US6111983A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Determination of image shapes using training and sectoring
Assignee: Columbia University
•
Filing: 1997-12-30
•
Granted: 2000-08-29
•
Expires: 2017-12-30
Abstract
Abstract For enhanced accuracy in determining a shape in an image, a shape model can be trained for parameter settings, based on training data for which the desired shape is specified. Such training can be applied to models in which a shape is treated in a sectored fashion, with training applied to each of the sectors. The suitability of an objective function, obtained by training or otherwise, can be assessed by using near-correct shapes which are generated by randomly perturbing a ground-truth shape.
Inventors
- 👤 Samuel D. Fenster
- 👤 John R. Kender
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 22:52
CPC Classifications
G; G06; G06V; G06V10/00; G06V10/70; G06V10/74; G06V10/75; G06V10/754
Claims Analysis
Total Claims: 26
Claim 1
Independent Claim
199 words
A method for generating an objective function for a deformable model of a shape boundary, comprising: dividing the model of the shape boundary into a plurality of sectors; selecting, for each one of the plurality of sectors of the model of the shape boundary, a continuous model of feature value density for at least one feature of the boundary; applying corresponding sectors of a plurality of training images to the plurality of sectors of the model of the shape boundary, each one of the corresponding sectors of the plurality of training images representing ground truth of corresponding sectors of the shape boundary; obtaining, from each of the corresponding sectors of the plurality of training images and for each of the sectors of the model of the shape boundary corresponding thereto, a feature value for at least one preselected feature, resulting in a respective distribution of feature values for each of the sectors of the model of the shape boundary; deriving, for each of the sectors of the model of the shape boundary, a distribution function from the continuous model and the respective plurality of feature values; and using the derived distribution functions in generating a sector dependent objective function.
Claim 2
Independent Claim
199 words
A method for generating an objective function for a deformable model of a shape boundary, comprising: dividing the model of the shape boundary into a plurality of sectors; selecting, for each one of the plurality of sectors of the model of the shape boundary, a continuous model of feature value density for at least one feature of the boundary; applying corresponding sectors of a plurality of training images to the plurality of sectors of the model of the shape boundary, each one of the corresponding sectors of the plurality of training images representing ground truth of corresponding sectors of the shape boundary; obtaining, from each of the corresponding sectors of the plurality of training images and for each of the sectors of the model of the shape boundary corresponding thereto, a feature value for at least one preselected feature, resulting in a respective distribution of feature values for each of the sectors of the model of the shape boundary; deriving, for each of the sectors of the model of the shape boundary, a distribution function from the continuous model and the respective plurality of feature values; and using the derived distribution functions in generating a sector dependent objective function.
Claim 3
Dependent Claim
12 words
The method according to claim 1, wherein the continuous model is Gaussian.
Claim 4
Dependent Claim
15 words
The method according to claim 1, wherein the continuous model is multi-dimensional, representing multiple features.
Claim 5
Dependent Claim
12 words
The method according to claim 3, wherein the continuous model is Gaussian.
Claim 6
Dependent Claim
16 words
The method according to claim 1, wherein the boundary is the contour of a two-dimensional shape.
Claim 7
Dependent Claim
13 words
The method according to claim 1, wherein the feature is an image feature.
Claim 8
Dependent Claim
13 words
The method according to claim 6, wherein the image feature is visual intensity.
Claim 9
Dependent Claim
18 words
The method according to claim 6, wherein the image feature is visual intensity gradient perpendicular to image shape.
Claim 10
Dependent Claim
13 words
The method according to claim 1, wherein the feature is a shape feature.
Claim 11
Dependent Claim
13 words
The method according to claim 9, wherein the shape feature is boundary smoothness.
Claim 12
Independent Claim
202 words
A system for venerating an objective function for a deformable model of a shape boundary, comprising a processor which is instructed for: dividing the model of the shape boundary into a plurality of sectors; selecting, for each of the plurality of sectors of the model of the shape boundary, a continuous model of feature value density for at least one feature value of the boundary; applying corresponding sectors of a plurality of training images to the plurality of sectors of the model of the shape boundary, each of the corresponding sectors of the plurality of training images representing around truth of corresponding sectors of the shape boundary; obtaining, from each of the corresponding sectors plurality of training images and for each of the sectors of the model of the shape boundary corresponding thereto, a feature value for at least one preselected feature, resulting in a respective distribution of feature values for each of the sectors of the model of the shape boundary; deriving, for each of the sectors of the model of the shape boundary, a distribution function from the continuous model and the respective plurality of feature values; and using the derived distribution functions in generating a sector dependent objective function.
Claim 13
Independent Claim
202 words
A system for venerating an objective function for a deformable model of a shape boundary, comprising a processor which is instructed for: dividing the model of the shape boundary into a plurality of sectors; selecting, for each of the plurality of sectors of the model of the shape boundary, a continuous model of feature value density for at least one feature value of the boundary; applying corresponding sectors of a plurality of training images to the plurality of sectors of the model of the shape boundary, each of the corresponding sectors of the plurality of training images representing around truth of corresponding sectors of the shape boundary; obtaining, from each of the corresponding sectors plurality of training images and for each of the sectors of the model of the shape boundary corresponding thereto, a feature value for at least one preselected feature, resulting in a respective distribution of feature values for each of the sectors of the model of the shape boundary; deriving, for each of the sectors of the model of the shape boundary, a distribution function from the continuous model and the respective plurality of feature values; and using the derived distribution functions in generating a sector dependent objective function.
