Feature encoding system and method for connected component labeling

• US 8,111,919 B2
• Filed: 03/12/2008
• Issued: 02/07/2012
• Est. Priority Date: 02/04/2008
• Status: Expired due to Fees

• Current Assignee: Eyep, Inc.
• Original Assignee: Eyep, Inc.
• Inventors: Sullender, Craig
• Primary Examiner(s): Sherali, Ishrat I
• Application Number:
  US12/046,464
• Publication Number:
  US 20090196505A1
• Time in Patent Office: 1,427 Days
• Field of Search: 382/180, 382162-167, 382/170, 382/171, 382/173, 382190-195, 382/141, 382/199, 382/201, 382/206, 382/224
• US Class Current: 382/180
• CPC Class Codes: G06V 10/457 by analysing connectivity, e.g. edge linking, connected component analysis or slices

Abstract

Embodiments disclosed include methods and systems for encoding one or more region features in connected components labeling including associating one or more labels for an object with a memory structure, the memory structure including the one or more region features; storing the one or more region features in the memory structure, the one or more region features processed in raster order to provide a correspondence between one or more region properties and an original location of the object; enabling the memory structure to receive one or more extents of the one or more region properties at an adjustable precision and with an adjustable data rate, the adjustable precision and the adjustable data rate determined as a function of an amount of detail to be stored; and enabling the memory structure to receive one or more extents at an adjustable data rate determined as a function of an amount of detail to be stored independent of pixel data.

Claims

1. A method for encoding one or more region features in connected components labeling, the method comprising:

associating one or more labels for an object with a memory structure, the memory structure including the one or more region features;

storing the one or more region features in the memory structure, the one or more region features processed in raster order to provide a correspondence between one or more region properties and an original location of the object;

enabling the memory structure to receive one or more extents of the one or more region properties at an adjustable precision and with an adjustable data rate, the adjustable precision and the adjustable data rate determined as a function of an amount of detail to be stored; and

enabling the memory structure to receive one or more extents at an adjustable data rate determined as a function of an amount of detail to be stored independent of pixel data.

2. The method of claim 1 wherein the associating one or more labels for an object with a memory structure, the memory structure including the one or more region features includes:

extracting the one or more region features as identifying a boundary of a region, the one or more region features including one or more of coefficients and/or constants descriptive of the boundary; and

combining the one or more region properties as one or more region features.

3. The method of claim 2 wherein the extracting the one or more region features as identifying a boundary of a region, the one or more region features including one or more of coefficients and/or constants descriptive of the boundary includes:

determining the boundary using a chain of border coordinates, the chain or border coordinates derived from the region extents.

4. The method of claim 1 wherein the enabling the memory structure to receive one or more extents of the one or more region properties at an adjustable precision and with an adjustable data rate, the adjustable precision and the adjustable data rate determined as a function of an amount of detail to be stored includes:

determining the adjustable precision according to a user determination.

5. The method of claim 1 wherein the enabling the memory structure to receive one or more extents of the one or more region properties at an adjustable precision and with an adjustable data rate, the adjustable precision and the adjustable data rate determined as a function of an amount of detail to be stored includes:

receiving a data rate from an application, the application determining an amount of detail required to be stored in the memory structure.

6. The method of claim 5 wherein the receiving a data rate from an application, the application determining an amount of detail required to be stored in the memory structure includes:

receiving the data rate from an application for motion detection as a low data rate; and

receiving the data rate from an application for object recognition as a high data rate.

7. The method of claim 1 further comprising:

grouping one or more segments of the object into subcomponents, each subcomponent representing one or more of the region feature.

8. The method of claim 7 wherein the grouping one or more segments of the object into subcomponents, each subcomponent representing one or more of the region feature includes:

storing the subcomponents to enable deriving one or more contours of the object;

deriving external contours by flagging one or more endpoints of an extent of the one or more subcomponents; and

deriving internal contours by identifying connected unflagged extents of the one or more subcomponents.

9. The method of claim 1 wherein the enabling the memory structure to receive one or more extents of the one or more region properties at an adjustable precision and with an adjustable data rate, the adjustable precision and the adjustable data rate determined as a function of an amount of detail to be stored includes:

enabling the adjustable precision extents to be referenced from the labels in a spatial order as a function of a location of the one or more objects in an image.

10. The method of claim 1 wherein the one or more region features are received from raster-organized data arrays wherein the data is image data arranged to display an image.

11. The method of claim 1 wherein one or more region features are stored in the memory structure to enable random access by one or more label designators according to a location of the one or more region features.

12. The method of claim 1 wherein one or more region features are stored in the memory structure to enable ordering the one or more region features based on one or more of size and/or shape.

13. A connected component feature encoding system for encoding one or more region features, the system comprising:

a memory structure configured to be associated with one or more labels for an object, the memory structure including;

one or more region features configured to be stored in the memory structure in raster order to provide a spatial correspondence between the memory structure and an original location of the object; and

one or more pointers configured to be stored in the memory structure and configured to at least provide a linking between the original location of the object, the one or more region features and the one or more labels for the object; and

a processor coupled to the memory structure, the processor configured to include;

a data rate module coupled to the memory structure to enable the memory structure to receive one or more extents of the one or more region features at an adjustable precision and with an adjustable data rate, the adjustable precision and the adjustable data rate determined as a function of an amount of detail to be stored in the memory structure; and

a boundary module coupled to the memory module, the boundary module configured to extract the one or more region features as identifying a boundary of a region, the one or more region features including one or more of coefficients and/or constants descriptive of the boundary, the boundary module configured to combine the one or more region features as one or more region properties.

14. The connected component feature encoding system of claim 13 wherein the boundary module is configured to determine the boundary using a chain of border coordinates, the chain or border coordinates derived from the one or more region features.

15. The connected component feature encoding system of claim 13 wherein the data rate module is configured to receive the adjustable precision from a user, an application coupled to the connected component feature encoding system.

16. The connected component feature encoding system of claim 13 wherein the data rate module is configured to reference adjustable precision extents from the labels in a spatial order as a function of a location of the one or more objects in an image.

17. The connected component feature encoding system of claim 13 wherein the boundary module is configured to determine the boundary using a chain of border coordinates, the chain or border coordinates derived from the one or more region features.