LEARNING DATA GENERATION METHOD FOR CLASSIFIER LEARNING HAVING REGIONAL FEATURES, AND SYSTEM THEREOF

To provide a learning data generation method for classifier learning having regional features, and a system thereof.SOLUTION: A learning data generation method includes: a step of removing a partial area from one sample image of two sample images, and thereafter mixing the two sample images by a met...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	HEO YOUNGUN, CHUN SANG HYUK, YUN SANG DOO, HAN DONGYOON
Format:	Patent
Sprache:	eng ; jpn
Schlagworte:	CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING IMAGE DATA PROCESSING OR GENERATION, IN GENERAL PHYSICS
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	To provide a learning data generation method for classifier learning having regional features, and a system thereof.SOLUTION: A learning data generation method includes: a step of removing a partial area from one sample image of two sample images, and thereafter mixing the two sample images by a method of replacing the removed area with a patch of the other sample image to generate a new image; and a step of learning a CNN (convolutional neural network) model by using the generated image as learning data.SELECTED DRAWING: Figure 3 【課題】地域的特徴を有する分類器学習のための学習データ生成方法およびそのシステムを提供する。【解決手段】学習データ生成方法は、２つのサンプルイメージのうちの１つのサンプルイメージから一部の領域を取り除いた後、取り除かれた領域を他の１つのサンプルイメージのパッチと入れ替える方式によって前記２つのサンプルイメージを混合して新たなイメージを生成する段階、および前記生成されたイメージを学習データとして使用してＣＮＮ（ｃｏｎｖｏｌｕｔｉｏｎａｌｎｅｕｒａｌｎｅｔｗｏｒｋ）モデルを学習する段階を含む。【選択図】図３