Combining Segmentation and Edge Detection for Efficient Ore Grain Detection in an Electromagnetic Mill Classification System

Abstract

This paper presents a machine vision method for detection and classification of copper ore grains. We proposed a new method that combines both seeded regions growing segmentation and edge detection, where region growing is limited only to grain boundaries. First, a 2D Fast Fourier Transform (2DFFT) and Gray-Level Co-occurrence Matrix (GLCM) are calculated to improve the detection results and processing time by eliminating poor quality samples. Next, detection of copper ore grains is performed, based on region growing, improved by the first and second derivatives with a modified Niblack’s theory and a threshold selection method. Finally, all the detected grains are characterized by a set of shape features, which are used to classify the grains into separate fractions. The efficiency of the algorithm was evaluated with real copper ore samples of known granularity. The proposed method generates information on different granularity fractions at a time with a number of grain shape features.

Measurement-Based Modelling of Material Moisture and Particle Classification for Control of Copper Ore Dry Grinding Process

“…The above size distribution was obtained using manual sieve analysis. Currently researched machine vision techniques for assessment of particle size distribution [ 21 , 22 ] may radically shorten the duration of this measurement.…”

“…Color and texture can also be useful, especially when considering the active surface of metal ore. This is the case in a method based on proper lighting and aperture control to extract metal surfaces on ore particles, developed in [ 21 , 22 ]. Firstly, paper [ 21 ] investigated machine-vision-based method of particle detection and classification in electromagnetic mill system for a wide range of particle sizes, shapes and positions in the prepared sample.…”

“…Numerous particle shape features were investigated with the final selection of Heywood, compactness and aspect ratio factors. Then, in [ 22 ] the authors improved the method of material quality checking based on a cascade of two-dimensional Fast Fourier Transform (2DFFT) and Gray Level Co-occurrence Matrix (GLCM), combined Seeded Region Growing (SRG) with boundaries information from edge detection, and proposed a modified Niblack homogeneity algorithm. Moreover, the development of the proposed method will be concentrated on detecting the active surface of copper ore, and on determining the relationship between material moisture and granularity.…”

Moisture Determination for Fine-Sized Copper Ore by Computer Vision and Thermovision Methods

“…Secondly, the authors have experience in using vision methods to assess the degree of grinding of copper ore [ 23 , 24 ]. The publications used segmentation algorithms with modification for the incorrect detection of particles.…”

“…The publications used segmentation algorithms with modification for the incorrect detection of particles. In [ 24 ], a number of modifications were applied to combine two classical approaches to the detection of copper ore particles, namely regions with growing segmentation and edge detection, where region growth was limited only to the particles boundaries. The proposed method showed that the solution based on image analysis can be used to test online materials.…”