Image processing for aberration correction
Processing of a digital image that has been recorded through an optical system is described, the image being in a raw format matrix of picture elements and comprising a plurality of subsets of picture elements, each subset corresponding to a color component. The method comprises analyzing values of picture elements of the color components in terms of spatial features, resulting in at least one calculated parameter that is representative of aberration caused by the optical system, and performing demosaicing of the digital image by, for each picture element under consideration, defining a kernel comprising picture elements that are to be considered during the demosaicing, the definition of the kernel being dependent on the at least one parameter that is representative of the aberration.
1. A method of processing a digital image recorded through an optical system, the image being in a raw format matrix of picture elements and comprising a plurality of subsets of picture elements, each subset corresponding to a color component, the method comprising:
analyzing values of picture elements of the color components of the digital image in terms of spatial features, resulting in at least one calculated parameter that is representative of aberration caused by the optical system; and
performing demosaicing of the digital image by, for each picture element under consideration, defining a kernel comprising picture elements that are to be considered during the demosaicing, the definition of the kernel being dependent on the at least one parameter that is representative of the aberration.
2. The method of claim 1 , comprising:
generating a plurality of color component images by interpolating the respective subsets of picture elements corresponding to a color component, and wherein:
the analysis of values of picture elements comprises analyzing values of picture elements of the color component images.
3. The method of claim 1 , wherein the analysis of values of picture elements in terms of spatial features comprises any of edge detection and image segmentation.
4. The method of claim 1 , wherein the analysis of values of picture elements in terms of spatial features comprises comparing values of picture elements between color component images by utilizing an image registration technique.
5. The method of claim 1 , wherein the definition of the kernel comprises calculation of at least a size of the kernel, a distance and direction of displacement of the kernel in relation to the picture element under consideration.
6. The method of claim 1 , comprising calculating a demosaicing map comprising kernel data in relation to all picture elements of the raw image, and utilizing the calculated demosaicing map during the demosaicing.
7. A communication terminal comprising an optical system, an image sensor and image processing circuitry configured to perform a method of processing a digital image recorded through the optical system, the image being in a raw format matrix of picture elements and comprising a plurality of subsets of picture elements, each subset corresponding to a color component, the method comprising:
analyzing values of picture elements of the color components of the digital image in terms of spatial features, resulting in at least one calculated parameter that is representative of aberration caused by the optical system; and
performing demosaicing of the digital image by, for each picture element under consideration, defining a kernel comprising picture elements that are to be considered during the demosaicing, the definition of the kernel being dependent on the at least one parameter that is representative of the aberration.
8. The communication terminal of claim 7 , comprising a Bayer arrangement of color filters arranged at the image sensor.
9. A non-transitory computer readable medium having stored therein software instructions that, when executed in a processor, causes the processor to perform a method of processing a digital image recorded through an optical system, the image being in a raw format matrix of picture elements and comprising a plurality of subsets of picture elements, each subset corresponding to a color component, the method comprising:
analyzing values of picture elements of the color components of the digital image in terms of spatial features, resulting in at least one calculated parameter that is representative of aberration caused by the optical system; and
performing demosaicing of the digital image by, for each picture element under consideration, defining a kernel comprising picture elements that are to be considered during the demosaicing, the definition of the kernel being dependent on the at least one parameter that is representative of the aberration.