Minimizing profile error when estimating the sieve-size distribution of iron ore pellets using ordinal logistic regression

Size measurement of pellets based on image analysis techniques would allow non-invasive, frequent and consistent measurement. We evaluate the ability of commonly used size and shape measurement methods to discriminate among different sieve-size classes of pellets. Backward elimination and forward selection of features are used to select two feature sets that are statistically significant for discriminating among different sieve-size classes.