SYSTEM AND METHOD TO FIND THE PRECISE LOCATION OF OBJECTS OF INTEREST IN DIGITAL IMAGES

• US 20120033852A1
• Filed: 03/30/2011
• Published: 02/09/2012
• Est. Priority Date: 08/06/2010
• Status: Active Grant

• Current Assignee: Daniel Curtis, Michael B. Kennedy
• Original Assignee: Daniel Curtis, Michael B. Kennedy
• Inventors: Kennedy, Michael B., Curtis, Daniel
• Granted Patent: US 8,170,372 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06F 18/00 Pattern recognition

Abstract

The present invention is a method and system to precisely locate objects of interest in any given image scene space, which finds the presence of objects based upon pattern matching geometric relationships to a master, known set. The method and system prepares images for feature and attribute detection and identifies the presence of potential objects of interest, then narrows down the objects based upon how well they match a pre designated master template. The method by which matching takes place is done through finding all objects, plotting its area, juxtaposing a sweet spot overlap of its area on master objects, which in turn forms a glyph shape. The glyph shape is recorded, along with all other formed glyphs in an image'"'"'s scene space and then mapped to form sets using a classifier and finally a pattern matching algorithm. The resulting objects of interest matches are then refined to plot the contour boundaries of the object'"'"'s grouped elements (arrangement of contiguous pixels of the given object called a Co-Glyph) and finally snapped to its component actual dimensions e.g., x, y of a character or individual living cell.

Claims

1. A method for determining relative position and orientation of Objects Of Interest in an image scene, comprising:

identifying candidate objects of interest by zonal image analysis;

determining contiguous pixel organization as vectors for each candidate object of interest using symmetric rule sets;

forming for each candidate object of interest an area of influence calculated from the origin of the centroid of said contiguous pixel organization;

creating an axonometric projection of a first area of influence onto a second area of influence along a constrained path formed by the tangents of the first and second areas of influence;

forming an overlap of the first area of influence with the second area of influence using an offset rule, the overlap comprising a glyph;

incorporating into the glyph the pixel organization vectors of the object of interest in the first area of influence;

calculating the coordinates of the glyph formed by the axonometric projection,

wherein, the coordinates of the glyph including said vectors define a unique relationship between the first area of influence encompassing a first object of interest and the second area of influence encompassing a second object of interest.

2. The method of claim 1, creating an axonometric projection of a third area of influence onto the second area of influence along a constrained path formed by the tangents of the third and second areas of influence.

3. The method of claim 2, forming an overlap of the third area of influence with the second area of influence using an offset rule, the overlap comprising a second glyph, and calculating the coordinates of the second glyph formed by the axonometric projection.

4. The method of claim 3, incorporating into the second glyph the pixel organization vectors of the object of interest in the second areas of influence.

5. The method of claim 4, wherein a plurality of glyphs are recorded in a visual database according to precise orientation and residing order in the image scene.

6. The method of claim 5, wherein the relative position and orientation of the objects of interest in the image scene are preserved by recording glyphs in the visual database comprising a concentric recordation work space.

7. A device for determining relative position and orientation of Objects Of Interest in an image scene, comprising:

an image capture component for acquiring at least one image;

a storage component for storing at least one acquired image;

a computational component for executing instruction sets;

an instruction set for identifying candidate objects of interest by zonal image analysis;

an instruction set for determining contiguous pixel organization as vectors for each candidate object of interest using symmetric rule sets;

an instruction set for forming for each candidate object of interest an area of influence calculated from the origin of the centroid of said contiguous pixel organization;

an instruction set for creating an axonometric projection of a first area of influence onto a second area of influence along a constrained path formed by the tangents of the first and second areas of influence;

an instruction set for forming an overlap of the first area of influence with the second area of influence using an offset rule, the overlap comprising a glyph;

an instruction set for incorporating into the glyph the pixel organization vectors of the object of interest in the first area of influence;

an instruction set for calculating the coordinates of the glyph formed by the axonometric projection,

wherein, the coordinates of the glyph including said vectors define a unique relationship between the first area of influence encompassing a first object of interest and the second area of influence encompassing a second object of interest.

8. The device of claim 7, further comprising an instruction set for creating an axonometric projection of a third area of influence onto the second area of influence along a constrained path formed by the tangents of the third and second areas of influence;

9. The device of claim 8, further comprising an instruction set forming an overlap of the third area of influence with the second area of influence using an offset rule, the overlap comprising a second glyph, and calculating the coordinates of the second glyph formed by the axonometric projection.

