Systems, methods, and devices for image matching and object recognition in images
AssigneeSlyce Acquisition Inc.
InventorsPhilip B. Romanik, Neil Lawrence Mayle
Life of this patent
Estimating the term…
Abstract
An image matching technique locates feature points in a template image such as a logo and then does the same in a test image. Feature points from the template image are then matched to the feature points in the test image. An additional matching technique boosts the number of points that match each other. The additional points improve the match quality and help discriminate true from false positive matches.
Claims20
A computer-implemented method for determining whether a portion of a first image is located within a second image, the method, implemented by hardware in combination with software, the method comprising:
- (A) determining a set of candidate matching feature points from a first set of feature points associated with said first image and a second set of feature points associated with said second image;
- (B) selectively producing a transformed image by transforming one of:
- (i) said first image into a coordinate system of said second image, and (ii) said second image into a coordinate system of said first image, said transforming being based on a relationship of said candidate matching feature points;
- and
- (C) attempting to boost said set candidate matching feature points to include additional candidate feature points associated with said first image and said second image, said attempting being based on (i) said candidate matching feature points and (ii) said transformed image;
- and then (D) determining whether said portion of said first image is located within said second image based on a size of said set candidate matching feature points.
A computer-implemented method for determining whether a portion of a first image is located within a second image, the method, implemented by hardware in combination with software, the method comprising:
- (A) determining a set of candidate matching feature points from a first set of feature points associated with said first image and a second set of feature points associated with said second image;
- (B) selectively producing a transformed image by transforming one of:
- (i) said first image into a first different coordinate system, and (ii) said second image into a second different coordinate, said transforming being based on a relationship of said candidate matching feature points;
- and
- (C) attempting to boost said set candidate matching feature points to include additional candidate feature points associated with said first image and said second image, said attempting being based on (i) said candidate matching feature points and (ii) said transformed image;
- and then
- (D) determining whether said portion of said first image is located within said second image based on a size of said set candidate matching feature points.
A computer-implemented method for determining whether a portion of a first image is located within a second image, the method, implemented by hardware in combination with software, the method comprising:
- (A) determining a set of candidate matching feature points from a first set of feature points associated with said first image and a second set of feature points associated with said second image;
- (B) selectively transforming one of:
- (i) said set of candidate matching feature points into a first different coordinate system, and (ii) said second set of feature points into a second different coordinate, said transforming being based on a relationship of said candidate matching feature points;
- and
- (C) attempting to boost said set candidate matching feature points to include additional candidate feature points associated with said first image and said second image, said attempting being based on (i) said candidate matching feature points and (ii) said transformed image;
- and then
- (D) determining whether said portion of said first image is located within said second image based on a size of said set candidate matching feature points.
Dependencies are inferred from the claim text.