Measurement of thin-layered structures in X-ray computer tomography

1. A computer-implemented method for reconstructing an image, the method comprising:

causing an electronic digital computer to receive tomographic data representative of an image signal;

causing the electronic digital computer to derive, from the image signal, a plurality of components;

causing the electronic digital computer to identify a spatial location associated with maximum phase congruency of the plurality of components;

causing the electronic digital computer to incorporate, into an image, an edge at the spatial location; and

causing the electronic digital computer to provide an output representative of the image.

2. The method of claim 1 , wherein causing the electronic digital computer to derive a plurality of components from the image signal comprises:

causing the electronic digital computer to pass the image signal through each of a plurality of filters;

whereby the output of each such filter results in one of the plurality of components.

3. The method of claim 2 , wherein causing the electronic digital computer to identify a spatial location associated with a maximum phase congruency comprises:

for each of a plurality of locations, causing the electronic digital computer to determine a consistency of local phase among the plurality of components; and

causing the electronic digital computer to identify a spatial location showing the maximum consistency to be the location associated with maximum phase congruency.

4. The method of claim 2 , further comprising selecting the filters to be quadrature pair filters.

5. The method of claim 1 , wherein causing the electronic digital computer to derive a plurality of components comprises causing the electronic digital computer to define a plurality of kernels, each of which corresponds to one of the plurality of components.

6. The method of claim 5 , wherein causing the electronic digital computer to identify a spatial location comprises causing the electronic digital computer to operate on the image signal with each of the kernels, thereby generating a corresponding plurality of responses, and estimating a common crossing point of the responses.

7. The method of claim 5 , further comprising selecting the kernels to be zero-phase kernels.

8. A manufacture comprising a non-transitory computer-readable medium having encoded thereon software for reconstructing an image, the software including instructions for:

receiving tomographic data representative of an image signal;

deriving, from the image signal, a plurality of components;

identifying a spatial location associated with maximum phase congruency of the plurality of components;

incorporating, into an image, an edge at the spatial location; and

providing an output representative of the image.

9. The manufacture of claim 8 , wherein the instructions for deriving a plurality of components from the image signal include instructions for:

passing the image signal through each of a plurality of filters;

whereby an output of each such filter results in one of the plurality of components.

10. The manufacture of claim 9 , wherein the instructions for identifying a spatial location associated with a maximum phase congruency include instructions for:

for each of a plurality of locations, determining a consistency of local phase among the plurality of components; and

identifying a spatial location showing the maximum consistency to be the location associated with maximum phase congruency.

11. The manufacture of claim 9 , wherein the software further includes instructions for selecting the filters to be log-Gabor filters.

12. The manufacture of claim 8 , wherein the instructions for deriving a plurality of components include instructions for defining a plurality of kernels, each of which corresponds to one of the plurality of components.

13. The manufacture of claim 12 , wherein the instructions for identifying a spatial location include instructions for operating on the image signal with each of the kernels, thereby generating a corresponding plurality of responses, and estimating a common crossing point of the responses.

14. The manufacture of claim 12 , wherein the instructions further include selecting the kernels to be zero-phase kernels.

15. An electronic data processing system for reconstructing an image on the basis of tomographic data representative of an image signal, the system comprising:

an electronic processor for executing instructions, and a memory in data communication with the electronic processor, the memory having recorded thereon instructions that, when executed by the electronic processor, cause the establishment of means for deriving, from the image signal, a plurality of components;

a phase-congruency detector for providing, on the basis of the plurality of components, a spatial location associated with maximum phase congruency of the plurality of components, and

an image-enhancement unit in data communication with the phase-congruency detector, the image-enhancement unit being configured to incorporate, into an image, an edge at the spatial location; and

an output device in data communication with the image-enhancement unit for providing output representative of the image.

16. The system of claim 15 , wherein the means for deriving an image signal comprises a plurality of filters, the output of each filter providing one of the plurality of components.

17. The system of claim 16 , wherein the plurality of filters comprises quadrature pair filters.

18. The system of claim 16 , wherein the plurality of filters comprises log-Gabor filters.

19. The system of claim 15 , wherein the means for deriving an image comprises a plurality of kernels, each of which corresponds to one of the components.

20. The system of claim 15 , further comprising a CT scanner for providing the tomographic data.