부산대학교 도서관

A new method for characterizing surface topography using a 2-dimensional discrete wavelet transform (2-D DWT) has been developed. The wavelet transform (WT) is the basis for a wide range of techniques applied in image processing and pattern recognition. Its main advantages over other feature extraction methods are the space-frequency localization, and the multi-resolution view of the frequency components of a signal. In this paper, an automatic grinding modes classification technique using a 2-D DWT is introduced, and comparisons of automatic grinding modes classification and human eye inspection are also examined.