Online Detection System for Ore Particle Size Distribution Based on Deep Learning

Abstract

This study presents a deep learning-based system for the online detection of ore particle size distribution (PSD) to enhance efficiency and enable real-time monitoring in mining operations. Traditional methods, such as sieving and manual sampling, are time-consuming, labor-intensive, and unsuitable for real-time applications. To address these limitations, a system was developed that integrates advanced computer vision techniques, robust hardware components, and intelligent software design. The system captures high-quality images of ore particles using industrial cameras and lighting systems, applies image preprocessing, and employs a deep learning model for real-time detection and classification. Evaluation in a simulated mining environment demonstrated high performance in terms of accuracy, latency, and robustness. The results indicate that the system effectively detects and classifies ore particles, providing real-time feedback on particle size distribution. This solution offers a scalable and efficient alternative to traditional methods, supporting more effective mining operations and improved resource utilization. The research contributes to smart mining technologies by delivering a practical and reliable tool for real-time ore particle size monitoring.