10. The device of claim 9, further comprising an instruction set further comprising means incorporating into the second glyph the pixel organization vectors of the object of interest in the second areas of influence.

11. The device of claim 10, wherein a plurality of glyphs are recorded in a visual database according to precise orientation and residing order in the image scene.

12. The device of claim 11, wherein the relative position and orientation of the objects of interest in the image scene are preserved by recording glyphs in the visual database comprising a concentric recordation work space.

13. A system for determining relative position and orientation of Objects Of Interest in an image scene, comprising:

a module for identifying candidate objects of interest by zonal image analysis;

a module for determining contiguous pixel organization as vectors for each candidate object of interest using symmetric rule sets;

a module for forming for each candidate object of interest an area of influence calculated from the origin of the centroid of said contiguous pixel organization;

a module for creating an axonometric projection of a first area of influence onto a second area of influence along a constrained path formed by the tangents of the first and second areas of influence;

a module for forming an overlap of the first area of influence with the second area of influence using an offset rule, the overlap comprising a glyph;

a module for incorporating into the glyph the pixel organization vectors of the object of interest in the first area of influence;

a module for calculating the coordinates of the glyph formed by the axonometric projection,

wherein, the coordinates of the glyph including said vectors define a unique relationship between the first area of influence encompassing a first object of interest and the second area of influence encompassing a second object of interest.

14. The system of claim 13, further comprising a module for creating an axonometric projection of a third area of influence onto the second area of influence along a constrained path formed by the tangents of the third and second areas of influence;

15. The system of claim 14, further comprising a module forming an overlap of the third area of influence with the second area of influence using an offset rule, the overlap comprising a second glyph, and calculating the coordinates of the second glyph formed by the axonometric projection.

16. The system of claim 15, further comprising a module comprising means incorporating into the second glyph the pixel organization vectors of the object of interest in the second areas of influence.

17. The system of claim 16, wherein a plurality of glyphs are recorded in a visual database according to precise orientation and residing order in the image scene.

18. The system of claim 17, wherein the relative position and orientation of the objects of interest in the image scene are preserved by recording glyphs in the visual database comprising a concentric recordation work space.

19. A processor-readable storage medium, storing processor-readable instructions which cause a device to perform operations, comprising:

identifying candidate objects of interest by zonal image analysis;

determining contiguous pixel organization as vectors for each candidate object of interest using symmetric rule sets;

forming for each candidate object of interest an area of influence calculated from the origin of the centroid of said contiguous pixel organization;

creating an axonometric projection of a first area of influence onto a second area of influence along a constrained path formed by the tangents of the first and second areas of influence;

forming an overlap of the first area of influence with the second area of influence using an offset rule, the overlap comprising a glyph;

incorporating into the glyph the pixel organization vectors of the object of interest in the first area of influence;

calculating the coordinates of the glyph formed by the axonometric projection;

recording at least one glyph in a visual database according to precise orientation and residing order in the image scene,

wherein, the coordinates of the glyph including said vectors define a unique relationship between the first area of influence encompassing a first object of interest and the second area of influence encompassing a second object of interest.

20. A device configured to perform operations, comprising:

identifying candidate objects of interest by zonal image analysis;

determining contiguous pixel organization as vectors for each candidate object of interest using symmetric rule sets;

forming for each candidate object of interest an area of influence calculated from the origin of the centroid of said contiguous pixel organization;

creating an axonometric projection of a first area of influence onto a second area of influence along a constrained path formed by the tangents of the first and second areas of influence;

forming an overlap of the first area of influence with the second area of influence using an offset rule, the overlap comprising a glyph;

incorporating into the glyph the pixel organization vectors of the object of interest in the first area of influence;

calculating the coordinates of the glyph formed by the axonometric projection;

creating an axonometric projection of a third area of influence onto the second area of influence along a constrained path formed by the tangents of the third and second areas of influence;

forming an overlap of the third area of influence with the second area of influence using an offset rule, the overlap comprising a second glyph, and calculating the coordinates of the second glyph formed by the axonometric projection;

incorporating into the second glyph the pixel organization vectors of the object of interest in the second areas of influence;

recording a plurality of glyphs in a visual database according to precise orientation and residing order in the image scene; and

preserving the relative position and orientation of the objects of interest in the image scene by recording glyphs in the visual database comprising a concentric recordation,

wherein the glyph coordinates including said vectors define a unique relationship between the first area of influence encompassing a first object of interest and the second area of influence encompassing a second object of interest.