Some texture images are analyzed and classified by the fuzzy morphological opening. The basic idea is to use a family of multi scaled structuring elements to open the texture image. A spectrum is then formed by calculating the differences among the areas of successive opened images. With different shapes of the structuring elements, a corresponding spectrum can be obtained, and consequently by comparing the sets of the spectra of the textures, proper classification can be made. Both spectral and structural information can be obtained with this technique and the accuracy of the classification can reach 84% with only one training sample
Published in:
Communications, Computers and Signal Processing, 1997. 10 Years PACRIM 1987-1997 - Networking the Pacific Rim. 1997 IEEE Pacific Rim Conference on
(Volume:2
)
Date of Conference: 20-22 Aug 1997