[11p-N107-1~16] 23.1 Joint Session N "Informatics"

△ [11p-N107-15] Training data optimization and error analysis
for machine learning-based crystal orientation estimation

〇(M1)Kyoka Hara¹, Takuto Kojima², Kentaro Kutsukake³, Hiroaki Kudo², Noritaka Usami¹ (1.Grad. Eng. Nagoya Univ., 2.Grad. Info. Nagoya Univ., 3.Riken AIP)

Keywords:machine learning, multicrystalline silicon, crystal orientation

We have developed a machine learning-based method for estimating crystal orientations from multiple optical images of textured multicrystalline wafers. We report on further improvement of the prediction accuracy by optimizing training data through statistical analysis of the estimation results in different conditions.