US6941016B1
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method for finding contours in an image of an object
Assignee: Cognex Technology and Investment LLC
•
Filing: 2001-12-31
•
Granted: 2005-09-06
•
Expires: 2021-12-31
Abstract
Abstract A method is provided for robustly and efficiently identifying contours in an image of an object. Features are extracted from an image, typically using a Sobel edge detector or a set of large oriented filters, and are locally chained to form contour segments of particular character, such as line segments. A spatial hash table is used to access the contour segments according to the location of their endpoints. Candidate sets for joining to grow contours are reduced by considering only segments having ends in adjacent spatial hash bins. Contour constraints are applied to appropriately choose among the candidates to form contours. The method facilitates robust and efficient scratch detection of fiber-optic end faces, or of any surface that may have smoothly curved scratches.
Inventors
- 👤 Adam Wagman
- 👤 Ivan Bachelder
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 10:18
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/0002; G06T7/0004; G06T5/00; G06T5/77; G06T7/10; G06T7/12; G06T7/181; G06V; G06V10/00; G06V10/40; G06V10/44; G06V10/457; G06V10/46; G06T2207/00; G06T2207/10; G06T2207/10056; G06T2207/20; G06T2207/20021; G06T2207/30; G06T2207/30108; G06V10/478
Claims Analysis
Total Claims: 32
Claim 1
Independent Claim
81 words
A method of identifying contours in an image comprising the steps of: extracting features from said image; forming chains from localized subsets of said features; computing the values of one or more hash functions for contour segments, wherein each said contour segment comprises at least one of said chains; identifying, using said values for said contour segments, candidate contour segments for joining; selecting contour segments for joining from said candidates; and joining said selected contour segments to form joined contour segments.
Claim 2
Independent Claim
81 words
A method of identifying contours in an image comprising the steps of: extracting features from said image; forming chains from localized subsets of said features; computing the values of one or more hash functions for contour segments, wherein each said contour segment comprises at least one of said chains; identifying, using said values for said contour segments, candidate contour segments for joining; selecting contour segments for joining from said candidates; and joining said selected contour segments to form joined contour segments.
Claim 3
Dependent Claim
23 words
The method according to claim 1 further comprising the step of repeating said steps of identifying, selecting, and joining for additional contour segments.
Claim 4
Dependent Claim
20 words
The method according to claim 2 further comprising the step of computing additional hash functions for said additional contour segments.
Claim 5
Dependent Claim
19 words
The method according to claim 1 wherein said step of forming chains includes applying one or more chaining constraints.
Claim 6
Dependent Claim
16 words
The method of claim 1 wherein said step of selecting contour segments includes using contour constraints.
Claim 7
Dependent Claim
51 words
The method according of claim 1 wherein said step of extracting features comprises the steps of: applying a set of filters to said image to produce a magnitude image and an angle image; and applying truepeak and subpixel interpolation to said magnitude image and said angle image to produce said features.
Claim 8
Dependent Claim
11 words
The method according to claim 6 wherein said features comprise edgelets.
Claim 9
Dependent Claim
13 words
The method according to claim 6 wherein said filters comprise large oriented filters.
Claim 10
Dependent Claim
12 words
The method according to claim 6 wherein said filters comprise Sobel filters.
Claim 11
Dependent Claim
37 words
The method according to claim 7 wherein said step of forming chains comprises the steps of: applying 3 by 3 connectivity and angle constraints to connect said edgelets into chains; and breaking said chains into linear sub-chains.
Claim 12
Dependent Claim
27 words
The method according to claim 10 wherein said linear sub-chains are based on Root Mean Square (RMS) error of a line fit to subsections of said chain.
Claim 13
Dependent Claim
20 words
The method according to claim 10 further comprising the step of discarding sub-chains that are shorter than a predetermined length.
Claim 14
Dependent Claim
19 words
The method according to claim 10 further comprising the steps of discarding sub-chains that have a low average magnitude.
Claim 15
Dependent Claim
30 words
The method according to claim 10 wherein said step of computing one or more hash functions comprises the step of computing a bounding box including all endpoints of said sub-chains.
Claim 16
Dependent Claim
44 words
The method according to claim 14 further comprising the steps of: constructing a spatial hash table using said bounding box and a bin size; making said sub-chains accessible with said hash table according to spatial hash values of endpoints of each of said sub-chains.
Claim 17
Dependent Claim
25 words
The method according to claim 15 wherein said bin size is based upon a maximum allowable gap between endpoints of sub-chains being considered for joining.
