Pattern classification system and method for collective learning
AssigneeADT Services AG
InventorsStewart E. Hall, David M. Salcedo
Life of this patent
Estimating the term…
Abstract
A method for configuring a pattern recognition system begins by receiving object recognition data from at least one first local image processing system. The object recognition data is stored in at least one global database. Configuration data is determined for a second local image processing system based at least in part upon the received object recognition data from the at least one first image processing system, and then transmitted to the second local image processing system.
Claims22
A method for adaptive learning in a pattern recognition system, the method comprising:
- receiving image data and local object recognition data from at least one first local image processing system, the local object recognition data based on a at least one first local classification algorithm;
- storing the object recognition data in at least one global database, the at least one global database associated with a centralized expert system;
- analyzing the image data using at least one expert system classification algorithm;
- determining accuracy of the first local classification algorithm based on analysis by the at least one expert system classification algorithm;
- and
- transmitting an update to the at least one first local classification algorithm to improve the accuracy thereof.
A method for adaptive learning in a pattern recognition system, the method comprising:
- transmitting image data from at least one local image processing system to a central expert system, the image data including a first pattern recognition dataset based upon a first pattern recognition algorithm, the central expert system performing the steps of:
- executing a second pattern recognition algorithm on the received image data to produce a second pattern recognition dataset;
- comparing the first pattern recognition dataset to the second pattern recognition dataset to reveal any discrepancies;
- and
- responsive to discovering discrepancies, transmitting updates to at least one of the first pattern recognition algorithm and a database of the at least one local image processing system.
A method of configuring a local image processing system, the method comprising:
- transmitting a first set of system parameters to a centralized expert pattern recognition system;
- and
- receiving configuration data from the centralized expert pattern recognition system, the configuration data including at least one local pattern recognition algorithm, the configuration data based at least in part on the first set of system parameters and object recognition data collected from at least one other local image processing system, the at least one other local image processing system including a second local pattern recognition algorithm, the central expert system determining the at least one local pattern recognition algorithm based at least in part on the first set of parameters and the second local pattern recognition algorithm.
A pattern recognition system comprising:
- a plurality of local image processing systems, each of the plurality of local image processing systems collecting local image data and producing local object recognition data based on at least one local classification algorithm;
- at least one centralized expert pattern recognition system communicatively coupled to each of the plurality of local image processing systems, the at least one centralized expert pattern recognition system:
- receiving image data and local object recognition data from at least one of the local image processing systems;
- analyzing the local image data using at least one expert system algorithm to produce expert object recognition data;
- comparing the expert object recognition data with the local object recognition data to determine the accuracy of the at least one local classification algorithm;
- and
- transmitting an update to the at least one local classification algorithm to improve the accuracy thereof.
Dependencies are inferred from the claim text.