We propose an approach for textured image segmentation based on amplitude-modulation frequency-modulation models. An image is modeled as a set of 2-D nonstationary sinusoids with spatially varying amplitudes and spatially varying frequency vectors. First, the demodulation procedure for the models furnishes a high-dimensional output at each pixel. Then, features including texture contrast, scale, and brightness are elaborately selected based on the high-dimensional output and the image itself. Next, a normalization and weighting scheme for feature combination is presented. Finally, simple K-means clustering is utilized for segmentation. The main characteristic of this work provides a feature vector that strengthens useful information and has fewer dimensionalities simultaneously. The proposed approach is compared with the dominant component analysis (DCA)+K-means algorithm and the DCA+ weighted curve evolution algorithm on three different datasets. The experimental results demonstrate that the proposed approach outperforms the others.
Texture segmentation is the process of partitioning an image into regions with different textures containing similar group
of pixels. Most of segmentation algorithms can be regarded as consisting of two successive processes: feature extraction
and feature-based segmentation. In this paper, a new texture segmentation method based on the combination of secondorder
features and spatial information is proposed. Our method has been compared with the Only Second-order
Features(OSF) based algorithm and the Krishnapuram and Freg's (KF) algorithm by segmenting images from the
Brodatz album and a real-life scenery image dataset. The results show that the proposed approach reduces the inaccurate
small regions and keeps the edge of contiguous target regions more smooth. In addition, the parameters of filter bank are
selected elaborately according to human psychophysics research, thus our algorithm is so intuitive and physiological
relevant that it reserves opportunities for further approaches.