Medical Imaging Technology
Online ISSN : 2185-3193
Print ISSN : 0288-450X
ISSN-L : 0288-450X
Main Topics / Towards Explainable Artificial Intelligence
Training Convolutional Neural Networks with Natural Principle
Hirokatsu KATAOKARyosuke YAMADAAsato MATSUMOTO
Author information
JOURNALS RESTRICTED ACCESS

2021 Volume 39 Issue 3 Pages 117-123

Details
Abstract

The paper introduces a framework for Formula-Driven Supervised Learning (FDSL) which automatically generates image patterns and their image labels for creating a large-scale image dataset. We mainly focus on Fractal DataBase (FractalDB) which consists of fractal geometry existing around our real world. That is, one of the important natural principle enables to train convolutional neural networks without any natural images taken by a camera and human-annotated labels in a pre-training phase.

Information related to the author
© 2021 The Japanese Society of Medical Imaging Technology
Previous article Next article
feedback
Top