Image reconstruction method and system

1. A method of producing a scalar image from a derivative field and a vector image, the derivative field being calculated or derived from the vector image, the method including the steps of:

a) retrieving, by a processing unit of at least one computing device, a function class c from a data store, wherein all members of the class c are a finite number of prepopulated functions which map each vector of the vector image to a unique scalar value;

b) selecting, by the processing unit of the at least one computing device, a function f from the class c which maps the vector image to a scalar image, a derivative field of the function f being closest to the derivative field from amongst derivative fields of other members of the function class c; and

c) generating, by the processing unit of the at least one computing device, the scalar image from the vector image by using the function f to calculate each scalar value in the scalar image from a corresponding vector in the vector image, wherein the calculation of each scalar value is performed without solving Poisson's equation.

2. The method of claim 1 , wherein all members of the class c are surjective functions.

3. The method of claim 1 , wherein all members of the class c are smoothly varying functions.

4. The method of claim 1 , wherein all members of the class c are polynomial functions.

5. The method of claim 4 , wherein all members of the class c are second order polynomial functions.

6. The method of claim 1 , wherein all members of the class c are look up table operations.

7. The method of claim 6 , wherein all members of the class c are look up table operations with local interpolation.

8. The method of claim 1 , wherein the derivative field is derived from the vector image.

9. The method of claim 1 , wherein the derivative field is based on 1st order derivative calculations.

10. The method of claim 1 , wherein the derivative field is based on 2nd order derivative calculations.

11. The method of claim 1 , wherein the derivative field is defined as a vector of components wherein each component is a linear combination of at least one order of derivative-type calculations.

12. The method of claim 1 , wherein the vector image is a multi-spectral image.

13. The method of claim 1 , wherein the vector image is a hyper-spectral image.

14. The method of claim 1 , wherein the vector image is a chromaticity image.

15. A method of producing a second vector image from a derivative field and a first vector image, including the steps of:

defining, by a processing unit of the at least one computing device, the second vector image to be composed of a separate scalar image in each spectral dimension;

producing, by the processing unit of the at least one computing device, the respective scalar image for each spectral dimension from the derivative field and the first vector image, wherein for each spectral dimension, the step of producing including the steps of:

a) retrieving, by the processing unit of the at least one computing device, a function class c from a data store, wherein all members of the class c are prepopulated functions which map each vector of the vector image to a unique scalar value;

b) selecting, by the processing unit of the at least one computing device, a function f from the class c which maps the first vector image to a scalar image, a derivative of the function f being closest to the derivative field from amongst derivative fields of other members of the function class c; and

c) generating, by the processing unit of the at least one computing device, the scalar image for the spectral dimension from the first vector image by using the function f to calculate each scalar value in the scalar image from a corresponding vector in the first vector image, wherein the calculation of each scalar value is performed without solving Poisson's equation.

16. A method of producing a master scalar image from a master derivative field and a master vector image including the steps of:

i) dividing, by a processing unit of at least one computing device, each of the master scalar image, the master derivative field and the master vector image into a plurality of corresponding regions, each region comprising a master scalar image region, a master derivative field region and a master vector image region; and

ii) for each region:

a) retrieving, by the processing unit of the at least one computing device, a function class c from a data store, wherein all members of the class c are prepopulated functions which map each vector of the vector image to a unique scalar value;

b) selecting, by the processing unit of the at least one computing device, a function f from the class c which maps the master vector image to a scalar image, a derivative of the function f being closest to the derivative field from amongst derivative fields of other members of the function class c; and

c) generating, by the processing unit of the at least one computing device, the scalar image for the spectral dimension from the master vector image by using the function f to calculate each scalar value in the scalar image from a corresponding vector in the master vector image, wherein the calculation of each scalar value is performed without solving Poisson's equation.

17. The method of claim 16 , wherein in step (c) of the method applied to each region, the step of using f to calculate each scalar value in the scalar image comprises applying, by the at least one computing device, a function g to calculate each scalar value in the scalar image from a corresponding vector in the master vector image region, wherein selecting the function f further comprises calculating g as a weighted combination of functions f for other proximal regions.

18. The method of claim 16 , further comprising producing a second master vector image from the master derivative field and master vector image, including the steps of:

defining, by the at least one computing device, the second master vector image to be composed of a separate master scalar image in each spectral dimension; and

performing the dividing, the selecting the function class c, the selecting the function f, and the generating the scalar region for each spectral dimension.

19. A computer implemented image reconstruction system comprising:

a first data receiving module configured to receive data on a vector image;

a second data receiving module configured to obtain data on a derivative field; and

a processing unit configured to execute computer program code to:

retrieve a function class c from a data store, wherein all members of the function class c are functions which map each vector of the vector image to a unique scalar value, wherein the function class c comprises a finite number of prepopulated functions;

select a function f from the class c which maps the vector image to a scalar image, a derivative of the function f being closest to the derivative field from amongst derivative fields of other members of the function class c; and

generate the scalar image from the vector image by using the function f to calculate each scalar value in the scalar image from a corresponding vector in the vector image, wherein the calculation of each scalar value is performed without solving Poisson's equation; and, output the scalar image.