/Reducing A Search Space For Item Identification Using Machine Learning
Abstract

An item tracking system includes a memory storing an encoded vector library comprising a plurality of encoded vectors, each encoded vector comprising a plurality of values associated with at least one attribute and associated with an item identifier for a first object; and a processor operably coupled to the memory. The processor is configured to receive a first encoded vector for a second object, remove one or more encoded vectors from the encoded vector library, compare the first encoded vector to encoded vectors in the encoded vector library, identify a second encoded vector that most closely matches the first encoded vector based on the plurality of values, identify a first item identifier associated with the second encoded vector, and output the first item identifier.

Full Text

What is claimed is:

An item tracking system includes a memory storing an encoded vector library comprising a plurality of encoded vectors, each encoded vector comprising a plurality of values associated with at least one attribute and associated with an item identifier for a first object; and a processor operably coupled to the memory. The processor is configured to receive a first encoded vector for a second object, remove one or more encoded vectors from the encoded vector library, compare the first encoded vector to encoded vectors in the encoded vector library, identify a second encoded vector that most closely matches the first encoded vector based on the plurality of values, identify a first item identifier associated with the second encoded vector, and output the first item identifier.
Timeline
Filed
06/03/2026
Published
09/24/2026
Granted
Not Available
IPC Codes(8)
G06V 10/56:relating to colour
G01G 19/414:using electronic computing means only
G06F 18/21:Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation