Features extracted online from froth images on flotation plants are potentially useful to the development of advanced control for flotation systems, provided that these features can be related to the key indicator variables of the plant, such as valuable metal loadings and recovery. Although such relationships have been established in a number of base metal plants in industry, this is not the case in the platinum industry. In this paper, estimation of flotation grades and recoveries from froth image data is therefore considered based on laboratory and industrial plant data. It is shown that grades and recoveries can be reliably estimated from a number of different features by use of linear and nonlinear models. This includes simple colour information that on the industrial plant showed a strong correlation with grade measurement.