Claim 18
Dependent Claim
18 words
The method according to claim 15 further comprising the step of identifying a seed sub-chain from said sub-chains.
Claim 19
Dependent Claim
34 words
The method according to claim 17 further comprising the step of identifying candidate sub-chains for joining through said spatial hash table using 3 by 3 bin neighborhoods around each endpoint of said seed sub-chain.
Claim 20
Dependent Claim
38 words
The method according to claim 18 further comprising the steps of: computing sub-chain endpoint to endpoint distances; and removing sub-chain candidates from consideration for joining if said endpoint to endpoint distances are not within predetermined allowable gap distances.
Claim 21
Dependent Claim
41 words
The method according to claim 19 further comprising the steps of: sorting candidate sub-chains by said endpoint distances; and considering for joining said sub-chains in order according said endpoint distances wherein said sub-chains having the smallest endpoint distances are considered first.
Claim 22
Dependent Claim
21 words
The method according to claim 20 further comprising the step of applying constraints to determine whether to join said candidate sub-chains.
Claim 23
Dependent Claim
15 words
The method according to claim 21 wherein said constraints include a line to line angle.
Claim 24
Dependent Claim
60 words
The method according to claim 21 further comprising the steps of: combining said seed sub-chain with a candidate sub-chain to form an extended seed sub-chain; removing said candidate sub-chain from consideration for further joining; removing said candidate sub-chain from said spatial hash table; and updating a hash value of said seed sub-chain according to end-points of said extended seed sub-chain.
Claim 25
Dependent Claim
43 words
The method according to claim 23 further comprising the steps of repeating said steps of applying constraints, combining, removing said candidate sub-chain from further consideration, removing said candidate sub-chain from said spatial hash table, and said updating step until no candidate sub-chains remain.
Claim 26
Dependent Claim
32 words
The method according to claim 24 further comprising the steps of repeating said steps including and following said step of identifying candidate sub-chains until no more candidate sub-chains are available for joining.
Claim 27
Dependent Claim
36 words
The method according to claim 25 further comprising the steps of repeating said steps including and following said step of identifying a seed subchain until no more sub-chains are available to use as a seed sub-chain.
Claim 28
Dependent Claim
31 words
The method according to claim 26 further comprising the step of: selecting a new maximum allowable gap size; and repeating said steps following said step of selecting a maximum allowable gap.
Claim 29
Dependent Claim
13 words
The method according to claim 1 where said joined contour segments comprise contours.
Claim 30
Dependent Claim
30 words
The method according to claim 7 wherein said step of forming chains comprises the step of applying 3 by 3 connectivity and angle constraints to connect said edgelets into chains.
Claim 31
Independent Claim
93 words
A method of identifying scratch defects in an image of a fiber end comprising the steps of: extracting features from said image using oriented filters; forming chains from localized subsets of said features using 3 by 3 connectivity rules; constructing a spatial hash table for contour segments using contour segment endpoints, wherein each said contour segment comprises at least one of said chains; identifying, using said spatial hash table, candidate contour segments for joining; selecting contour segments for joining from said candidates; and joining said selected contour segments to form identified scratch defects.
Claim 32
Independent Claim
93 words
A method of identifying scratch defects in an image of a fiber end comprising the steps of: extracting features from said image using oriented filters; forming chains from localized subsets of said features using 3 by 3 connectivity rules; constructing a spatial hash table for contour segments using contour segment endpoints, wherein each said contour segment comprises at least one of said chains; identifying, using said spatial hash table, candidate contour segments for joining; selecting contour segments for joining from said candidates; and joining said selected contour segments to form identified scratch defects.
References Cited (Backward Citations)
US4876729A
Cited
US5179419A
Cited
US5319734A
Cited
US5596672A
Cited
US5727327A
Cited
US5535002A
Cited
US5543915A
Cited
US5600439A
Cited
US5636020A
Cited
US5898494A
Cited
US5671049A
Cited
US5768401A
Cited
US5768409A
Cited
US5729622A
Cited
US5862250A
Cited
US5657131A
Cited
US5809162A
Cited
US5923781A
Cited
US5857047A
Cited
US5729966A
Cited
US6183343B1
Cited
US6088498A
Cited
US6069991A
Cited
US5857049A
Cited
US6108444A
Cited
US6105396A
Cited
US5995212A
Cited
US6753965B2
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2008-09-05 | 2009-03-07 | 2009-09-06 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2012-09-05 | 2013-03-07 | 2013-09-06 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2016-09-05 | 2017-03-07 | 2017-09-06 | $7700.00 | Window Closed / Lapsed |