Method and system for representing the depths of the floors of the oceans

• US 6,721,694 B1
• Filed: 10/13/1998
• Issued: 04/13/2004
• Est. Priority Date: 10/13/1998
• Status: Expired due to Term

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Lambrecht, Mark A., Molinelli, Eugene J.
• Primary Examiner(s): Thomson, W. D.
• Application Number:
  US09/170,821
• Time in Patent Office: 2,009 Days
• Field of Search: 703/2, 703/9, 703/10, 434/6, 702/17, 702/16, 702/97, 702/100, 702/103, 702/106, 702/152, 702/153, 702/155, 702/156, 702/166, 702/167, 702/189, 702/190, 702/193, 702/194, 702/195-199, 367/7, 367/12, 367/21, 367/13, 367/72, 367/73, 367/87, 367/126, 362/25, 362/159, 362/195, 362/191, 362/180
• US Class Current: 703/2
• CPC Class Codes: G01S 15/89 for mapping or imaging
• CPC Class Codes: G01V 11/00 Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00

Abstract

A method for generating a model of a portion of the floor of a body of water from a plurality of depth measurement sources includes determining an overlap of any of the plurality of depth measurement sources with another depth measurement source and generating a relative shift between any two depth measurement sources that overlap. For each depth measurement source that overlaps with at least one other source, generating an overall shift based on the generated relative shifts. The method also includes generating a desired grid having a plurality of grid nodes and generating a model depth at a plurality of the grid nodes based, at least in part, on a global shift for the measurements of the plurality of sources based on the relative shift. Further, an output is generated of all model depths at respective grid nodes.

Claims

1. A method for generating a model of a portion of a floor of a body of water from a plurality of depth measurements at a plurality of locations from a plurality of depth measurement sources, the method comprising the steps of:

deriving a wavelength spectrum from the plurality of depth measurements from each source of the plurality of depth measurement sources;

fitting each wavelength spectrum to a third order polynomial to produce a fitted wavelength spectrum for each source of the plurality of depth measurement sources;

determining a real world Nyquist wavelength for each source of the plurality of sources based on the fitted wavelength spectrum for each source of the plurality of depth measurement sources;

determining a model grid with variable model grid spacing based on the real world Nyquist wavelengths for the plurality of depth measurement sources; and

determining depths for the model grid based on the plurality of depth measurements from the plurality of depth measurement sources.

2. The method of claim 1, said step of determining depths for the model grid further comprising decimating the plurality of depth measurements from each source based on the real world Nyquist wavelength for each source.

3. The method of claim 1, said step of determining the real world Nyquist wavelength for each source further comprising:

determining a first inflection point for the fitted wavelength spectrum; and

determining the real world Nyquist wavelength for the source based on the first inflection point.

4. The method of claim 3, said step of determining the real world Nyquist wavelength for the source based on the first inflection point further comprising determining the closest wavelength longer than a wavelength of the first inflection point, where the wavelength spectrum at the closest wavelength has an amplitude greater than a threshold amount above a minimum amplitude of the wavelength spectrum.

5. A method for generating a model of a portion of a floor of a body of water from a plurality of depth measurements at a plurality of locations from a plurality of depth measurement sources, the method comprising the steps of:

h. determining a real world Nyquist wavelength associated with each source of the plurality of depth measurement sources;

i. determining a uniform horizontal grid with a grid spacing based on a minimum real world Nyquist wavelength among a plurality of real world Nyquist wavelengths associated with a plurality of overlapping depth measurement sources of the plurality of depth measurement sources;

j. interpolating the plurality of depth measurements for each source of the plurality of overlapping depth measurement sources to the uniform horizontal grid;

a. determining an overlap region between two sources of the plurality of overlapping depth measurement sources;

b. determining a first measure of difference between depths of the two sources in the overlapping region using the depth values interpolated to the uniform horizontal grid;

c. determining a pairwise shift between the two sources that produces a second measure of difference between depths of the two sources using the depth values interpolated to the uniform horizontal grid, which second measure is substantially less than the first measure;

d. determining an error in the direction of shift associated with each source;

e. determining an absolute shift associated with each source based on the pairwise shift and the error associated with each source;

f. generating a plurality of shifted depth measurements from the plurality of depth measurement sources based on the absolute shift associated with each source; and

g. determining depths for the model based on the plurality of shifted depth measurements from the plurality of depth measurement sources.

