Stone quarrying and processing represents a multibillion-dollar industry, with tens of millions of tons of stone being annually produced worldwide. Production of ornamental stone rose to 47.4 million tons in 1998, representing an increased output of 117% in only twelve years. Globalisation of the stone market has lead to increased competition and more rigorous product quality and economic requirements. In order to be competitive it is increasingly necessary to find ways to minimise production costs. One way of achieving this is to reduce waste, since currently in excess of 30% of quarried stone is lost during processing. Stone processing is a relatively complex process, involving around eighteen separate stages. At the same time, stone surfaces contain a mixture of complex 2D patterns and 3D topographic geometry. The challenge for automated inspection of polished stone is to develop systems able to reliably identify 3D topographic defects (either naturally occurring or resulting from polishing), in the presence of concomitant complex 2D stochastic colour patterns. The resulting real-time analysis of the defects may be used in adaptive process control, in order to avoid the wasteful production of defective product. The utilisation of high technology machine vision for quality control and inspection can help to achieve this. An innovative approach to surface inspection is under development at UWE. The technique is based upon an adaptation of the photometric stereo method, and is able to isolate and characterise mixed 2D and 3D surface features. The method is object-centred, is insensitive to variations in object pose, and able to undertake tasks considered beyond the capabilities of existing surface inspection techniques. The approach has been successfully applied to real stone samples, and a selection of experimental results are presented.
