected to basis conversion processing from the HV polarization base to the LR polarization base, and LL, LR, A second basis conversion processing unit that arranges the strengths of RL and RR to obtain a four-dimensional real vector ξ;
For each pixel, the first full polarimetry data base-transformed by the first base transform processing unit and the second full polarimetry data base-transformed by the second base conversion processing unit A similarity calculation processing unit for calculating the similarity,
A polarimetric SAR device configured to quantitatively identify the shape of the imaged feature based on the similarity for each pixel calculated by the similarity calculation processing unit .
Here, HH is reception data observed by a combination of horizontal polarization transmission and horizontal polarization reception, HV is reception data observed by a combination of horizontal polarization transmission and vertical polarization reception, and VH is vertical polarization. Reception data observed by a combination of wave transmission and horizontal polarization reception, VV is reception data observed by a combination of vertical polarization transmission and vertical polarization reception.
LL is reception data converted into a combination of left-handed circular polarization transmission and left-handed circular polarization reception, LR is reception data converted into a combination of left-handed circular polarization transmission and right-handed circular polarization reception, and RL is Received data converted into a combination of right-handed circularly polarized wave transmission and left-handed circularly polarized wave reception, RR is received data converted into a combination of right-handed circularly polarized wave transmission and right-handed circularly polarized wave reception. The similarity r [%] is

The polarimetric SAR device according to claim 1 or 2 , characterized by:
ξ: Four-dimensional real vector of the known feature η: Four-dimensional real vector of the imaged feature (ξ / | ξ |, η / | η |): inner product of the vectors The polarimetric SAR device according to claim 1 or 2 , wherein the similarity of each pixel is converted into gray and displayed as an image. First full polarimetry data, which is a full polarimetry complex image of a feature imaged by polarimetry SAR (synthetic aperture radar), is acquired, and the acquired first full polarimetry data and full polarimetry complex image data of a known feature are obtained. A target identification method using polarimetry SAR data for calculating similarity with certain second full polarimetry data and identifying the shape of the imaged feature based on the calculated similarity,
The first full polarimetry data consisting of four types of HH, HV, VH, and VV is subjected to base conversion processing from the HV polarization base to the LR polarization base, and LL, LR, A first step of arranging the intensities of RL and RR to obtain a four-dimensional real vector η;
The second full polarimetry data consisting of four types of HH, HV, VH, and VV is subjected to basis conversion processing from the HV polarization base to the LR polarization base, and LL, LR, A second step of arranging the intensities of RL and RR to obtain a four-dimensional real vector ξ;
For each pixel, a first degree of similarity between the first full polarimetry data subjected to the base conversion in the first step and the second full polarimetry data subjected to the base conversion in the second step is calculated. 3 steps,
A target identification method using polarimetric SAR data , wherein the shape of the imaged feature is quantitatively identified based on the similarity for each pixel calculated in the third step .
Here, HH is reception data observed by a combination of horizontal polarization transmission and horizontal polarization reception, HV is reception data observed by a combination of horizontal polarization transmission and vertical polarization reception, and VH is vertical polarization transmission.・ Reception data observed with a combination of horizontal polarization reception, VV is reception data observed with a combination of vertical polarization transmission and vertical polarization reception.
LL is reception data converted into a combination of left-handed circularly polarized wave transmission and left-handed circularly polarized wave reception, LR is reception data converted into a combination of left-handed circularly polarized wave transmission and right-handed circularly polarized wave reception, and RL is right-handed rotation Received data converted into a combination of circularly polarized wave transmission and left-handed circularly polarized wave reception, RR is received data converted into a combination of right-handed circularly polarized wave transmission and right-handed circularly polarized wave reception. The similarity r [%] is

The target identification method using polarimetry SAR data according to claim 5, wherein
ξ: Four-dimensional real vector of the known feature η: Four-dimensional real vector of the imaged feature (ξ / | ξ |, η / | η |): inner product of the vectors 6. The target identification method using polarimetric SAR data according to claim 5 , wherein the similarity of each pixel is converted into gray and displayed as an image.

Images