Browsing by Subject "Bayesian classification"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Unknown Microscopic image classification via WT-based covariance descriptors using Kullback-Leibler distance(IEEE, 2012) Keskin, Furkan; Çetin, A. Enis; Erşahin, Tülin; Çetin-Atalay, RengülIn this paper, we present a novel method for classification of cancer cell line images using complex wavelet-based region covariance matrix descriptors. Microscopic images containing irregular carcinoma cell patterns are represented by randomly selected subwindows which possibly correspond to foreground pixels. For each subwindow, a new region descriptor utilizing the dual-tree complex wavelet transform coefficients as pixel features is computed. WT as a feature extraction tool is preferred primarily because of its ability to characterize singularities at multiple orientations, which often arise in carcinoma cell lines, and approximate shift invariance property. We propose new dissimilarity measures between covariance matrices based on Kullback-Leibler (KL) divergence and L 2-norm, which turn out to be as successful as the classical KL divergence, but with much less computational complexity. Experimental results demonstrate the effectiveness of the proposed image classification framework. The proposed algorithm outperforms the recently published eigenvalue-based Bayesian classification method. © 2012 IEEE.Item Unknown Modeling urbanization using building patterns(2007) Doǧrusöz, E.; Aksoy, S.Automatic extraction of buildings and modeling of their spatial arrangements provide essential information for urban applications. This paper describes our work on modeling urbanization using spatial building patterns. Building detection is done using Bayesian classification of multi-spectral information. The individual buildings are used as textural primitives, and co-occurrence based spatial domain features and Fourier spectrum-based frequency domain features are used to model their repetitiveness and periodicity at particular orientations. These features are used to classify image neighborhoods as organized (regular) and unorganized (irregular). Experiments with high-resolution Ikonos imagery show that the proposed technique can be used for automatic segmentation of urban scenes and extraction of valuable information about urban growth.