System and method for detecting generic items in image sequence
First Claim
1. A merchandise checkout system for detecting a merchandise item from sequentially captured images, the system comprising:
- a feature extractor for extracting visual features from the sequentially captured images, the extracted visual features characterized by feature vectors;
a feature comparison module operable to compare visual features extracted from a first image of the sequentially captured images to visual features extracted from a second image of the sequentially captured images to identify a group of extracted visual features that match between the first and second images;
a classifier operable to measure, for the group of extracted visual features, a classification parameter so as to generate a parameter measurement that collectively characterizes the group of extracted visual features for comparison of the parameter measurement to an associated classification criterion, the associated classification criterion defining a first classification representing presence of merchandise items and a second classification representing presence of non-merchandise items, and the classifier further operable to classify the group of extracted visual features into either the first classification or the second classification based on whether the parameter measurement satisfies the associated classification criterion, wherein the associated classification criterion includes a threshold amount of entropy of an area occupied by the group of extracted visual features; and
an alarm configured to generate an alert in response to the classifier classifying the group of extracted visual features as belonging to the first classification and thereby indicating the presence of the merchandise item.
2 Assignments
0 Petitions
Accused Products
Abstract
A system and method for detecting the presence of known or unknown objects based on visual features is disclosed. In the preferred embodiment, the system is a checkout system for detecting items of merchandise on a shopping cart. The merchandise checkout system preferably includes a feature extractor for extracting visual features from a plurality of images; a motion detector configured to detect one or more groups of the visual features present in at least two of the plurality of images; a classifier to classify each of said groups of the visual features based on one or more classification criteria, wherein each of the one or more parameters is associated with one of said groups of visual features; and an alarm configured to generate an alert if the one or more parameters for any of said groups of the visual features satisfy one or more classification criteria.
-
Citations
26 Claims
-
1. A merchandise checkout system for detecting a merchandise item from sequentially captured images, the system comprising:
-
a feature extractor for extracting visual features from the sequentially captured images, the extracted visual features characterized by feature vectors; a feature comparison module operable to compare visual features extracted from a first image of the sequentially captured images to visual features extracted from a second image of the sequentially captured images to identify a group of extracted visual features that match between the first and second images; a classifier operable to measure, for the group of extracted visual features, a classification parameter so as to generate a parameter measurement that collectively characterizes the group of extracted visual features for comparison of the parameter measurement to an associated classification criterion, the associated classification criterion defining a first classification representing presence of merchandise items and a second classification representing presence of non-merchandise items, and the classifier further operable to classify the group of extracted visual features into either the first classification or the second classification based on whether the parameter measurement satisfies the associated classification criterion, wherein the associated classification criterion includes a threshold amount of entropy of an area occupied by the group of extracted visual features; and an alarm configured to generate an alert in response to the classifier classifying the group of extracted visual features as belonging to the first classification and thereby indicating the presence of the merchandise item. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A method of detecting a merchandise item from sequentially captured images, the method comprising:
-
extracting visual features from the sequentially captured images, the extracted visual features characterized by feature vectors; comparing visual features extracted from a first image of the sequentially captured images to visual features extracted from a second image of the sequentially captured images to identify a group of extracted visual features that match between the first and second images; measuring one or both motion parameters or appearance parameters so as to generate a parameter measurement that collectively characterizes the group of extracted visual features for comparison of the parameter measurement to an associated classification criterion, the associated classification criterion defining a first classification representing presence of merchandise items and a second classification representing presence of non-merchandise items, wherein the associated classification criterion includes a threshold amount of entropy of an area occupied by the group of extracted visual features; classifying the group of extracted visual features into either the first classification or the second classification based on whether the parameter measurement satisfies the associated classification criterion; and generating an alert if the one or both motion parameters or appearance parameters for the group of extracted visual features satisfy the associated classification criterion. - View Dependent Claims (25, 26)
-
Specification