US6532302B2
Expired / Lapsed
US
⏱️ 0.0 years remaining
Multiple size reductions for image segmentation
Assignee: Canon Inc
•
Filing: 1998-04-08
•
Granted: 2003-03-11
•
Expires: 2018-04-08
Abstract
Abstract A sample image ( 144 ) is segmented by an image segmentation system ( 120 ) including a size reduction unit ( 134 ), which reduces the size of the image ( 144 ), and, at the same time, fills small gaps between foreground pixels. Thereafter, a connected component analyzer ( 136 ) identifies connected components and their associated minimum bounding rectangles in the reduced image ( 145 ). Next, a target object filter ( 138 ) searches the connected components for target objects, making use of a target object library ( 146 ) to identify target objects characterized by such parameters as size, shape, and texture. Finally, an inverse mapper ( 140 ) locates the bounding rectangles of the target objects in the original sample image ( 144 ), and extracts the associated portions of the image ( 144 ) for analysis in a conventional image classifier ( 142 ).
Inventors
- 👤 Radovan V. Krtolica
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 10:16
CPC Classifications
G; G06; G06V; G06V30/00; G06V30/40; G06V30/41; G06V30/414
Claims Analysis
Total Claims: 34
Claim 1
Independent Claim
76 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising the steps of: reducing the size of the sample image to create a reduced image containing target objects; identifying at least one target object in the reduced image; and locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image.
Claim 2
Dependent Claim
76 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising the steps of: reducing the size of the sample image to create a reduced image containing target objects; identifying at least one target object in the reduced image; and locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image.
Claim 3
Dependent Claim
32 words
The method of claim 1 , further comprising the step of: extracting at least one portion of the sample image defined by the at least one bounding rectangle in the sample image.
Claim 4
Dependent Claim
24 words
The method of claim 1 , wherein the reducing step comprises: calculating a reduction factor; and mapping the sample image onto the reduced image.
Claim 5
Dependent Claim
134 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising the steps of: reducing the size of the sample image to create a reduced image; identifying at least one target object in the reduced image; and locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image; wherein the reducing step comprises: calculating a reduction factor; and mapping the sample image onto the reduced image; and the calculating step comprises: determining a gap size; and calculating a reduction factor, wherein the reduction factor is defined by the equation: R = 1 G + 1 wherein R is the reduction factor, and G is the gap size in pixels.
Claim 6
Dependent Claim
134 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising the steps of: reducing the size of the sample image to create a reduced image; identifying at least one target object in the reduced image; and locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image; wherein the reducing step comprises: calculating a reduction factor; and mapping the sample image onto the reduced image; and the calculating step comprises: determining a gap size; and calculating a reduction factor, wherein the reduction factor is defined by the equation: R = 1 G + 1 wherein R is the reduction factor, and G is the gap size in pixels.
Claim 7
Dependent Claim
76 words
The method of claim 1 , wherein the reducing step comprises: calculating a reduction factor; and mapping the sample image onto the reduced image, and the mapping step comprises: mapping the sample image onto a grid of boxes, each box corresponding to a pixel of the reduced image; and setting a color of a pixel in the reduced image according to the percentage of pixels of that color in the corresponding box of the sample image.
Claim 8
Dependent Claim
15 words
The method of claim 4 , wherein the gap size is determined from user input.
Claim 9
Dependent Claim
25 words
The method of claim 4 , wherein the gap size is determined by selecting one of a plurality of plausible gap sizes for an image.
Claim 10
Dependent Claim
36 words
The method of claim 5 , wherein the dimensions of the grid are determined by multiplying the reduction factor by the dimensions of the sample image and rounding any fractions to the next highest integer value.
Claim 11
Dependent Claim
26 words
The method of claim 1 , wherein each target object is characterized by a target object definition, each target object definition comprising at least one parameter.
Claim 12
Dependent Claim
29 words
The method of claim 9 , wherein the at least one parameter is taken from the group of parameters comprising size, shape, granularity, texture, intensity of color, and location.
Claim 13
Dependent Claim
59 words
The method of claim 9 , wherein the step of identifying the at least one target object in the reduced image comprises the substeps of: identifying at least one connected component in the reduced image; assigning a bounding rectangle to the at least one connected component; and determining whether the at least one connected component is a target object.
Claim 14
Dependent Claim
88 words
The method of claim 11 , wherein the step of determining whether the at least one connected component is a target object comprises the substeps of: comparing the at least one connected component with at least one target object definition from a target library, the at least one target object definition comprising at least one parameter characterizing a target object; and responsive to the at least one connected component satisfying the at least one target object definition, selecting the at least one connected component as a target object.
Claim 15
Dependent Claim
44 words
The method of claim 1 , wherein the locating step comprises: multiplying a set of coordinates associated with a bounding rectangle of the target object by an enlargement factor, wherein the enlargement factor is the reciprocal of a reduction factor for the reduced image.
Claim 16
Dependent Claim
96 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising: selecting a plurality of gap sizes; and for each selected gap size: reducing the size of the sample image, wherein the sample image is reduced by applying a reduction factor defined by the equation: R = 1 G + 1 wherein R is the reduction factor, and G is the gap size in pixels; identifying each target object in the reduced image; and recording the number of identified target objects for the selected gap size.
