Skin assessment using image fusion
Apparatuses and methods are disclosed for assessing the texture of skin using images thereof. In exemplary embodiments, a texture map of an area of skin is generated from a combination of a standard white light image, a parallel-polarized image, and a cross-polarized image of the area of skin. The texture map is then flattened to remove the underlying curvature of the skin. A texture roughness metric is then generated based on the flattened texture map. An image of the texture map and the metric can be displayed to provide visual and alphanumeric representations of the texture of skin, thereby facilitating the comparison of baseline and follow-up images of the skin, such as those taken before and after treatment.
1. An apparatus comprising:
a storage device configured to store instructions;
a processor configured to execute instructions stored in the storage device to:
obtain a standard image of a skin area;
obtain a parallel-polarized image of the skin area;
obtain a cross-polarized image of the skin area;
generate a texture map from a combination of the standard, parallel-polarized and cross-polarized images;
generate at least one of a texture map image or a texture metric from the texture map; and
cause the at least one of the texture map, the texture map image, or the texture metric to be displayed, stored, communicated, or processed;
an illumination source, wherein the illumination source is configured to illuminate the skin area with unpolarized light or polarized light of a first polarization orientation; and
an image capture device configured to capture an image without polarization or with polarization of a second polarization orientation,
wherein the first and second polarization orientations are the same for capturing the parallel-polarized image of the skin area and mutually orthogonal for capturing the cross-polarized image of the skin area.
2. The apparatus of claim 1 , wherein the combination includes:
an intensity of the parallel-polarized image divided by an intensity of the cross-polarized image,and
a fraction of an intensity of the standard image.
3. The apparatus of claim 1 , wherein the processor is configured to execute instructions to perform an image registration operation on the standard, parallel-polarized and cross-polarized images before generating the texture map.
4. The apparatus of claim 1 , wherein the processor is configured to execute instructions to generate a flattened texture map.
5. The apparatus of claim 1 , wherein generating the texture metric includes:
applying a two-dimensional wavelet transform to the texture map to determine low- and high-frequency components, and
determining a ratio of an energy of the high-frequency components to an energy of the low-frequency components.
6. The apparatus of claim 1 , wherein generating the texture map image includes determining a gray scale value for each of a plurality of pixels of the texture map.
7. The apparatus of claim 1
wherein at least one of the first or second polarization orientations is selectable.
8. The apparatus of claim 1 , wherein the processor is configured to execute instructions to detect a skin feature by performing a segmentation operation using at least one of the texture map or the texture map image.
9. The apparatus of claim 1 , wherein the processor is configured to execute instructions to generate a further image by modifying one or more of the standard, parallel-polarized and cross-polarized images in accordance with the texture map.
10. The apparatus of claim 1 , wherein the processor is configured to execute instructions to generate at least one of a treatment evaluation or a treatment suggestion.
11. A method comprising:
obtaining a standard image of a skin area;
obtaining a parallel-polarized image of the skin area;
obtaining a cross-polarized image of the skin area;
generating a texture map from a combination of the standard, parallel-polarized and cross-polarized images;
generating at least one of a texture map image or a texture metric from the texture map; and
causing the at least one of the texture map, the texture map image, or the texture metric to be displayed, stored, communicated, or processed,
wherein obtaining the standard, parallel-polarized and cross-polarized images of the skin area includes:
illuminating the skin area with unpolarized light or polarized light of a first polarization orientation; and
capturing an image without polarization or with polarization of a second polarization orientation,
wherein the first and second polarization orientations are the same for capturing the parallel-polarized image of the skin area and mutually orthogonal for capturing the cross-polarized image of the skin area.
12. The method of claim 11 , wherein the combination includes:
an intensity of the parallel-polarized image divided by an intensity of the cross-polarized image, and
a fraction of an intensity of the standard image.
13. The method of claim 11 comprising performing an image registration operation on the standard, parallel-polarized and cross-polarized images before generating the texture map.
14. The method of claim 11 , comprising generating a flattened texture map.
15. The method of claim 11 , wherein generating the texture metric includes:
applying a two-dimensional wavelet transform to the texture map to determine low- and high-frequency components, and
determining a ratio of an energy of the high-frequency components to an energy of the low-frequency components.
16. The method of claim 11 , wherein generating the texture map image includes determining a gray scale value for each of a plurality of pixels of the texture map.
17. The method of claim 11
wherein at least one of the first or second polarization orientations is selectable.
18. The method of claim 11 comprising detecting a skin feature by performing a segmentation operation using at least one of the texture map or the texture map image.
19. The method of claim 11 comprising generating a further image by modifying one or more of the standard, parallel-polarized and cross-polarized images in accordance with the texture map.
20. A non-transitory computer readable storage medium containing instructions for execution by a processor for carrying out the method of claim 11 .