Claim 14
Independent Claim
207 words
A system for generating an objective function for a deformable model of a shape boundary, comprising: means for dividing the model of the shape boundary into a plurality of sectors; means for selecting, for each of the plurality of sectors of the model of the shape boundary, a continuous model of feature value density for at least one feature of the boundary; means for applying corresponding sectors of a plurality of training images to the sectors of the model of the shape boundary, each of the corresponding sectors of the plurality of training images representing ground truth of corresponding sectors of the shape boundary; means for obtaining, from each of the corresponding sectors of the plurality of training images and for each of the sectors of the model of the shape boundary corresponding thereto, a feature value for at least one preselected feature, resulting in a respective distribution of feature values for each of the sectors of tile model of the shape boundary; means for deriving, for each of the sectors of the model of the shape boundary, a distribution function for the continuous model and the respective plurality of feature values; and means for using the derived distribution functions in generating a sector dependent objective function.
Claim 15
Independent Claim
207 words
A system for generating an objective function for a deformable model of a shape boundary, comprising: means for dividing the model of the shape boundary into a plurality of sectors; means for selecting, for each of the plurality of sectors of the model of the shape boundary, a continuous model of feature value density for at least one feature of the boundary; means for applying corresponding sectors of a plurality of training images to the sectors of the model of the shape boundary, each of the corresponding sectors of the plurality of training images representing ground truth of corresponding sectors of the shape boundary; means for obtaining, from each of the corresponding sectors of the plurality of training images and for each of the sectors of the model of the shape boundary corresponding thereto, a feature value for at least one preselected feature, resulting in a respective distribution of feature values for each of the sectors of tile model of the shape boundary; means for deriving, for each of the sectors of the model of the shape boundary, a distribution function for the continuous model and the respective plurality of feature values; and means for using the derived distribution functions in generating a sector dependent objective function.
Claim 16
Independent Claim
67 words
A method for rating an objective function for a deformable model of a specified shape boundary, comprising: generating a plurality of perturbed boundaries which differ from the specified shape boundary to different degrees; evaluating the objective function for each of the perturbed boundaries; and statistically analyzing how the objective function values are related to the degree of difference between the specified shape boundary and the perturbed boundaries.
Claim 17
Independent Claim
67 words
A method for rating an objective function for a deformable model of a specified shape boundary, comprising: generating a plurality of perturbed boundaries which differ from the specified shape boundary to different degrees; evaluating the objective function for each of the perturbed boundaries; and statistically analyzing how the objective function values are related to the degree of difference between the specified shape boundary and the perturbed boundaries.
Claim 18
Dependent Claim
22 words
The method according to claim 13, wherein generating the perturbed boundaries comprises applying normally distributed first-order deformations to the specified shape boundary.
Claim 19
Dependent Claim
27 words
The method according to claim 13, wherein the degree of difference between the specified shape boundary and each of the perturbed boundaries is determined by chamfer distance.
Claim 20
Dependent Claim
14 words
The method according to claim 13, wherein statistically analyzing comprises using a scatter plot.
Claim 21
Dependent Claim
14 words
The method according to claim 13, wherein statistically analyzing comprises calculating a correlation coefficient.
Claim 22
Dependent Claim
13 words
The method according to claim 13, wherein statistically analyzing comprises counting false positives.
Claim 23
Independent Claim
73 words
A system for rating an objective function for a deformable model of a specified shape boundary, comprising a processor which is instructed for: generating a plurality of perturbed boundaries which differ from the specified shape boundary to different degrees; evaluating the objective function for each of the perturbed boundaries; and statistically analyzing how the objective function values are related to the degree of difference between the specified shape boundary and the perturbed boundaries.
Claim 24
Independent Claim
73 words
A system for rating an objective function for a deformable model of a specified shape boundary, comprising a processor which is instructed for: generating a plurality of perturbed boundaries which differ from the specified shape boundary to different degrees; evaluating the objective function for each of the perturbed boundaries; and statistically analyzing how the objective function values are related to the degree of difference between the specified shape boundary and the perturbed boundaries.
Claim 25
Independent Claim
73 words
A system for rating an objective function for a deformable model of a specified shape boundary, comprising: means for generating a plurality of perturbed boundaries which differ from the specified shape boundary to different degrees; means for evaluating the objective function for each of the perturbed boundaries; and means for statistically analyzing how the objective function values are related to the degree of difference between the specified shape boundary and the perturbed boundaries.
Claim 26
Independent Claim
73 words
A system for rating an objective function for a deformable model of a specified shape boundary, comprising: means for generating a plurality of perturbed boundaries which differ from the specified shape boundary to different degrees; means for evaluating the objective function for each of the perturbed boundaries; and means for statistically analyzing how the objective function values are related to the degree of difference between the specified shape boundary and the perturbed boundaries.
References Cited (Backward Citations)
US5233670A
Cited
US5458126A
Cited
US5644656A
Cited
US5862245A
Cited
US5768413A
Cited
US5845009A
Cited
US5926568A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2003-08-29 | 2004-02-28 | 2004-08-29 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2007-08-29 | 2008-02-28 | 2008-08-29 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2011-08-29 | 2012-02-28 | 2012-08-29 | $7700.00 | Window Closed / Lapsed |