Object recognition for exception handling in automatic machine-readable symbol reader systems
First Claim
1. A method of operation for an automated machine-readable symbol reader system, the method comprising:
- receiving a plurality of images of an exception object in at least one processor-readable medium, the exception object having not been successfully identified by a machine-readable symbol reader of the machine-readable symbol reader system;
extracting, by at least one processor, one or more visual features from each of the plurality of received images;
accessing a database of visual model data associated with a plurality of known objects;
comparing, by the at least one processor, the extracted one or more visual features of each of the plurality of images with the visual model data associated with the plurality of known objects;
determining, by the at least one processor, a descriptive measure for each of the plurality of images based on the comparison of the extracted one or more visual features with the visual model data; and
sequentially displaying on a display device two or more of the plurality of images in an order dependent on the determined descriptive measures.
1 Assignment
0 Petitions
Accused Products
Abstract
Systems and methods for exception handling in an automated machine-readable symbol reader system having a machine-readable symbol reader that captures machine-readable symbols within a view volume. One or more image capture devices obtain a plurality of images of an exception object in response to an exception generated in the view volume. A processor receives the images, identities visual object recognition features from each image, and compares the features to determine one or more descriptive measures indicative of a likelihood that an operator (e.g., store employee) will be able to identify the exception object by viewing the image. The processor displays at least one of the images (e.g., the most descriptive image) on a display device according to the descriptive measure so that an operator can rapidly identify the identity of the exception object and take steps to resolve the exception.
-
Citations
35 Claims
-
1. A method of operation for an automated machine-readable symbol reader system, the method comprising:
-
receiving a plurality of images of an exception object in at least one processor-readable medium, the exception object having not been successfully identified by a machine-readable symbol reader of the machine-readable symbol reader system; extracting, by at least one processor, one or more visual features from each of the plurality of received images; accessing a database of visual model data associated with a plurality of known objects; comparing, by the at least one processor, the extracted one or more visual features of each of the plurality of images with the visual model data associated with the plurality of known objects; determining, by the at least one processor, a descriptive measure for each of the plurality of images based on the comparison of the extracted one or more visual features with the visual model data; and sequentially displaying on a display device two or more of the plurality of images in an order dependent on the determined descriptive measures. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14)
-
-
15. A method of operation for an automated machine-readable symbol reader system, the method comprising:
-
receiving a plurality of images of an exception object in at least one processor-readable medium, the exception object having not been successfully identified by a machine-readable symbol reader of the machine-readable symbol reader system; extracting, by at least one processor, one or more visual features from each of the plurality of received images; accessing a database of visual model data associated with a plurality of known objects; identifying matches between the extracted visual features and the visual model data associated with the plurality of known objects; determining a descriptive measure ranking for each of the plurality of images based on the number of identified matches; and sequentially displaying on a display device at least two of the plurality of images in an order dependent on the respective ranking of each of the at least two of the plurality of images.
-
-
16. An automated machine-readable symbol reader system, comprising:
-
a machine-readable symbol reader that reads machine-readable symbols associated with objects transported through a view volume; an exception identification subsystem that detects an exception object that has been transported into or through the view volume without being successfully identified by the machine-readable symbol reader; an exception handling subsystem that includes; a display device; a database of stored visual features data of known objects; at least one image capture device that obtains a plurality of images of the exception object; and at least one processor operatively coupled to the display device, the at least one database and the at least one image capture device, the at least one processor; extracts a plurality of visual features from each of the plurality of images of the exception object; identifies matches between the extracted visual features and the stored visual features data of known objects; determines a descriptive measure for each of the plurality of images based on the number of identified matches; and causes two or more of the plurality of images of the exception object to be displayed sequentially on the display device in an order dependent on the determined descriptive measures. - View Dependent Claims (17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30)
-
-
31. An automated machine-readable symbol reader system, comprising:
-
a machine-readable symbol reader that reads machine-readable symbols associated with objects transported through a view volume; an exception identification subsystem that detects an exception object that has been transported into or through the view volume without being successfully identified by the machine-readable symbol reader; an exception handling subsystem that includes; a display device; a database of stored visual features data of known objects; at least one image capture device that obtains a plurality of images of the exception object; and at least one processor operatively coupled to the display device, the at least one database and the at least one image capture device, the at least one processor; extracts a plurality of visual features from each of the plurality of images of the exception object; identifies matches between the extracted visual features and the stored visual features data of known objects to determine whether a logo is present or absent in each of the plurality of images; determines a descriptive measure for each of the plurality of images based on the determined presence or absence of a logo in each of the plurality of images; and causes at least one of the plurality of images of the exception object to be displayed on the display device based on the determined descriptive measure. - View Dependent Claims (32, 33, 34, 35)
-
Specification