Robust automatic indentation localisation and size approximation for Vickers microindentation hardness indentations

Commonly used algorithms for the automatic measurement of Vickers microindentation hardness testing images do not handle a broad range of real world indentation images well enough to be applicable in industrial environments. In this paper we propose a simple and fast algorithm for a processing step which approximates the location and size of the indentation in the image in a very robust and reliable manner using template matching and multiresolution techniques. The algorithm is evaluated against two image databases of substantial size (150 and 217 images) and compared to established indentation detection and recognition approaches.