FEATURE QUANTITY DETERMINATION METHOD, LEARNING DATA GENERATION METHOD, LEARNING DATA SET, EVALUATION SYSTEM, PROGRAM, AND LEARNING METHOD

• Main Authors: HAYAMA MASAAKI, MAMIYA YASUHIRO, SOFUE YASUYUKI, TANAKA HIROYUKI
• Format: Patent
• Language: English; Japanese
• Published: 30.07.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; PHYSICS; TESTING

To provide a feature quantity determination method, a learning data generation method, a learning data set, an evaluation system, a program, and a learning method which allow improvement in accuracy of evaluation of a sample.SOLUTION: The feature quantity determination method determines one or more second calculation methods for obtaining one or more effective feature quantities for use in evaluation of a sample, from a plurality of first calculation methods on the basis of respective contributions to evaluation of the sample of a plurality of feature quantities which are calculated from a part of interest of an image for evaluation of the sample by the plurality of first calculation methods.SELECTED DRAWING: Figure 1 【課題】課題は、試料の評価の精度を向上できる、特徴量決定方法、学習用データ生成方法、学習用データセット、評価システム、プログラム、及び、学習方法を提供することである。【解決手段】特徴量決定方法は、試料の評価用の画像の着目部分から複数の第１算出方法によりそれぞれ算出される複数の特徴量それぞれの前記試料の評価への寄与度に基づいて、前記複数の第１算出方法から、前記試料の評価に利用する１以上の有効特徴量を得るための１以上の第２算出方法を決定する。【選択図】 図１