Machine vision technology now offers a viable means of monitoring and control of froth flotation systems. In this study the relationship between process conditions and the surface bubble size as well as the process performance in the batch flotation of a copper sulfide ore is discussed and modeled by neural networks. Flotation experiments are conducted at a wide range of process conditions (i.e., gas flow rate, slurry solids %, frother/collector dosage, and pH) and the froth mean bubble size along with the metallurgical parameters are determined for each run. An adaptive marker based watershed algorithm is successfully developed for segmentation of the froth images and measurement of the bubble size at different conditions. The results show that there is a strong correlation between process conditions and the froth mean bubble size, which is of great importance for control purposes. Even though the metallurgical parameters can be estimated from the froth mean bubble size alone, other froth features (i.e., froth velocity, color, and stability) are required to be measured in order to achieve more accurate predictions of the process performance.

In this article, an innovative classification framework for hyperspectral image data, based on both spectral and spatial information, is proposed. The main objective of this method is to improve the accuracy and efficiency of high-resolution land-cover mapping in urban areas. The spatial information is obtained by an enhanced marker-based minimum spanning forest (MMSF) algorithm. A pixel-based support vector machine (SVM) algorithm is first used to classify the hyperspectral image data, then the enhanced MMSF algorithm is applied in order to increase the accuracy of less accurately classified land-cover types. The enhanced MMSF algorithm is used as a binary classifier. These two classes are the low-accuracy class and remaining classes. Finally, the SVM algorithm is trained for classes with acceptable accuracy. In the proposed approach, namely MSF-SVM, the markers are extracted from the classification maps obtained by both SVM and watershed segmentation algorithms, and are then used to build the MSF. Three benchmark hyperspectral data sets are used for the assessment: Berlin, Washington DC Mall, and Quebec City. Experimental results demonstrate the superiority of the proposed approach compared with SVM and the original MMSF algorithms. It achieves approximately 5, 6, and 7% higher rates in kappa coefficients of agreement in comparison with the original MMSF algorithm for the Berlin, Washington DC Mall, and Quebec City data sets, respectively.