Evaluating fabric fuzziness using wavelet transforms

Fabric fuzziness refers to the severity of hairiness of a fabric caused by untangled fiber ends protruding from the surface, and indicates fabric dete rioration in structure and appearance after repetitive abrasions in service . Although the average height and density of protruding fiber ends are the information that can be directly used to evaluate fuzziness, fiber ends are often too short and too fine to permit reliable measurements of fiber heig hts and density over a large area. In practice, visual examinations are oft en used for fuzziness evaluation despite the subjective nature of the metho d. An instrument that can perform reliable and quick evaluations of fuzzy a ppearance has been long demanded. This paper introduces the application of the wavelet transform to quantifying fabric fuzziness in a customized image analysis system. An index is constructed from the average powers of the co mponents from different frequency bands, and is demonstrated to be indicati ve of fabric fuzziness. (C) 2000 Society of Photo-Optical Instrumentation E ngineers. [S0091-3286(00)00409-8].