The technical article by Atsushi Makata of our Deep Learning Solution Team was published in the January 2020 issue of Image Labo, a technical magazine on image processing technology published by Nihon Kougyou Shuppan Co.

Contents of the article
Appearance Inspection Method by Direct Distance Estimation
Co-author: Chihiro Noguchi, Tokyo Institute of Technology / Atsushi Makata, Araya Inc. / Masao Yamanaka, Toyota Motor Corporation
Many methods based on neural networks have been proposed as appearance inspection methods using machine learning, but these methods have some drawbacks. In this paper, we propose an appearance inspection method based on direct distance estimation as a more lightweight method that complements these methods.

Journal information is here.