Determining color values in hyperspectral or multispectral images

Patent metadata

Field	Value
Publication number	US-9754181-B2
Application number	US-201214355796-A
Country	US
Kind code	B2
Filing date	Nov 2, 2012
Priority date	Nov 4, 2011
Publication date	Sep 5, 2017
Grant date	Sep 5, 2017

Abstract

The disclosure concerns processing of electronic images, such as hyperspectral or multispectral images ( 302 ) to determine an output color value for a component spectrum of a decomposed multispectral or hyperspectral image. A processor ( 120 ) first determines or accesses the component spectrum and a first color value for the component spectrum and receives an input color value for a pixel location in the image, wherein the component spectrum contributes to the pixel. The processor ( 120 ) then determines the output color value for the component spectrum based on the first color value for the component spectrum, the input color value and a utility score and stores the output color value for the component spectrum on a datastore ( 118 ). It is an advantage that the processor ( 120 ) receives a color value for a pixel location but the output color value is not exactly the received color value but one that is determined based on a utility score.

First claim

The invention claimed is:

1. A computer implemented method for converting a hyperspectral image into a color image, the method comprising:
receiving wavelength indexed hyperspectral image data captured by a hyperspectral sensor and decomposed into multiple component spectra, such that a mixture of the multiple component spectra forms the hyperspectral image data;
determining or accessing a first color value for each of the multiple component spectra;
receiving from a user, user input color value for a pixel location in the image, wherein one or more of the multiple component spectra contributes to the pixel according to the mixture;
determining the output color value for each of the one or more multiple component spectra by minimizing a difference between the user input color value and a pixel color value, the pixel color value based on the first color value for the one or more multiple component spectra according to the mixture and minimising the difference is based on a utility score;
creating a color image by determining for each pixel of the color image the pixel color value by mixing for each pixel the output color value for each of the one or more multiple component spectra according to the mixture; and
storing the color image on a datastore.

2. The computer implemented method of claim 1 , wherein determining the output color value comprises optimising a cost function.

3. The computer implemented method of claim 2 , wherein the method comprises determining or accessing a component contribution for the pixel location and the cost function is based on the component contribution.

4. The computer implemented method of claim 2 , wherein the cost function is based on a utility score for each component spectrum.

5. The computer implemented method of claim 1 , wherein the image is decomposed into material spectra and the component spectrum is one of the material spectra.

6. The computer implemented method of claim 5 , wherein the method comprises:
determining or accessing a decomposition of the material spectra into endmember spectra; and
determining an output color value for each endmember spectrum based on the output color value for the material spectrum.

7. The computer implemented method of claim 1 , wherein the image is decomposed into endmember spectra and the component spectrum is one of the endmember spectra.

8. The computer implemented method of claim 1 , wherein the method is performed, such that an output color value for each component spectrum represented in the image is determined.

9. The computer implemented method of claim 1 , wherein the utility score is any one or more of:
an aesthetic score;
a score based on a predetermined scale of danger, benefit, profit or added financial or monetary value;
enhanced visualisation of multiple images; or
the nature of the rendering context.

10. The computer implemented method of claim 1 , wherein the utility score is based on a classifier.

11. The computer implemented method of claim 10 , wherein the classifier is any one or more of:
a quadratic discriminant;
a support vector machine;
a linear discriminant; or
a logistic regressor.

12. The computer implemented method of claim 10 , wherein the classifier is based on one or more of:
abundance of material spectra in the image;
most abundant component spectrum;
second most abundant component spectrum;
rule of thirds;
whole image average;
macro feature;
domain knowledge; or
rules and regulations pertaining the utility measure.

13. The computer implemented method of claim 1 , wherein the utility score is based on the contrast of the image.

14. The computer implemented method of claim 1 , wherein the utility score is based on the hazardousness of the component spectrum.

15. The computer implemented method of claim 1 , wherein the method comprises accessing a library of component spectra.

16. The computer implemented method of claim 1 , wherein the method comprises determining or accessing illuminant spectra and illuminant contributions.

17. The computer implemented method of claim 16 , wherein the method comprises determining a color value for each illuminant spectrum.

18. The computer implemented method of claim 16 , wherein the method comprises determining a color value for each combination of component spectra and illuminant spectra.

19. The computer implemented method of claim 1 , wherein the method comprises displaying to a user the color image.

20. The computer implemented method of claim 1 , wherein the method comprises initializing the first color value for the component spectrum with a default color.

21. A non-transitory computer readable medium with an executable program stored thereon that when executed causes a computer to perform the method of claim 1 .

22. A computer system for converting a hyperspectral image into a color image, the system comprising:
a port for receiving wavelength indexed hyperspectral image data captured by a hyperspectral sensor and decomposed into multiple component spectra, such that a mixture of the multiple component spectra forms the hyperspectral image data;
a processor to determine or access a first color value for each of the multiple component spectra, receive from a user, a user input colour value for a pixel location in the image wherein one or more of the multiple component spectra contributes to the pixel according to the mixture, determine the output color value for each of the one or more multiple component spectra by minimising a difference between the user input color value and a pixel color value, the pixel color value based on the first color value for the one or more multiple component spectra according to the mixture and minimising the difference based on a utility score, create a color image by determining for each pixel of the color image the pixel color value by mixing for each pixel the output color value for each of the one or more multiple component spectra according to the mixture, and store a color image including the pixel color value on a datastore.

23. The computer system of claim 22 , wherein the computer system comprises a storage to store the image data, the component spectra and the color image.

Assignees

• Nat Ict Australia Limitied
• Nat Ict Australia Ltd

Inventors

• Robles-Kelly Antonio
• Huynh Cong Phuoc
• Gu Lin

Classifications

• H04N1/6058 Primary

  Reduction of colour to a range of reproducible colours, e.g. to ink- reproducible colour gamut · CPC title

• G06T11/10

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06T11/001

  Physics · mapped topic

• H04N1/6086

  by scene illuminant, i.e. conditions at the time of picture capture, e.g. flash, optical filter used, evening, cloud, daylight, artificial lighting, white point measurement, colour temperature · CPC title

• G06K9/4652 Primary

  Physics · mapped topic