METHOD AND APPARATUS FOR CLASSIFICATING IMAGE-DATA USING TENSOR-BASED DATA GENERATION MODEL

METHOD AND APPARATUS FOR CLASSIFICATING IMAGE-DATA USING TENSOR-BASED DATA GENERATION MODEL

PURPOSE: A video data classifying method using a tensor-based data generation model and an apparatus thereof are provided to extract critical information required for the classification by using low level tensor data accurately and stably, thereby increasing the reliability of a classification resul...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: PARK, HYE YOUNG, YOON, DONG WOO
Format: Patent
Sprache:eng ; kor
Schlagworte:
CALCULATING
COMPUTING
COUNTING
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
INFORMATION STORAGE
INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORDCARRIER AND TRANSDUCER
PHYSICS
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:PURPOSE: A video data classifying method using a tensor-based data generation model and an apparatus thereof are provided to extract critical information required for the classification by using low level tensor data accurately and stably, thereby increasing the reliability of a classification result. CONSTITUTION: A video data classifying method using a tensor-based data generation model comprises the following steps of: converting inputted video data into low level tensor data when the video data to be classified is inputted(S120); calculating the similarity by substituting the low level tensor data for a defined similarity function proportionally to the probability density of the low level tensor data difference between learning videos(S140); and classifying video data by using the calculated similarity.