US2015199386A1PendingUtilityA1
System and Method for Candidate Sorting and Clustering
Est. expiryAug 29, 2031(~5.1 yrs left)· nominal 20-yr term from priority
Inventors:Kevin W. Stokes
G06F 17/30271G06F 17/30705G06F 17/30268G06F 7/24G06F 16/51G06F 16/56G06V 30/418G06F 16/35G06F 16/5866G06F 16/5838
46
PatentIndex Score
0
Cited by
0
References
0
Claims
Abstract
A system for organizing a plurality of candidates based on the relative similarity of a first candidate with respect to the remaining plurality of candidates is disclosed. The system includes a controller in communication with a storage device configured to receive and accessibly store a generated plurality of candidate images. The controller operable to analyze each of the plurality of candidate images to determined a numeric thumbnail based on a number of identified features in each of a plurality of grid elements of an array and a sum total number of all identified features in the array, calculate a similarity score between one of the plurality of determined numeric thumbnails and each of the remaining plurality of determined numeric thumbnails; and generate a logical group image order as a function of the highest similarity score between the one of the plurality of determined numeric thumbnails and each of the remaining plurality of determined numeric thumbnails.
Claims
exact text as granted — not AI-modifiedWhat is claimed is:
1 . A system for organizing a plurality of candidate images comprising:
a controller operable to sub-divide each candidate image into a plurality of grid elements of an array and determine a corresponding numeric thumbnail for the candidate image, the corresponding numeric thumbnail being based on a count of identified features found in each of the plurality of grid elements and a sum total number of all identified features found in all of the plurality of grid elements, calculate a similarity score between each of the plurality of determined numeric thumbnails and the remaining plurality of determined numeric thumbnails, and generate an ordering of the plurality of candidate images as a function of the highest similarity score between each of the plurality of determined numeric thumbnails and the remaining plurality of determined numeric thumbnails.
2 . The system of claim 1 , wherein the similarity score includes a penalty factor which compensates for size of the grid elements and the count of identified features found in each thereof.
3 . The system of claim 1 , wherein the controller is configured to group similarity scores within the logical group image order to define a cluster further defined by a cluster threshold.
4 . The system of claim 1 , wherein the controller is configured to calculate the similarity score in an iterative manner and remove the numeric thumbnail having the highest similarity score from the plurality of determined numerical thumbnails once it has been ordered, wherein the removed numeric thumbnail is a starting point for the next iterative calculation of the similarity score.
5 . The system of claim 1 , wherein the similarity score is defined as:
SS
=
∑
i
=
1
k
x
i
2
+
[
∑
i
=
1
k
(
x
i
n
k
)
]
2
-
∑
i
=
1
k
x
i
-
y
i
2
-
[
∑
i
=
1
k
(
x
i
n
k
)
-
∑
i
=
1
k
(
y
i
n
k
)
]
2
∑
i
=
1
k
x
i
2
+
[
∑
i
=
1
k
(
x
i
n
k
)
]
2
+
1
where:
k is a total number of grid elements being evaluated;
i is a current number of the grid elements being evaluated;
x i is a grid value associated with the one of the plurality of determined numeric thumbnails;
y i is a grid value associated with the one of the remaining plurality of determined numeric thumbnails; and
n k is an empirical tuning factor.
6 . The system of claim 1 , wherein the similarity score is defined as:
SS
=
∑
i
=
1
k
max
(
x
i
,
y
i
)
-
∑
i
=
1
k
x
i
-
y
i
∑
i
=
1
k
max
(
x
i
,
y
i
)
+
1
where:
k is a total number of grid elements being evaluated;
i is a current number of the grid elements being evaluated;
x i is a grid value associated with the one of the plurality of determined numeric thumbnails; and
y i is a grid value associated with the one of the remaining plurality of determined numeric thumbnails.
7 . A method of organizing a plurality of candidate images, the method comprising:
sub-dividing each of the plurality of candidate images into a plurality of grid elements and determining a numeric thumbnail, wherein the numeric thumbnail is based on grid values corresponding to each of the of the plurality of grid elements and a grid total of all of the grid values of the plurality of grid elements; and calculating a similarity score based the determined grid values corresponding to each of the plurality of candidate images and the determined grid values corresponding to each of the remaining of the plurality of candidate images.
8 . The method of claim 7 further comprising:
ordering the plurality of candidate images as a function of the highest similarity score determined between the one of the plurality of candidate images and each of the remaining of the plurality of candidate images.
9 . The method of claim 8 further comprising:
storing the ordered plurality of candidate images in an accessible memory device.
10 . The method of claim 7 , wherein the determining of the numeric thumbnail further comprises determining a number of non-white pixels within each grid element.
11 . The method of claim 7 , wherein the grid is a multi-dimensional array of grid elements arranged for processing color components of a color model.
12 . The method of claim 7 , wherein the sub-dividing further comprises dividing the candidate images into a two-dimensional array of grid elements.
13 . The method of claim 7 , wherein the similarity score is computed by a computation of a first squared sum of the grid values associated with the one of the plurality of determined numeric thumbnails, and then a computation of a second squared sum of the difference between the grid values associated with the one of the plurality of determined numeric thumbnails and one of the remaining plurality of determined numeric thumbnails, wherein the difference between the first second squared sums is then divided by a sum of one plus the squared sum of the grid values associated with the one of the plurality of determined numeric thumbnails.
14 . The system of claim 7 , wherein the similarity score is computed by a computation of a first sum of maximum grid values from each of the plurality of determined numeric thumbnails being compared, and then computation of a second sum of the absolute value of the difference between the grid values associated with the one of the plurality of determined numeric thumbnails and one of the remaining plurality of determined numeric thumbnails, wherein the difference between the first and second sums is then divided by a sum of one plus the sum of maximum grid values from each of the plurality of determined numeric thumbnails being compared.
15 . The method of claim 7 further comprising:
grouping similarity scores within the logical group image order to define a cluster further defined as a function of a cluster threshold.
16 . The method of claim 7 , wherein calculating the similarity score includes calculating the similarity score in an iterative manner, removing the candidate image having the highest similarity score from the remaining of the plurality of candidate images once it has been placed in logical group image order, and initiating the iterative calculation of the similarity score using the removed candidate image as a starting point.
17 . A method of organizing a plurality of candidate images, the method comprising:
analyzing each of a plurality of candidate images to determine a similarity score, wherein the plurality of candidates images includes an unknown number of candidate clusters, the analyzing comprising defining a numeric thumbnail for each of the plurality of candidate images based on grid values determined by sub-dividing the candidate image into two or more grid elements and determining the grid values corresponding to each portion of the candidate image contained within each grid element, and a grid total of all of the determined grid values for the candidate image; and organizing each of a plurality of candidate images into a number of clusters based on the similarity score, wherein the number of clusters is unknown prior to organizing.
18 . The method of claim 17 , wherein the grid values represent a number of identified features found in each of the plurality of grid elements.
19 . The method of claim 18 further comprising:
calculating the similarity score based the determined grid values corresponding to one of the plurality of candidate images and the determined grid values corresponding to each of the remaining of the plurality of candidate images; and
generating a logical group image order as a function of the highest similarity score determined between the one of the plurality of candidate images and each of the remaining of the plurality of candidate images.
20 . The method of claim 18 , wherein the grid values represent a number of non-white pixels found within each of the plurality of grid elements.Join the waitlist — get patent alerts
Track US2015199386A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.