Review of wavelet-based unsupervised texture segmentation, advantage of adaptive wavelets

TL;DR

The authors assess the influence of the chosen wavelet and propose to use the recently introduced empirical wavelets and show that the adaptability of the empirical wavelet permits to reach better results than classic wavelets.

Abstract

Wavelet-based segmentation approaches are widely used for texture segmentation purposes because of their ability to characterize different textures. In this paper, we assess the influence of the chosen wavelet and propose to use the recently introduced empirical wavelets. We show that the adaptability of the empirical wavelet permits to reach better results than classic wavelets. In order to focus only on the textural information, we also propose to perform a cartoon + texture decomposition step before applying the segmentation algorithm. The proposed method is tested on six classic benchmarks, based on several popular texture images.

Figures (10)

• Figure 1: General steps in unsupervised texture segmentation.
• Figure 2: Dyadic decimated wavelet transform for $k=2$.
• Figure 3: Dyadic undecimated wavelet transform for $K=2$.
• Figure 4: Magnitude of the Fourier spectrum segmentation and empirical wavelet construction principle.
• Figure 5: Fourier partitions corresponding to the different 2D empirical wavelets.