Smartphone-based asset management system
First Claim
Patent Images
1. A computer-implemented method for image processing and object recognition, the method comprising:
- receiving, by a remote image processing server, an image of a plurality of objects and location information corresponding to the image from a mobile communications device via a wireless network;
performing, by an image analyzer of the remote image processing server, segmentation to divide the received image into a plurality of regions, wherein performing segmentation comprises;
generating an edge map from the image;
removing components of the edge map that are smaller than a predetermined threshold size from the edge map;
determining a dominant angle for substantially parallel lines of the edge map; and
identifying the plurality of regions of the image, wherein the plurality of regions of the image correspond to regions of the edge map between lines of the edge map within a predetermined threshold angle of the dominant angle;
for at least one region of the plurality of regions, extracting, by the remote image processing server, features of the region and matching the features of the region with a database entry to determine an object identity of a single object corresponding to the region;
determining, by the remote image processing server, a localized position of the single object within a room based on the received location information and further based on previously acquired training information;
storing, by the remote image processing server, the localized position of the single object with the corresponding object identity in a database;
receiving, by the remote image processing server, a query for an object;
determining;
by the remote image processor server, in response to the query, a localized position for the object based on information stored in the database; and
outputting, by the remote image processor server, the localized position for the object.
2 Assignments
0 Petitions
Accused Products
Abstract
A method of tracking an inventory of objects via a mobile communications device includes acquiring an image of one or more of the objects via the mobile communications device, which also collects a location of the mobile communications device while acquiring the image of the one or more of the objects. The location and image are transferred from the mobile communications device to a remote server via a wireless network, such that the one or more of the objects are identified at the server based on the image, and the location and identity of the one or more objects are stored on a database associated with the server.
-
Citations
16 Claims
-
1. A computer-implemented method for image processing and object recognition, the method comprising:
-
receiving, by a remote image processing server, an image of a plurality of objects and location information corresponding to the image from a mobile communications device via a wireless network; performing, by an image analyzer of the remote image processing server, segmentation to divide the received image into a plurality of regions, wherein performing segmentation comprises; generating an edge map from the image; removing components of the edge map that are smaller than a predetermined threshold size from the edge map; determining a dominant angle for substantially parallel lines of the edge map; and identifying the plurality of regions of the image, wherein the plurality of regions of the image correspond to regions of the edge map between lines of the edge map within a predetermined threshold angle of the dominant angle; for at least one region of the plurality of regions, extracting, by the remote image processing server, features of the region and matching the features of the region with a database entry to determine an object identity of a single object corresponding to the region; determining, by the remote image processing server, a localized position of the single object within a room based on the received location information and further based on previously acquired training information; storing, by the remote image processing server, the localized position of the single object with the corresponding object identity in a database; receiving, by the remote image processing server, a query for an object;
determining;
by the remote image processor server, in response to the query, a localized position for the object based on information stored in the database; andoutputting, by the remote image processor server, the localized position for the object. - View Dependent Claims (2, 3, 4, 14, 15, 16)
-
-
5. A non-transitory computer-readable medium, part of a remote image processing server, having processor-executable instructions stored thereon for image processing and object recognition, the processor-executable instructions, when executed by a processor, causing the following steps to be performed:
-
receiving an image of a plurality of objects and location information corresponding to the image from a mobile communications device via a wireless network; performing segmentation to divide the received image into a plurality of regions, wherein performing segmentation comprises; generating an edge map from the image; removing components of the edge map that are smaller than a predetermined threshold size from the edge map; determining a dominant angle for substantially parallel lines of the edge map; and identifying the plurality of regions of the image, wherein the plurality of regions of the image correspond to regions of the edge map between lines of the edge map within a predetermined threshold angle of the dominant angle; for at least one region of the plurality of regions, extracting features of the region and matching the features of the region with a database entry to determine an object identity of a single object corresponding to the region; determining a localized position of the single object within a room based on the received location information and farther based on previously acquired training information; and storing the localized position of the single object with the corresponding object identity in a database; receiving a query for an object; determining, in response to the query, a localized position for the object based o information stored in the database; and outputting the localized position for the object. - View Dependent Claims (6, 7, 8, 9, 10, 11)
-
-
12. A remote image processing server, the server comprising:
-
a receiver, configured to receive an image of a plurality of objects and insertion information corresponding to the image from a mobile communications devise via a wireless network; a processor, configured to; perform segmentation to divide the received image into a plurality of regions, wherein performing segmentation comprises; generating an edge map from the image; removing components of the edge map that are smaller than a predetermined threshold size from the edge map;
determining a dominant angle for substantially parallel lines of the edge map; andidentifying the plurality of regions of the image, wherein the plurality of regions of the image correspond to regions of the edge map between lines of the edge map within a predetermined threshold angle of the dominant angle;
for at least one region of the plurality of regions extract features of the region and matching the features of the region with a database entry to determine an object identity of a single object corresponding to the region, wherein a Speeded Up Robust Features (SURF) algorithm is used to match the features of the region with the database entry; and
determine a localized position of the single object within a room leased on the received location information and farther based on previously acquired training information; anda database, configured to store the localized position of the single object with the corresponding object identity; wherein the receiver is further configured to receive a query for an object; and wherein the processor is further configured to determine and output the localized position for the object based on information stored in the database. - View Dependent Claims (13)
-
Specification