Measurement of Vickers Hardness Using Image Processing Techniques. 2nd Report. Application of Image Feature Extraction.

Abstract

We investigated a method for automated measurement of micro Vickers hardness using image processing techniques. A TV image of the indentation from a microscope was input into an image processing system, and a method was developed to measure the size of Vickers indentation. When an indentation is marked on an unclean surface, the difference of gray levels in the material surface region and the indentation region is small in the input image of the specimen, and the boundary of the indentation is not clear in this case. It is not easy to separate the indentation part from the material surface. Since the features of the gray level distribution in the indentation region are different from those at the material surface region, we applied the image feature extraction methods to detect the indentation region. The result measured by this method was similar to the result measured manually. It was concluded that the accuracy of this method is sufficient for practical use.