Object segmentation at a self-checkout
Techniques for segmenting an object at a self-checkout are provided. The techniques include capturing an image of an object at a self-checkout, dividing the image into one or more blocks, computing a confidence value for each of the one or more blocks, and eliminating one or more blocks from consideration based on the confidence value for each of the one or more blocks, wherein the one or more blocks remaining map to a region of the image containing the object.
1. A method for segmenting an object at a self-checkout, wherein the method comprises:
capturing an image of an object at a self-checkout, said capturing carried out by a distinct module executing on a hardware processor;
dividing the image into one or more blocks, said dividing carried out by a distinct module executing on a hardware processor;
computing a confidence value for each of the one or more blocks, said computing carried out by a distinct module executing on a hardware processor; and
eliminating one or more blocks from consideration based on the confidence value for each of the one or more blocks, wherein the one or more blocks remaining map to a region of the image containing the object, said eliminating carried out by a distinct module executing on a hardware processor.
2. The method of claim 1 , further comprising:
reading a barcode for the object at the self-checkout; and
using a database to identify one or more known features corresponding to the barcode.
3. The method of claim 2 , further comprising comparing the one or more computed features of the image with the one or more known features corresponding to the barcode to determine whether or not the image matches the object's barcode identity.
4. The method of claim 1 , further comprising creating and maintaining a set of one or more reference background images.
5. The method of claim 4 , wherein the set of one or more background reference images cover one or more visual variations of the self-checkout.
6. The method of claim 4 , wherein creating and maintaining a set of one or more reference background images comprises creating and maintaining a set of one or more reference background images over two or more different time periods.
7. The method of claim 4 , further comprising dividing each reference background image into a set of one or more non-overlapping blocks.
8. The method of claim 7 , further comprising extracting one or more features from each of the one or more non-overlapping blocks.
9. The method of claim 1 , further comprising using the confidence value of each block to weight a color histogram used for object classification.
10. The method of claim 1 , wherein capturing an image of an object at a self-checkout comprises capturing, via a camera, an image of an object as it moves across an open conveyor belt of a self-checkout.
11. The method of claim 1 , wherein said eliminating comprises eliminating one or more blocks from consideration via use of an adaptive threshold computed on the confidence value for each of the one or more blocks.
12. The method of claim 11 , wherein eliminating one or more blocks from consideration via use of an adaptive threshold computed on the confidence value for each of the one or more blocks comprises iteratively eliminating one or more blocks from consideration via use of an adaptive threshold computed on the confidence value for each of the one or more blocks.
13. The method of claim 1 , further comprising computing one or more features of the image.
14. The method of claim 1 , wherein said computing a confidence value for each of the one or more blocks comprises using a minimum feature distance from one or more reference background blocks.
15. A computer program product comprising a tangible computer readable recordable storage memory device including computer useable program code for segmenting an object at a self-checkout, the computer program product including:
computer useable program code for capturing an image of an object at a self-checkout;
computer useable program code for dividing the image into one or more blocks;
computer useable program code for computing a confidence value for each of the one or more blocks; and
computer useable program code for eliminating one or more blocks from consideration based on the confidence value for each of the one or more blocks, wherein the one or more blocks remaining map to a region of the image containing the object.
16. The computer program product of claim 15 , further comprising:
computer useable program code for reading a barcode for the object at the self-checkout; and
computer useable program code for using a database to identify one or more known features corresponding to the barcode.
17. The computer program product of claim 15 , further comprising computer useable program code for creating and maintaining a set of one or more reference background images.
18. The computer program product of claim 15 , wherein said eliminating comprises eliminating one or more blocks from consideration via use of an adaptive threshold computed on the confidence value for each of the one or more blocks.
19. The computer program product of claim 15 , wherein said computing a confidence value for each of the one or more blocks comprises using a minimum feature distance from one or more reference background blocks.
20. A system for segmenting an object at a self-checkout, comprising:
a memory; and
at least one processor coupled to the memory and operative to:
capture an image of an object at a self-checkout;
divide the image into one or more blocks;
compute a confidence value for each of the one or more blocks; and
eliminate one or more blocks from consideration based on the confidence value for each of the one or more blocks, wherein the one or more blocks remaining map to a region of the image containing the object.