Filtering for texture classification: a comparative study

TLDR

The study examines how different filtering techniques affect texture classification while keeping the local energy function and classification algorithm constant. The authors review major filtering approaches for texture feature extraction and conduct a comparative study. They evaluate a range of filters—including Laws masks, ring/wedge, dyadic Gabor, wavelet transforms, wavelet packets, wavelet frames, quadrature mirror filters, DCT, eigenfilters, optimized Gabor, linear predictors, and optimized FIR filters—computing local energy features and comparing them to co‑occurrence and autoregressive methods. The experiments yield a ranking of the tested approaches, indicating which filters perform best for texture classification.

Abstract

In this paper, we review most major filtering approaches to texture feature extraction and perform a comparative study. Filtering approaches included are Laws masks (1980), ring/wedge filters, dyadic Gabor filter banks, wavelet transforms, wavelet packets and wavelet frames, quadrature mirror filters, discrete cosine transform, eigenfilters, optimized Gabor filters, linear predictors, and optimized finite impulse response filters. The features are computed as the local energy of the filter responses. The effect of the filtering is highlighted, keeping the local energy function and the classification algorithm identical for most approaches. For reference, comparisons with two classical nonfiltering approaches, co-occurrence (statistical) and autoregressive (model based) features, are given. We present a ranking of the tested approaches based on extensive experiments.