Locating curvilinear objects using feathered fiducials
First Claim
1. A method for locating a curvilinear object in a digitized image, comprising:
- a) defining a range of scale variations;
b) generating an enlarged submodel having dimensions larger than those expected for said curvilinear object within said range of scale variations;
c) generating a reduced size submodel of said curvilinear object having dimensions smaller than those expected for said curvilinear object and within said range of scale variations;
d) disposing said reduced size submodel within said enlarged submodel and tangent to said enlarged submodel;
e) generating a feathered submodel by varying pixel grey scale values between said enlarged and said reduced size submodels;
f) conducting a correlation search of said digitized image of said curvilinear object using said feathered submodel, whereby the point in said digitized image reported by said correlation search as having the highest correlation to said feathered submodel is deemed to be at an edge of said curvilinear object and used to compute a center of said curvilinear object to locate it.
1 Assignment
0 Petitions
Accused Products
Abstract
A method and apparatus for locating curvilinear objects of variable scale in digitized images using feathered fiducials. The leathered fiducials are constructed by: defining a range of tolerated object size scale variations; generating an enlarged submodel within the range of tolerated scale variations and assigning background pixel values exterior to the enlarged submodel; generating a reduced size submodel within the range of tolerated scale variations and assigning foreground pixel values to the reduced size submodel; disposing the reduced size submodel within and tangent to the enlarged submodel at the application point; uniformly changing grey scale pixel values in a graduated manner from foreground to background between the reduced size submodel and the enlarged one. The curvilinear objects are then located by conducting searches using the constructed fiducials.
71 Citations
22 Claims
-
1. A method for locating a curvilinear object in a digitized image, comprising:
-
a) defining a range of scale variations; b) generating an enlarged submodel having dimensions larger than those expected for said curvilinear object within said range of scale variations; c) generating a reduced size submodel of said curvilinear object having dimensions smaller than those expected for said curvilinear object and within said range of scale variations; d) disposing said reduced size submodel within said enlarged submodel and tangent to said enlarged submodel; e) generating a feathered submodel by varying pixel grey scale values between said enlarged and said reduced size submodels; f) conducting a correlation search of said digitized image of said curvilinear object using said feathered submodel, whereby the point in said digitized image reported by said correlation search as having the highest correlation to said feathered submodel is deemed to be at an edge of said curvilinear object and used to compute a center of said curvilinear object to locate it. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A method for locating a curvilinear object in a digitized image, comprising:
-
a) defining a range of scale variations by specifying a nominal submodel size and a percentage of acceptable size variations larger and smaller than said nominal submodel size; b) generating an enlarged submodel having dimensions larger than those expected for said curvilinear object within said range of scale variations; c) generating a reduced size submodel of said curvilinear object having dimensions smaller than those expected for said curvilinear object and within said range of scale variations; d) disposing said reduced size submodel within said enlarged submodel and tangent to said enlarged submodel; e) computing a fill area between said enlarged submodel and said reduced size submodel; f) generating a feathered submodel having varying pixel grey scale values between said enlarged and said reduced size submodels by creating a smoothly varying grey scale fill in said fill area; g) conducting a correlation search of said digitized image using said feathered submodel, whereby the point in said digitized image reported by said correlation search as having the highest correlation to said feathered submodel is deemed to be at an edge of said curvilinear object and used to compute a center of said curvilinear object to locate it.
-
-
11. A method for locating a curvilinear object in a digitized image, comprising:
-
a) defining a range of scale variations; b) generating an enlarged submodel having dimensions larger than those expected for said curvilinear object within said range of scale variations; c) generating a reduced size submodel of said curvilinear object having dimensions smaller than those expected for said curvilinear object and within said range of scale variations; d) disposing said reduced size submodel within said enlarged submodel and tangent to said enlarged submodel; e) computing a fill area between said enlarged submodel and said reduced size submodel; f) generating a feathered submodel having varying pixel grey scale values between said enlarged and said reduced size submodels by creating a smoothly varying grey scale fill in said fill area; g) conducting a correlation search of a digitized image of said curvilinear object using said feathered submodel, whereby the point in said digitized image reported by said correlation search as having the highest correlation to said feathered submodel is deemed to be at an edge of said curvilinear object and used to compute a center of said curvilinear object to locate it.
-
-
12. An apparatus for locating a curvilinear object in a digitized image, comprising:
-
a) means for defining a range of scale variations; b) means for generating an enlarged submodel having dimensions larger than those expected for said curvilinear object within said range of scale variations; c) means for generating a reduced size submodel of said curvilinear object having dimensions smaller than those expected for said curvilinear object and within said range of scale variations; d) means for disposing said reduced size submodel within said enlarged submodel and tangent to said enlarged submodel; e) means for generating a feathered submodel by varying pixel grey scale values between said enlarged and said reduced size submodels; f) means for conducting a correlation search of said digitized image of said object using said feathered submodel, whereby the point in said digitized image reported by said correlation search as having the highest correlation to said feathered submodel is deemed to be at an edge of said curvilinear object and used to compute a center of said curvilinear object to locate it. - View Dependent Claims (13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. An apparatus for locating a curvilinear object in a digitized image, comprising:
-
a) means for defining a range of scale variations by specifying a nominal submodel size and a percentage of acceptable size variations larger and smaller than said nominal submodel size; b) means for generating an enlarged submodel having dimensions larger than those expected for said curvilinear object within said range of scale variations; c) means for generating a reduced size submodel of said curvilinear object having dimensions smaller than those expected for said curvilinear object and within said range of scale variations; d) means for disposing said reduced size submodel within said enlarged submodel and tangent to said enlarged submodel; e) means for computing a fill area between said enlarged submodel and said reduced size submodel; f) means for generating a feathered submodel having varying pixel grey scale values between said enlarged and said reduced size submodels by creating a smoothly varying grey scale fill in said fill area; g) means for conducting a correlation search of said digitized image of said object using said feathered submodel, whereby the point in said digitized image reported by said correlation search as having the highest correlation to said feathered submodel is deemed to be at an edge of said curvilinear object and used to compute a center of said curvilinear object to locate it.
-
-
22. An apparatus for locating a curvilinear object, comprising:
-
a) means for defining a range of scale variations; b) means for generating an enlarged submodel having dimensions larger than those expected for said curvilinear object within said range of scale variations; c) means for generating a reduced size submodel of said curvilinear object having dimensions smaller than those expected for said curvilinear object and within said range of scale variations; d) means for disposing said reduced size submodel within said enlarged submodel and tangent to said enlarged submodel; e) means for computing a fill area between said enlarged submodel and said reduced size submodel; f) means for generating a feathered submodel having varying pixel grey scale values between said enlarged and said reduced size submodels by creating a smoothly varying grey scale fill in said fill area; g) means for conducting a correlation search of a digitized image of said object using said feathered submodel, whereby the point in said digitized image reported by said correlation search as having the highest correlation to said feathered submodel is deemed to be at an edge of said curvilinear object and used to compute a center of said curvilinear object to locate it.
-
Specification