Deep convolutional neural network for mill feed size characterization

Loading...
Thumbnail Image

Date

Authors

Olivier, Laurentz Eugene
Maritz, M.G. (Michael)
Craig, Ian Keith

Journal Title

Journal ISSN

Volume Title

Publisher

International Federation of Automatic Control

Abstract

Knowing the characteristics of the feed ore size is an important consideration for operations and control of a run-of-mine ore milling circuit. Large feed ore variations are important to detect as they require intervention, whether it be manual by the operator or by an automatic controller. A deep convolutional neural network is used in this work to classify the feed ore images into one of four classes. A VGG16 architecture is used and the classifier is trained making use of transfer learning.

Description

Keywords

Deep learning, Convolutional neural network, Transfer learning, Milling, Run-of-mine ore

Sustainable Development Goals

Citation

Olivier, L.E., Maritz, M.G. & Craig, I.K. 2019, 'Deep convolutional neural network for mill feed size characterization', IFAC-PapersOnLine, vol. 54, no. 14, pp. 105-110.