Apparatus and method for image navigation and registration of geostationary remote sensing satellites

{
  "publication_number": "US10262403B2",
  "country": "US",
  "kind": "B2",
  "title": "Apparatus and method for image navigation and registration of geostationary remote sensing satellites",
  "abstract": "Provided is an apparatus for geometrically correcting an image from a geostationary remote sensing satellite, the apparatus being configured to perform the image navigation and registration on image data from a geostationary remote sensing satellite. The apparatus includes the landmark determination part configured to acquire landmark information associated with a first image and the navigation filter configured to calculate a state vector for correcting at least one of the attitude error, the orbit error, and the payload misalignment error with respect to the first image based on the landmark information.",
  "claims": [
    "1. An apparatus for geometrically correcting an image from a geostationary remote sensing satellite, the apparatus comprising: a landmark determination part configured to acquire landmark information associated with a first image captured by a geostationary remote sensing satellite; and a navigation filter configured to calculate a state vector for processing a geometric correction on the first image by correcting at least one of an attitude error, an orbit error, and a payload misalignment error associated with the first image based on the landmark information; wherein when the navigation filter does not receive an input of orbit information from an outside source, the navigation filter is configured to autonomously estimate orbit information associated with the first image using Kalman filter orbit information, wherein the Kalman filter orbit information is calculated based on the landmark information, and wherein the navigation filter is configured to calculate the state vector including the estimated orbit information.",
    "2. The apparatus of claim 1, wherein the landmark determination part is configured to select a first position of at least one landmark from the first image, calculate a difference between the selected first position and a second position, and acquire the landmark information based on a result of the calculating, the second position being an actual landmark position corresponding to the first position.",
    "3. The apparatus of claim 1, further comprising: a preprocessor configured to generate the first image by performing a radiometric calibration on an image received from a geostationary remote sensing satellite.",
    "4. The apparatus of claim 1, wherein the navigation filter is configured to calculate the state vector using a Kalman filter algorithm.",
    "5. The apparatus of claim 1, wherein when the orbit information is an input from an outside source, the navigation filter is configured to additionally refine the orbit information using Kalman filter orbit information that is calculated based on the landmark information, and calculate the state vector including the additionally refined orbit information.",
    "6. The apparatus of claim 1, further comprising: a resampler configured to resample pixel positions of the first image based on the calculated state vector.",
    "7. A method of geometrically correcting an image from a geostationary remote sensing satellite, the method comprising: acquiring landmark information associated with a first image, the first image captured by a geostationary remote sensing satellite; and calculating a state vector for processing a geometric correction on the first image by correcting at least one of an attitude error, an orbit error, and a payload misalignment error associated with the first image based on the landmark information; wherein when orbit information is not received from an outside source, the calculating the state vector includes estimating orbit information associated with the first image using Kalman filter orbit information that is calculated based on the landmark information and calculating the state vector including the estimated orbit information.",
    "8. The method of claim 7, further comprising: preprocessing for generating the first image by performing a radiometric calibration on an image received from a geostationary remote sensing satellite.",
    "9. The method of claim 7, wherein the acquiring of the landmark information includes selecting a first position of at least one landmark from the first image, calculating a difference between the selected first position and a second position, and acquiring the landmark information based on a result of the calculating, the second position being an actual landmark position corresponding to the first position.",
    "10. The method of claim 7, wherein when the orbit information is an input from an outside source, the calculating of the state vector includes additionally refining the orbit information using Kalman filter orbit information that is calculated based on the landmark information and calculating the state vector including the additionally refined orbit information.",
    "11. The method of claim 7, wherein the calculating of the state vector includes calculating the state vector using a Kalman filter algorithm.",
    "12. The method of claim 7, further comprising: resampling pixel positions of the first image based on the calculated state vector.",
    "13. A non-transitory computer-readable medium storing program instructions for controlling a processor to perform a method of geometrically correcting an image from a geostationary remote sensing satellite, the method comprising: acquiring landmark information associated with a first image, the first image captured by a geostationary remote sensing satellite; and calculating a state vector for processing a geometric correction on the first image by correcting at least one of an attitude error, an orbit error, and a payload misalignment error associated with the first image based on the landmark information; wherein when orbit information is not received from an outside source, the calculating the state vector includes estimating orbit information associated with the first image using Kalman filter orbit information that is calculated based on the landmark information and calculating the state vector including the estimated orbit information.",
    "14. The non-transitory computer-readable medium of claim 13, wherein the method further comprising: preprocessing for generating the first image by performing a radiometric calibration on an image received from a geostationary remote sensing satellite.",
    "15. The non-transitory computer-readable medium of claim 13, wherein the acquiring of the landmark information includes selecting a first position of at least one landmark from the first image, calculating a difference between the selected first position and a second position, and acquiring the landmark information based on a result of the calculating, the second position being an actual landmark position corresponding to the first position.",
    "16. The non-transitory computer-readable medium of claim 13, wherein when the orbit information is an input from an outside source, the calculating of the state vector includes additionally refining the orbit information using Kalman filter orbit information that is calculated based on the landmark information and calculating the state vector including the additionally refined orbit information.",
    "17. The non-transitory computer-readable medium of claim 13, wherein the calculating of the state vector includes calculating the state vector using a Kalman filter algorithm."
  ],
  "description_excerpt": "The following descriptions relate to technology for performing image navigation and registration on image data from a geostationary remote sensing satellite and, more particularly, to a method of correcting a geometric distortion in an image from a geostationary remote sensing satellite based on landmark information.\n\nIn general, an observation satellite, or a remote sensing satellite, located on the geostationary orbit may employ the image navigation and registration (INR) scheme to correct a geometric distortion in a satellite image for providing accurate observation information. A satellite INR system may model an error-caused process with respect to a pixel position in an image and correct the error such that the error is maintained to be within an allowable range. For example, the attitude, orbit, and satellite payload misalignment errors may be the source of the errors and the targets to be corrected.\n\nIn the INR system, the reference point selection may affect the overall system configuration, the design of interfaces between the subsystems, and various other aspects such as the developmental composition, costs, a future operation plan, and the like. When the INR system for geostationary orbital 3-axis satellite was first developed in the early 1990s, the initial INR system has employed the reference points composed of both landmarks and stars (GOES-I to M satellites in the US). This INR system using the combination of the landmarks and the starts has been continually applied to GOES-N to P in the US and MASAT-1R in Japan.",
  "cpc": [
    "G06T 5/20",
    "B64G 1/1021",
    "B64G 1/1028",
    "B64G 1/242",
    "B64G 1/244",
    "B64G 2001/1028",
    "G01C 21/24",
    "G01S 19/20",
    "G01S 19/40",
    "G06T 2207/20024"
  ],
  "ipc": [
    "B64G 1/10",
    "G01C 21/24",
    "G06T 5/20"
  ],
  "assignees": [
    "Korea Aerospace Research Institute KARI"
  ],
  "inventors": [
    "Handol Kim",
    "Ahmed Kamel",
    "Dochul Yang",
    "Chulmin Park",
    "Jin WOO"
  ],
  "filing_date": "2017-04-24",
  "publication_date": "2019-04-16",
  "grant_date": "2019-04-16",
  "priority_date": "2017-04-24",
  "application_number": "US-201715495228-A",
  "family_id": "63852329",
  "cited_by_count": 2,
  "citations": [
    "US20110196550A1",
    "US20120078510A1"
  ]
}