6. The method of claim 5, said step of determining the pairwise shift further comprises:

generating a set of shifts;

generating a set of measures of differences between the depths of the two sources, wherein each measure in the set of measures is associated with each shift in the set of shifts;

selecting as the pairwise shift a shift of the set of shifts that is associated with the smallest measure of difference among the set of measures of differences; and

determining the second measure based on the smallest measure.

7. The method of claim 6, said step of generating the set of shifts further comprising generating the set of shifts to extend up to a distance corresponding to about twice the maximum error in the direction of shift among the plurality of depth measurement sources.

8. The method of claim 7, said step of determining the pairwise shift further comprises:

determining that the pairwise shift is in a horizontal direction;

generating a set of horizontal shifts that are integral multiples of the grid spacing;

generating a set of measures of differences between the depths of the two sources, wherein each measure in the set of measures is associated with each shift in the set of horizontal shifts;

selecting as the pairwise shift a shift of the set of shifts that is associated with the smallest measure of difference among the set of measures of differences; and

determining the second measure based on the smallest measure.

9. The method of claim 5, wherein:

the first measure and the second measure include a chi-square statistic of the differences between depths of the two sources; and

said step of determining the pairwise shift further comprises

determining whether the chi-square statistic of the second measure is not less than one tenth of the chi square value statistic of the first measure, and

if it is determined that the chi-square statistic of the second measure is not less than one tenth of the chi square value statistic of the first measure, then setting the pairwise shift to zero.

10. The method of claim 5, said step of determining the pairwise shift further comprising:

determining a first correlation between depths of the two sources in the overlapping region;

determining a second correlation between depths of the two sources in an overlapping region after shifting by the pairwise shift between the two sources;

determining whether the second correlation is substantially greater than the first correlation, and

if it is determined that the first correlation is not substantially greater than the second correlation, then setting the pairwise shift to zero.

11. The method of claim 5, wherein steps c, d and e are performed for a two-dimensional horizontal shift;

and the method further comprises repeating steps c, d and e for a vertical shift before performing steps f and g.

12. The method of claim 11, further comprising determining a substantial depth scale difference in an overlap region after shifting the sources based on the absolute horizontal shift and before repeating steps c, d and e for determining the absolute vertical shift.

13. A method for generating a model of a portion of a floor of a body of water from a plurality of depth measurements at a plurality of locations from a plurality of depth measurement sources, the method comprising the steps of:

determining a real world Nyquist wavelength associated with each source of the plurality of depth measurement sources;

determining a uniform grid with grid spacing based on a minimum real world Nyquist wavelength among a plurality of real world Nyquist wavelengths associated with the plurality of depth measurement sources;

interpolating the plurality of depth measurements for each source of the plurality of depth measurement sources to generate interpolated depths on the uniform grid for each source;

filtering out wavelengths less than about ten times the minimum real world Nyquist wavelength from a combination of the interpolated depths on the uniform grid from all sources to generate a low pass surface on the uniform grid;

determining residual depths on the uniform grid by differencing the interpolated depths on the uniform grid for each source;

determining a plurality of semi-variograms for each source of the plurality of depth measurement sources based on the residuals from each source;

determining a model grid with variable model grid spacing based on the real world Nyquist wavelengths for the plurality of depth measurement sources; and

determining depths for the model grid based on the residuals and the plurality of semi-variograms for each source.

14. The method of claim 13, said step of determining the plurality of semi-variograms for each source further comprising determining a first semi-variogram for a first source based on the residuals of the first source at a first set of contiguous grid positions on the uniform grid.

15. The method of claim 14, said step of determining the plurality of semi-variograms for each source further comprising determining a different, second semi-variogram based on the residuals at a different, second set of contiguous grid positions on the uniform grid.

16. The method of claim 15, wherein the number of grid positions in the first set and the number of grid positions in the second set are based on the number of grid positions that fit in an area that is proportional to an integral multiple of the grid spacing.

17. The method of claim 16, wherein the integral multiple is between about ten and about thirty.

18. The method of claim 15, wherein the first set is offset from the second set by about ten times the grid spacing.

19. The method of claim 14, said step of determining depths for the model grid based on the residuals and the plurality of semi-variograms for each source further comprising using the first variogram to generate a model depth value at a particular model grid position if the first set of contiguous grid positions on the uniform grid is within a threshold distance of the particular model grid position.