Multi-resolution analysis in image segregation
View Patent ↗In a first exemplary embodiment of the present invention, an automated, computerized method is provided for processing an image. According to a feature of the present invention, the method comprises the steps of providing an image file depicting an image, in a computer memory, forming a set of selectively varied representations of the image file and performing an image segregation operation on at least one preselected representation of the image of the image file, to generate intrinsic images corresponding to the image. According to a feature of the exemplary embodiment of the present invention, the selectively varied representations comprise multi-resolution representations such as a scale-spaced pyramid of representations. In a further feature of the exemplary embodiment of the present invention, the intrinsic images comprise a material image and an illumination image.
1. An automated, computerized method for processing an image, comprising the steps of:
providing an image file depicting an image, in a computer memory;
forming a set of selectively varied representations of the image file each expressed as a multi-band representation with each band corresponding to a segment of the electro-magnetic spectrum; and
performing an image segregation operation on at least one preselected representation of the image of the image file to generate intrinsic images corresponding to the image, such that at least one of the intrinsic images comprises a material representation and at least another one of the intrinsic images comprises an illumination representation independent of said at least one of the intrinsic images comprising a material representation, wherein the image segregation operation is performed on the bands of the at least one preselected image directly by utilizing a spatio-spectral operator/constraint/solver model.
2. The method of claim 1 wherein the step of forming a set of selectively varied representations of the image file is carried out by forming multi-resolution representations of the image.
3. The method of claim 2 wherein the multi-resolution representations of the image comprise a scale-spaced pyramid of representations.
4. The method of claim 1 wherein the step of utilizing a spatio-spectral operator/constraint/solver model is carried out by generating spatio-spectral operators at each of different ones of the selectively varied representations of the image, and implementing a constraint as a function of the spatio-spectral operators generated at each of different ones of the selectively varied representations of the image.
5. The method of claim 4 wherein the spatio-spectral operators include a weighted color average.
6. The method of claim 1 wherein the step of performing an image segregation operation on at least one preselected representation of the image of the image file, to generate intrinsic images corresponding to the image, is carried out by performing an image segregation operation in respect of preselected regions of at least one preselected representation of the image of the image file.
7. The method of claim 6 wherein the step of performing an image segregation operation in respect of preselected regions of at least one preselected representation of the image of the image file is carried out by performing an image segregation operation in respect of different preselected regions of at least two of the set of selectively varied representations of the image.
8. The method of claim 1 wherein the step of performing an image segregation operation on at least one preselected representation of the image of the image file, to generate intrinsic images corresponding to the image, is carried out as a function of an image segregation of another one of the set of selectively varied representations of the image.
9. The method of claim 8 wherein the function of an image segregation of another one of the set of selectively varied representations of the image includes a weighted color average.
10. The method of claim 8 wherein the step of performing an image segregation operation on at least one preselected representation of the image of the image file is carried out over a limited extent of the at least one preselected representation of the image.
11. The method of claim 10 wherein the function of an image segregation of another one of the set of selectively varied representations of the image includes a projection of the image segregation operation carried out over a limited extent of the at least one preselected representation of the image to another one of the set of selectively varied representations of the image.
12. A computer system which comprises:
a CPU; and
a memory storing an image file containing an image;
the CPU is arranged and configured to execute a routine to form a set of selectively varied representations of the image each expressed as a multi-band representation with each band corresponding to a segment of the electro-magnetic spectrum and to perform an image segregation operation on at least one preselected representation of the image of the image file, to generate intrinsic images corresponding to the image, such that at least one of the intrinsic images comprises a material representation and at least another one of the intrinsic images comprises an illumination representation independent of said at least one of the intrinsic images comprising a material representation, wherein the image segregation operation is performed on the bands of the at least one preselected image directly by utilizing a spatio-spectral operator/constraint/solver model.
13. A computer program product, disposed on a non-transitory computer readable media, the product including computer executable instructions operable to control a computer to carry out the method of claim 1 .