Depth-based image element removal
Various embodiments herein each include at least one of systems, methods, and software to enable depth-based image element removal. Some embodiments may be implemented in a store checkout context, while other embodiments may be implemented in other contexts such as at price-checking kiosks or devices that may be deployed within a store or other retail establishment, a library at a checkout terminal, and the like. Some embodiments include removing elements of images based at least in part on depth data.
1. A method comprising:
receiving image data including at least one image and depth data captured with regard to an imaging area of a product scanner;
for each of the at least one images:
removing pixels from the image for which pixel depth data of corresponding pixels is not included in the depth data; and
identifying and removing matching shapes between a skeletal representation of the image and at least one shape wire-frame model; and
providing each of the at least one images to another process for further processing.
2. The method of claim 1 , further comprising:
obtaining a skeletal representation of a respective image, the obtaining comprising:
performing edge detection on the image to identify dominant shapes in the image; and
applying a skeletal transform to the image and the edges detected therein to obtain the skeletal image.
3. The method of claim 1 , wherein the at least one shape wire-frame model includes shape wire-frame models of human arms, hands, and fingers.
4. The method of claim 3 , wherein for each of the at least one images, the method further comprising:
identifying skin color of at least one of an arm, hand, and finger identified by the wire-frame matching with regard to at least one of the images; and
removing pixels matching the identified skin color.
5. The method of claim 4 , wherein matching the identified skin color is performed according to probabilistic matching algorithm.
6. The method of claim 1 , further comprising:
removing from each of the at least one images, a respective background.
7. The method of claim 1 , wherein the further processing includes a process that compares the image data to image data of a product identified via barcode scanning to determine whether the received image data represents a product of the product identified via the barcode scanning.
8. A method comprising:
receiving image data including at least one image and depth data captured with regard to an imaging area;
for each of the at least one images:
removing pixels from the image for which pixel depth data of corresponding pixels is not included in the depth data;
obtaining a skeletal image of the image;
perform wire-frame matching between the skeletal image and at least one shape wire-frame model to identify matching shapes within the respective image; and
removing pixels of matching shapes from respective images;
providing each of the at least one images to another process for further processing.
9. The method of claim 8 , wherein obtaining a skeletal image of the image includes:
performing edge detection on the image to identify dominant shapes in the image; and
applying a skeletal transform to the image and the edges detected therein to obtain the skeletal image.
10. The method of claim 8 , wherein the at least one shape wire-frame model includes shape wire-frame models of human arms, hands, and fingers.
11. The method of claim 10 , wherein for each of the at least one images, the method further comprising:
identifying skin color of at least one of an arm, hand, and finger identified by the wire-frame matching with regard to at least one of the images; and
removing pixels matching the identified skin color.
12. The method of claim 11 , wherein matching the identified skin color is performed according to probabilistic matching algorithm.
13. The method of claim 8 , further comprising:
removing from each of the at least one images, a respective background.
14. The method of claim 8 , wherein the further processing includes a process that compares the image data to image data of a product identified via barcode scanning to determine whether the received image data represents a product of the product identified via the barcode scanning.
15. A system comprising:
at least one processor, at least one memory device, and an image data interface;
an instruction set accessible in the at least one memory device and executable by the at least one processor to perform data processing activities, the data processing activities comprising:
receiving, via the image data interface, image data including at least one image and depth data captured with regard to an imaging area of a product scanner;
for each of the at least one images:
removing a calibrated background from the image;
removing pixels from the image for which pixel depth data of corresponding pixels is not included in the depth data; and
identifying and removing matching shapes between a skeletal representation of the image and at least one shape wire-frame model; and
providing each of the at least one images to another process for further processing.
16. The system of claim 15 , the data processing activities further comprising:
obtaining a skeletal representation of a respective image, the obtaining comprising:
performing edge detection on the image to identify dominant shapes in the image; and
applying a skeletal transform to the image and the edges detected therein to obtain the skeletal image.
17. The system of claim 15 , wherein the at least one shape wire-frame model includes shape wire-frame models of human arms, hands, and fingers.
18. The system of claim 3 , wherein for each of the at least one images, the method further comprising:
identifying skin color of at least one of an arm, hand, and finger identified by the wire-frame matching with regard to at least one of the images; and
removing pixels matching the identified skin color.
19. The system of claim 18 , wherein matching the identified skin color is performed according to probabilistic matching algorithm.
20. The system of claim 15 , wherein the further processing includes a process that compares the image data to image data of a product identified via barcode scanning to determine whether the received image data represents a product of the product identified via the barcode scanning.