Determine the shape of a representation of an object

Patent metadata

Field	Value
Publication number	US-9846949-B2
Application number	US-201315036796-A
Country	US
Kind code	B2
Filing date	Nov 27, 2013
Priority date	Nov 27, 2013
Publication date	Dec 19, 2017
Grant date	Dec 19, 2017

Abstract

Examples disclosed herein relate to determining the shape of an object representation. In one implementation, a processor determines contours of a silhouette of an object representation above a contour degree threshold, where a contour degree is determined based on the area of the contour with respect to itself. The processor may identify a shape type of the object representation based on a comparison of the determined contours to shape definition information.

First claim

The invention claimed is:

1. A computing system, comprising:
a storage to store shape definition information; and
a processor to:
determine contours of a silhouette of an object representation above a contour degree threshold, wherein a contour degree is determined based on the area of the contour with respect to itself; and
identify a shape type of the object based on a comparison of the determined contours to the image shape definition information,
wherein the processor is further to determine silhouette information of an inner portion of the object representation based on a second contour degree threshold.

2. The computing system of claim 1 , wherein the object representation comprises coordinates related to a file for printing an object using 3D printing technology.

3. The computing system of claim 1 , wherein the processor selects a type of additional processing on the object representation based on the identified shape type.

4. The computing system of claim 1 , wherein determining contours of a silhouette comprises:
determining a first contour in a first direction associated with the width of the silhouette; and
determining a second contour in a second direction associated with the height of the silhouette.

5. The computing system of claim 1 , wherein determining a contour comprises:
comparing three points along the silhouette a particular distance apart; and
identifying the three points as part of a contour where the area of a triangle created between the three points is above the contour degree threshold.

6. A method, comprising:
determining, by a processor, a description of a shape of a silhouette of a representation of an object, including:
detecting a first contour of the silhouette wherein the area of the first contour with respect to itself is over a contour threshold; and
detecting a second contour of a portion of the silhouette wherein the area of the second contour with respect to itself is over the contour threshold; and

identifying the shape of the silhouette based on the shape description.

7. The method of claim 6 , wherein determining a description of the shape of the silhouette comprises:
determining a description of the shape a portion of the silhouette in a first direction associated with the width of the silhouette; and
determining a description of the shape of a portion of the silhouette in a second direction associated with the height of the silhouette.

8. The method of claim 7 , wherein identifying the type of object comprises identifying the type of object based on at least one of the number and position of contours in the silhouette.

9. The method of claim 6 , wherein detecting the first contour comprises:
comparing three points along the silhouette a particular distance apart; and
identifying the three points as part of a contour where the area of a triangle created between the three points is above the contour threshold.

10. The method of claim 6 , further comprising merging the first contour and the second contour into a single contour.

11. The method of claim 6 , further comprising identifying the type of object based on the shape of the silhouette.

12. The method of claim 6 , further comprising determining the shape of an inner portion of the silhouette based on a second contour threshold.

13. The method of claim 6 , further comprising determining whether a detected contour is concave or convex.

14. A machine-readable non-transitory storage medium comprising instructions executable by a processor to:
determine contoured regions of a silhouette of a representation of an object wherein the contoured regions have a contour with an area with respect to itself above a threshold;
determine the shape of an inner portion of the silhouette based on detecting contour regions having a contour with an area with respect to itself above a second threshold; and
recognize the shape of the silhouette based on a comparison of the contour region information to shape information.

15. The machine-readable non-transitory storage medium of claim 14 , wherein the representation of the object comprises coordinates related to a file for printing the object using 3D printing technology.

16. The machine-readable non-transitory storage medium of claim 14 wherein instructions to recognize the silhouette comprise instructions to:
determine a description of the shape a portion of the silhouette in a first direction associated with the width of the silhouette; and
determine a description of the shape of a portion of the silhouette in a second direction associated with the height of the silhouette.

17. The machine-readable non-transitory storage medium of claim 14 , wherein instructions to determine a contour region comprises instructions to:
compare three points along the silhouette a particular distance apart; and
identify the three points as part of a contour region where the area of a triangle created between the three points is above the contour threshold.

18. The machine-readable non-transitory storage medium of claim 14 , further comprising instructions to merge a first and second determined contour region into a single contour region.

19. The machine-readable non-transitory storage medium of claim 14 , wherein instructions to recognize the shape comprise instructions to recognize the shape based on at least one of the number and position of contours in the silhouette.

20. The machine-readable non-transitory storage medium of claim 14 , further comprising instructions to determine whether a contour within a contour region is concave or convex.

Assignees

• Hewlett Packard Development Co Lp

Inventors

• Moroney Nathan

Classifications

• G06V10/752

  Contour matching · CPC title

• G06V10/46

  Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features (colour feature extraction G06V10/56) · CPC title

• G06T7/12

  Edge-based segmentation · CPC title

• G06T7/136

  involving thresholding · CPC title

• G06T7/64 Primary

  of convexity or concavity · CPC title