Efficient re-sampling of discrete curves

1. A computer-implemented method for re-sampling discrete curves, wherein the discrete curves correspond to data, the method comprising:

normalizing a plurality of discrete curves, wherein each discrete curve comprises a respective sequence of points;

computing a weight function based on the plurality of discrete curves, wherein the weight function enhances differences between weighted discrete curves;

determining a set of orthonormal polynomials based on the computed weight function, wherein the set of orthonormal polynomials comprises a set of basis functions;

determining values for a plurality of zeros for one of the set of orthonormal polynomials, wherein the values for the plurality of zeros comprise re-sampling points for the plurality of discrete curves; and

re-sampling each of the plurality of discrete curves based on the determined values of the plurality of zeros, thereby generating re-sampled discrete curves;

wherein the re-sampled discrete curves are usable to analyze the data.

2. The method of claim 1 , further comprising:

prior to said normalizing, uniformly re-sampling each of the plurality of discrete curves, wherein, after said uniform re-sampling, each of the plurality of discrete curves has the same number of points.

3. The method of claim 1 , wherein each respective sequence of points comprises a sequence of complex values.

4. The method of claim 1 , wherein said normalizing comprises normalizing one or more of:

energy of the discrete curves; and

length of the discrete curves.

5. The method of claim 1 , wherein said computing the weight function comprises:

computing a weight vector w, such that

< c n ,c m > w =δ nm

wherein c n and C m comprise any two discrete curves of the normalized discrete curves; and

wherein δ nm is the Kronecker delta.

6. The method of claim 1 , wherein components of the weight function are non-negative.

7. The method of claim 1 , wherein the weight function comprises a Sturm-Liouville weight function.

8. The method of claim 1 , wherein the re-sampling points for the plurality of discrete curves comprise a set of optimal re-sampling points for the plurality of discrete curves.

9. The method of claim 1 , wherein the re-sampling points for the plurality of discrete curves comprise a re-sampling scheme that enhances differences between the plurality of discrete curves.

10. The method of claim 1 , wherein the plurality of discrete curves comprise one or more of:

image objects from one or more images; and

data objects from one or more data sets.

11. A computer readable medium which stores program instructions for re-sampling discrete curves, wherein the discrete curves correspond to data, wherein the program instructions are executable by a processor to perform:

normalizing a plurality of discrete curves, wherein each discrete curve comprises a respective sequence of points;

computing a weight function based on the plurality of discrete curves, wherein the weight function enhances differences between weighted discrete curves;

determining a set of orthonormal polynomials based on the computed weight function, wherein the set of orthonormal polynomials comprises a set of basis functions;

determining values for a plurality of zeros for one of the set of orthonormal polynomials, wherein the values for the plurality of zeros comprise re-sampling points for the plurality of discrete curves; and

re-sampling each of the plurality of discrete curves based on the determined values of the plurality of zeros, thereby generating re-sampled discrete curves;

wherein the re-sampled discrete curves are usable to analyze the data.

12. The computer readable medium of claim 11 , wherein the program instructions are further executable by a processor to perform:

prior to said normalizing, uniformly re-sampling each of the plurality of discrete curves, wherein, after said uniform re-sampling, each of the plurality of discrete curves has the same number of points.

13. The computer readable medium of claim 11 , wherein each respective sequence of points comprises a sequence of complex values.

14. The computer readable medium of claim 11 , wherein said normalizing comprises normalizing one or more of:

energy of the discrete curves; and

length of the discrete curves.

15. The computer readable medium of claim 11 , wherein said computing the weight function comprises:

computing a weight vector w, such that

< c n ,c m > w =δ nm

wherein C n and C m comprise any two discrete curves of the normalized discrete curves; and

wherein δ nm is the Kronecker delta.

16. The computer readable medium of claim 11 , wherein components of the weight function are non-negative.

17. The computer readable medium of claim 11 , wherein the weight function comprises a Sturm-Liouville weight function.

18. The computer readable medium of claim 11 , wherein the re-sampling points for the plurality of discrete curves comprise a set of optimal re-sampling points for the plurality of discrete curves.

19. The computer readable medium of claim 11 , wherein the re-sampling points for the plurality of discrete curves comprise a re-sampling scheme that enhances differences between the plurality of discrete curves.

