IEEJ Transactions on Electronics, Information and Systems
Online ISSN : 1348-8155
Print ISSN : 0385-4221
ISSN-L : 0385-4221
<Softcomputing, Learning>
How to Create Massive Training Data for Deep Learning without Any Experts Knowledge and Its Application to Artery-Vein Classification for Fundus Blood Vessels Using CNN
Mashiho MukaidaYuki OkamiHiroaki KogaNoriaki SuetakeEiji Uchino
Author information
JOURNAL RESTRICTED ACCESS

2020 Volume 140 Issue 5 Pages 549-550

Details
Abstract

Automatic classification of fundus blood vessels into artery and vein is one of the important topics in fundus image analysis. The conventional method, which employs pixel-wise training data, takes high costs for creating the training data. In this paper, we propose a very simple method to create as many training data as possible enough for deep learning, which only needs clipping an image of small size from the whole blood vessel image without any experts knowledge. The effectiveness of the proposed method has been confirmed.

Content from these authors
© 2020 by the Institute of Electrical Engineers of Japan
Previous article Next article
feedback
Top