Image processing apparatus
An image processing apparatus comprises a normalisation module operatively connected across a bus to a memory storing an image in which a region of interest (ROI) has been identified within the image. The ROI is bound by a rectangle having a non-orthogonal orientation within the image. In one embodiment, the normalisation module is arranged to divide the ROI into one or more slices, each slice comprising a plurality of adjacent rectangular tiles. For each slice, the apparatus successively reads ROI information for each tile from the memory including: reading a portion of the image extending across at least a width of the slice line-by-line along an extent of a slice. For each tile, the apparatus downsamples the ROI information to a buffer to within a scale S D <2 of a required scale for a normalised version of the ROI. The apparatus then fractionally downsamples and rotates downsampled information for a tile within the buffer to produce a respective normalised portion of the ROI at the required scale for the normalised ROI. Downsampled and rotated information is accumulated for each tile within a normalised ROI buffer for subsequent processing by the image processing apparatus.
1. An electronic device comprising:
one or more processors; and
one or more computer-readable media storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
storing first image data representing a first image;
identifying a region of interest (ROI) of the first image;
identifying a plurality of portions of the ROI, the plurality of portions of the ROI including at least a first portion of the ROI and a second portion of the ROI;
generating a second image by at least rotating the first portion of the ROI from a first orientation to a second orientation;
storing second image data representing the second image;
generating a third image by at least rotating the second portion of the ROI from a third orientation to a fourth orientation; and
storing third image data representing the third image.
2. The electronic device as recited in claim 1 , the operations further comprising:
generating a fourth image by downsampling the first portion of the ROI, the fourth image including the first orientation,
wherein generating the second image comprises generating the second image by rotating the fourth image from the first orientation to the second orientation.
3. The electronic device as recited in claim 1 , wherein at least one of:
the first orientation of the first portion of the ROI includes a first same orientation as the third orientation of the second portion of the ROI; or
the second orientation of the second image includes a second same orientation as the fourth orientation of the third image.
4. The electronic device as recited in claim 1 , the operations further comprising:
identifying, based at least in part on a size of the ROI, a number of portions into which to divide the ROI,
wherein identifying the plurality of portions of the ROI is based at least in part on the number of portions.
5. The electronic device as recited in claim 1 , wherein identifying the plurality of portions of the ROI comprises at least:
identifying the first portion of the ROI; and
identifying the second portion of the ROI as including an adjacent portion to the first portion of the ROI.
6. The electronic device as recited in claim 1 , wherein the ROI is a first ROI and the plurality of portions is a first plurality of portions, and wherein the operations further comprise:
identifying a second ROI of the first image;
identifying a second plurality of portions of the second ROI, the second plurality of portions of the second ROI including at least a third portion of the second ROI and a fourth portion of the second ROI;
generating a fourth image by at least rotating the third portion of the second ROI from a fifth orientation to a sixth orientation; and
storing fourth image data representing the fourth image.
7. The electronic device as recited in claim 6 , wherein the sixth orientation of the third image includes a different orientation than the second orientation of the second image.
8. The electronic device as recited in claim 1 , wherein:
the ROI is associated with a rectangle within the first image; and
the operations further comprise determining to process the plurality of portions of the ROI starting with the first portion of the ROI and ending with the second portion of the ROI, the first portion of the ROI being located at a first edge of the ROI and the second portion of the ROI being located at a second edge of the ROI, the second edge being opposite to the first edge.
9. The electronic device as recited in claim 1 , wherein identifying the ROI of the first image comprises at least:
identifying a face depicted by the first image; and
identifying the ROI as a third portion of the first image that depicts at least the face.
10. A method comprising:
storing first image data representing a first image;
identifying a region of interest (ROI) of the first image;
identifying a plurality of portions of the ROI, the plurality of portions of the ROI including at least a first portion of the ROI and a second portion of the ROI;
generating a second image by downsampling the first portion of the ROI;
generating a third image by at least rotating the second image from a first orientation to a second orientation; and
storing second image data representing the third image.
11. The method as recited in claim 10 , further comprising:
generating a fourth image by at least rotating the second portion of the ROI from a third orientation to a fourth orientation; and
storing third image data representing the fourth image.
12. The method as recited in claim 11 , wherein the second orientation of the third image includes a same orientation as the fourth orientation of the fourth image.
13. The method as recited in claim 10 , further comprising:
identifying, based at least in part on a size of the ROI, a number of portions into which to divide the ROI,
wherein identifying the plurality of portions of the ROI is based at least in part on the number of portions.
14. The method as recited in claim 10 , wherein identifying the plurality of portions of the ROI comprises at least:
identifying the first portion of the ROI; and
identifying the second portion of the ROI as including an adjacent portion to the first portion of the ROI.
15. The method as recited in claim 10 , wherein the ROI is a first ROI and the plurality of portions is a first plurality of portions, and wherein the method further comprises:
identifying a second ROI of the first image;
identifying a second plurality of portions of the second ROI, the second plurality of portions of the second ROI including at least a third portion of the second ROI and a fourth portion of the second ROI;
generating a fourth image by at least rotating the third portion of the second ROI from a third orientation to a fourth orientation; and
storing third image data representing the fourth image.
16. The method as recited in claim 15 , wherein the fourth orientation of the fourth image includes a different orientation than the second orientation of the third image.
17. The method as recited in claim 10 , wherein:
the ROI is associated with a rectangle within the image; and
the method further comprising determining to process the plurality of portions of the ROI starting with the first portion of the ROI and ending with the second portion of the ROI, the first portion of the ROI being located at a first edge of the ROI and the second portion of the ROI being located at a second edge of the ROI, the second edge being opposite to the first edge.
18. The method as recited in claim 10 , wherein identifying the ROI of the first image comprises at least:
identifying a face depicted by the first image; and
identifying the ROI as a third portion of the first image that depicts at least the face.
19. The electronic device as recited in claim 1 , the operations further comprising:
generating a fourth image that includes at least the second image and the third image; and
storing fourth image data representing the fourth image.
20. The electronic device as recited in claim 1 , the operations further comprising:
reading a first portion of the first image data out of memory for processing, the first portion of the first image data representing the first portion of the ROI; and
after generating the second image, reading a second portion of the first image data out of the memory for processing, the second portion of the first image data representing the second portion of the ROI.