Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
720814 | IFAC Proceedings Volumes | 2009 | 6 Pages |
Abstract
The visible and near infrared reflectance spectra of rock samples from a zinc-copper mine are analyzed by data-based multivariate methods. The effect of data preprocessing to the feature extraction of the spectra is discussed. Based on the measurement data, a classification model is developed to automatically separate gangue rocks from ore. The model is detected to classify rock samples with good accuracy and robustness.
Related Topics
Physical Sciences and Engineering
Engineering
Computational Mechanics