Detecting objects of interest in still images
First Claim
1. A computer implemented method for detecting interest sections in a still image, comprising:
- segmenting one or more sub images of said still image based on pixels of interest in one or more color spaces, said one or more sub images being categorized into one or more of interest sub images and a plurality of non interest sub images based on said segmentation;
matching one of a gray scale version of said interest sub images and a binary image version of said interest sub images with a predefined template for filtering said interest sub images based on relative densities of said pixels of interest and relative average intensities of predetermined features in said interest sub images;
determining a plurality of prospective image sections comprising one or more of a plurality of prospective interest sections and a plurality of prospective near interest sections by performing discriminative feature analyses of said filtered interest sub images, wherein said discriminative feature analyses are processed by a boosted cascade of classifiers; and
detecting said interest sections in said still image from said prospective interest sections and said prospective near interest sections by said boosted cascade of classifiers.
2 Assignments
0 Petitions
Accused Products
Abstract
A computer implemented method and system for detecting interest sections in a still image are provided. One or more sub images of the still image is subjected to segmentation. A gray scale version of interest sub images and/or a binary image version of the interest sub images are matched with a predefined template for filtering the interest sub images. Multiple prospective image sections comprising one or more of prospective interest sections and prospective near interest sections are determined by performing discriminative feature analyses of the filtered interest sub images using a gabor feature filter. The discriminative feature analyses are processed by a boosted cascade of classifiers. The boosted cascade of classifiers detects the interest sections in the still image from the prospective interest sections and the prospective near interest sections. The detected interest sections are subjected to a support vector machine classifier for further detecting interest sections.
-
Citations
25 Claims
-
1. A computer implemented method for detecting interest sections in a still image, comprising:
-
segmenting one or more sub images of said still image based on pixels of interest in one or more color spaces, said one or more sub images being categorized into one or more of interest sub images and a plurality of non interest sub images based on said segmentation; matching one of a gray scale version of said interest sub images and a binary image version of said interest sub images with a predefined template for filtering said interest sub images based on relative densities of said pixels of interest and relative average intensities of predetermined features in said interest sub images; determining a plurality of prospective image sections comprising one or more of a plurality of prospective interest sections and a plurality of prospective near interest sections by performing discriminative feature analyses of said filtered interest sub images, wherein said discriminative feature analyses are processed by a boosted cascade of classifiers; and detecting said interest sections in said still image from said prospective interest sections and said prospective near interest sections by said boosted cascade of classifiers. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A computer implemented system for detecting interest sections in a still image, comprising:
-
an image segmentation module that segments one or more sub images of said still image based on pixels of interest in one or more color spaces, said one or more sub images being categorized into one or more of interest sub images and a plurality of non interest sub images based on said segmentation; a template matching engine that matches one of a gray scale version of said interest sub images and a binary image version of said interest sub images with a predefined template for filtering said interest sub images based on relative densities of said pixels of interest and relative average intensities of predetermined features in said interest sub images; a feature analyses module that determines a plurality of prospective image sections comprising one or more of a plurality of prospective interest sections and a plurality of prospective near interest sections by performing discriminative feature analyses of said filtered interest sub images using a gabor feature filter; and a boosted cascade of classifiers that detects said interest sections in said still image from said prospective interest sections and prospective near interest sections. - View Dependent Claims (18, 19, 20, 21, 22)
-
-
23. A computer program product comprising computer executable instructions embodied in a non-transitory computer readable storage medium, wherein said computer program product comprises:
-
a first computer parsable program code for segmenting one or more sub images of said still image based on pixels of interest in one or more color spaces, said one or more sub images being categorized into one or more of interest sub images and a plurality of non interest sub images based on said segmentation; a second computer parsable program code for matching one of a gray scale version of said interest sub images and a binary image version of said interest sub images with a predefined template for filtering said interest sub images based on relative densities of said pixels of interest and relative average intensities of predetermined features in said interest sub images; a third computer parsable program code for determining a plurality of prospective image sections comprising one or more of a plurality of prospective interest sections and a plurality of prospective near interest sections by performing discriminative feature analyses of said filtered interest sub images using a gabor feature filter, wherein said discriminative feature analyses are processed by a boosted cascade of classifiers; and a fourth computer parsable program code for detecting interest sections in said still image from said prospective interest sections and said prospective near interest sections by said boosted cascade of classifiers. - View Dependent Claims (24, 25)
-
Specification