A New Weakly Supervised Learning Approach for Real-time Iron Ore Feed Load Estimation

TL;DR

This paper introduces a weakly supervised deep learning approach for real-time estimation of iron ore feed load from pellet images, addressing data annotation challenges and enabling process optimization.

Contribution

It proposes a novel two-stage training algorithm and neural network architectures for weakly supervised ore feed load estimation from images.

Findings

Achieved competitive accuracy in feed load estimation.

Enabled real-time application for process optimization.

Addressed data annotation challenges in mineral processing.

Abstract

Iron ore feed load control is one of the most critical settings in a mineral grinding process, directly impacting the quality of final products. The setting of the feed load is mainly determined by the characteristics of the ore pellets. However, the characterisation of ore is challenging to acquire in many production environments, leading to poor feed load settings and inefficient production processes. This paper presents our work using deep learning models for direct ore feed load estimation from ore pellet images. To address the challenges caused by the large size of a full ore pellets image and the shortage of accurately annotated data, we treat the whole modelling process as a weakly supervised learning problem. A two-stage model training algorithm and two neural network architectures are proposed. The experiment results show competitive model performance, and the trained models can be used for real-time feed load estimation for grind process optimisation.