Claim 17
Dependent Claim
96 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising: selecting a plurality of gap sizes; and for each selected gap size: reducing the size of the sample image, wherein the sample image is reduced by applying a reduction factor defined by the equation: R = 1 G + 1 wherein R is the reduction factor, and G is the gap size in pixels; identifying each target object in the reduced image; and recording the number of identified target objects for the selected gap size.
Claim 18
Dependent Claim
43 words
The method of claim 14 , wherein the step of identifying each target object in the reduced image comprises: performing connected component analysis on the reduced image to identify a number of connected components; and filtering the identified connected components for target objects.
Claim 19
Dependent Claim
70 words
The method of claim 15 , wherein the filtering step comprises the substeps of: comparing the identified connected components with a number of target object definitions in a target object library, each target object definition comprising at least one parameter characterizing a target object; and responsive to a connected component satisfying at least one target object definition in the target object library, selecting the connected component as a target object.
Claim 20
Dependent Claim
49 words
The method of claim 14 , further comprising the steps of: selecting at least one preferred target object from the target objects; locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one preferred target object in the reduced image.
Claim 21
Dependent Claim
94 words
The method of claim 17 , wherein the step of selecting at least one preferred target object comprises the substeps of: responsive to one reduced image comprising more target objects than any other reduced image, selecting as preferred target objects the target objects found in the one reduced image; responsive to at least two reduced images comprising more target objects than any other reduced image, selecting as preferred target objects the target objects found in the one of the at least two reduced images that is least reduced with respect to the sample image.
Claim 22
Dependent Claim
23 words
The method of claim 18 , further comprising the steps of: discarding the target objects that are not selected as preferred target objects.
Claim 23
Independent Claim
63 words
A system for segmenting a sample image into at least one target object, the system comprising: a size reduction unit for converting the sample image into a reduced image; and coupled to the size reduction unit, a connected component analyzer for identifying at least one connected component in the reduced image and assigning to the at least one connected component a bounding rectangle.
Claim 24
Independent Claim
63 words
A system for segmenting a sample image into at least one target object, the system comprising: a size reduction unit for converting the sample image into a reduced image; and coupled to the size reduction unit, a connected component analyzer for identifying at least one connected component in the reduced image and assigning to the at least one connected component a bounding rectangle.
Claim 25
Dependent Claim
33 words
The system of claim 20 , further comprising: coupled to the connected component analyzer, a target object filter for determining whether the at least one connected component is at least one target object.
Claim 26
Dependent Claim
34 words
The system of claim 21 , further comprising: coupled to the target object filter, a target object library for storing at least one target object definition, each target object definition characterizing a target object.
Claim 27
Dependent Claim
35 words
The system of claim 22 , wherein the at least one target object definition comprises at least one parameter taken from the group of parameters comprising size, shape, granularity, texture, intensity of color, and location.
Claim 28
Dependent Claim
44 words
The system of claim 21 , further comprising: coupled to the target object filter, an inverse mapper for locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image.
Claim 29
Dependent Claim
92 words
A computer-readable medium having computer-readable program code devices embodied therein for segmenting a sample image into at least one target object, the computer-readable medium comprising: computer-readable program code devices configured to reduce the size of the sample image to create a reduced image; computer-readable program code devices configured to identify at least one target object in the reduced image; and computer-readable program code devices configured to locate at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image.
Claim 30
Dependent Claim
92 words
A computer-readable medium having computer-readable program code devices embodied therein for segmenting a sample image into at least one target object, the computer-readable medium comprising: computer-readable program code devices configured to reduce the size of the sample image to create a reduced image; computer-readable program code devices configured to identify at least one target object in the reduced image; and computer-readable program code devices configured to locate at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image.
Claim 31
Dependent Claim
28 words
The method of claim 1 wherein the reducing step comprises calculating a reduction factor according to a gap size; and mapping the sample image onto the reduced image.
Claim 32
Dependent Claim
14 words
The method of claim 1 wherein the reduced image contains only extracted target objects.
Claim 33
Dependent Claim
73 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising the steps of: reducing the size of the sample image to create a reduced image; identifying at least one target object in the reduced image; and locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image.
Claim 34
Dependent Claim
73 words
In an image segmentation system, a method for segmenting a sample image into at least one target object, the method comprising the steps of: reducing the size of the sample image to create a reduced image; identifying at least one target object in the reduced image; and locating at least one bounding rectangle in the sample image corresponding to a bounding rectangle of the at least one target object in the reduced image.
References Cited (Backward Citations)
US5202933A
Cited
US5434953A
Cited
US5680479A
Cited
EP0657838A2
Cited
US5809167A
Cited
US5848185A
Cited
US5903904A
Cited
US5778092A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2006-03-10 | 2006-09-09 | 2007-03-11 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2010-03-10 | 2010-09-09 | 2011-03-11 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2014-03-10 | 2014-09-09 | 2015-03-11 | $7700.00 | Window Closed / Lapsed |