Classifying Fleece Fabric Appearance by Extended Morphological Fractal Analysis

Image analysis techniques have been widely accepted as objective methods for evalu ating fabric appearance. This paper presents the development of a fairly new fractal analysis method (extended morphological fractal analysis) for characterizing polar fleece fabric appearance after abrasion. The digital gray level image is treated as a three- dimensional surface whose fractal dimension is calculated by performing a series of dilations and erosions on this surface and plotting the area of the resulting set of surfaces against the size of the structuring element. In contrast to a single morphological fractal .parameter, which is scale-invariant, extended fractal analysis is able to characterize fabric textures where the roughness of these textures is not necessarily scale-invariant. This approach can be used to physically describe surface roughness and texture regularity with the parameter MFV (multiscale fractal vector) and to classify the appearance grade with the Bayes classification method. Our experimental results demonstrate that a good correlation can be established between estimated grades and subjective grades.