20. The computer readable medium of claim 11 , wherein the plurality of discrete curves comprise one or more of:

image objects from one or more images; and

data objects from one or more data sets.

21. A system for re-sampling discrete curves, comprising:

a computer system, comprising:

a processor; and

a memory medium coupled to the processor; and

wherein the memory medium stores program instructions which are executable by the processor to:

normalize a plurality of discrete curves, wherein each discrete curve comprises a respective sequence of points, and wherein the discrete curves correspond to data;

compute a weight function based on the plurality of discrete curves, wherein the weight function enhances differences between weighted discrete curves;

determine a set of orthonormal polynomials based on the computed weight function, wherein the set of orthonormal polynomials comprises a set of basis functions;

determine values for a plurality of zeros for one of the set of orthonormal polynomials, wherein the values for the plurality of zeros comprise re-sampling points for the plurality of discrete curves; and

re-sample each of the plurality of discrete curves based on the determined values of the plurality of zeros, thereby generating re-sampled discrete curves;

wherein the re-sampled discrete curves are usable to analyze the data.

22. A system for re-sampling discrete curves, comprising:

means for normalizing a plurality of discrete curves, wherein each discrete curve comprises a respective sequence of points;

means for computing a weight function based on the plurality of discrete curves, wherein the weight function enhances differences between weighted discrete curves;

means for determining a set of orthonormal polynomials based on the computed weight function, wherein the set of orthonormal polynomials comprises a set of basis functions;

means for determining values for a plurality of zeros for one of the set of orthonormal polynomials, wherein the values for the plurality of zeros comprise re-sampling points for the plurality of discrete curves; and

means for re-sampling each of the plurality of discrete curves based on the determined values of the plurality of zeros.

23. A computer-implemented method for re-sampling discrete curves, the method comprising:

receiving one or more images, wherein each image contains one or more image objects;

determining a plurality of discrete curves from the one or more images, wherein each of the discrete curves corresponds to one of the one or more image objects, and wherein each discrete curve comprises a respective sequence of points;

normalizing each of the plurality of discrete curves,

computing a weight function based on the plurality of discrete curves, wherein the weight function enhances differences between weighted discrete curves;

determining a set of orthonormal polynomials based on the computed weight function, wherein the set of orthonormal polynomials comprises a set of basis functions;

determining values for a plurality of zeros for one of the set of orthonormal polynomials, wherein the values for the plurality of zeros comprise re-sampling points for the plurality of discrete curves; and

re-sampling each of the plurality of discrete curves based on the determined values of the plurality of zeros, thereby generating re-sampled discrete curves;

wherein the re-sampled discrete curves are usable to analyze the one or more images.

24. A computer-implemented method for re-sampling discrete curves, the method comprising:

receiving one or more data sets, wherein each data set contains one or more data objects;

determining a plurality of discrete curves from the one or more data sets, wherein each of the discrete curves corresponds to one of the one or more data objects, and wherein each discrete curve comprises a respective sequence of points;

normalizing each of the plurality of discrete curves,

computing a weight function based on the plurality of discrete curves, wherein the weight function enhances differences between weighted discrete curves;

determining a set of orthonormal polynomials based on the computed weight function, wherein the set of orthonormal polynomials comprises a set of basis functions;

determining values for a plurality of zeros for one of the set of orthonormal polynomials, wherein the values for the plurality of zeros comprise re-sampling points for the plurality of discrete curves; and

re-sampling each of the plurality of discrete curves based on the determined values of the plurality of zeros; thereby generating re-sampled discrete curves;

wherein the re-sampled discrete curves are usable to analyze the one or more data sets.

25. A computer-implemented method for re-sampling discrete curves, wherein the discrete curves correspond to data, the method comprising:

computing a weight function based on a plurality of discrete curves, wherein the weight function enhances differences between weighted discrete curves;

determining a set of orthonormal polynomials based on the computed weight function, wherein the set of orthonormal polynomials comprises a set of orthogonal eigenfunctions of a Sturm-Liouville differential equation;

determining values for a plurality of zeros for one of the set of orthonormal polynomials, wherein the values for the plurality of zeros comprise re-sampling points for the plurality of discrete curves; and

re-sampling each of the plurality of discrete curves based on the determined values of the plurality of zeros, thereby generating re-sampled discrete curves;

wherein the re-sampled discrete curves are usable to analyze the data.