Universal product labeling for vision-based commerce
AssigneeGrabango Co.
InventorWilliam Glaser
Life of this patent
Estimating the term…
Abstract
A system and method for computer vision assisted physical-labels functions to use functional, large format product labeling for better computer vision (CV) identification. The product labels and detection capabilities of the system and method enhance product label detection and interpretation for use with image sensing at a distance. The product labels are non-obtrusive and can mitigate changes to sizing or visual appearance of packaging of goods or products. The system and method can enable universal product labeling for vision-based commerce and other applications.
Claims24
A method for sensor detectable physical labels comprising:
- receiving a product identifier of a product;
- encoding a partial product identifier into a graphical machine-readable code, wherein the partial product identifier is unique for a local classification group and is associated with the product identifier;
- at the imaging devices of the imaging system, collecting image data;
- detecting, by processing the image data using a code detection computer vision model, an imaged product label with an imaged graphic machine readable code;
- decoding the imaged graphic machine-readable code into an imaged partial identifier;
- determining a classification group of the imaged product label;
- and
- compiling a product identifier output from the imaged partial identifier and the classification group of the imaged product label.
A system for product label detection comprising:
- a computer vision monitoring system with a plurality of imaging devices distributed in an environment, the computer vision monitoring system configured to collect image data;
- a label management system comprising one or more computer-readable mediums storing instructions that, when executed by the one or more computer processors, cause a computing platform to perform operations comprising:
- receiving a product identifier of a product;
- encoding a partial product identifier into a graphical machine-readable code, wherein the partial product identifier is unique for a local classification group and is associated with the product identifier;
- detecting, by processing the image data using a code detection computer vision mode, an imaged product label with an imaged graphic machine readable code;
- decoding the imaged graphic machine-readable code into an imaged partial identifier;
- determining a classification group of the imaged product label;
- and
- compiling a product identifier output from the imaged partial identifier and the classification group of the imaged product label.
A non-transitory computer-readable medium storing instructions that, when executed by one or more computer processors of a computing platform, cause the computing platform to:
- receiving a product identifier of a product;
- encoding a partial product identifier into a graphical machine-readable code wherein the partial product identifier is unique for a local classification group and is associated with the product identifier;
- at the imaging devices of the imaging system, collecting image data;
- detecting, by processing the image data using a code detection computer vision model, an imaged product label with an imaged graphic machine readable code;
- decoding the imaged graphic machine-readable code into an imaged partial identifier;
- determining a classification group of the imaged product label;
- and
- compiling a product identifier output from the imaged partial identifier and the classification group of the imaged product label.
Dependencies are inferred from the claim text.