Conference Paper

# Determining multiscale image feature angles from complex wavelet phases

DOI: 10.1007/11559573_61 Conference: Proceedings

Source: DBLP

- [Show abstract] [Hide abstract]

**ABSTRACT:**In most of complex wavelet based fusion methods, only magnitude (or absolute value) of a complex coefficient is considered and phase information is neglected. However, more salient image features can be determined by the phase. In this paper, a multimodality image fusion algorithm is proposed with the shiftable complex directional pyramid transform (SCDPT), where phase and magnitudes of complex coefficients are jointly considered. Firstly, a novel similarity index (CCC-EM) is presented by combining the circular correlation coefficient (CCC) of relative phase angles and the traditional energy matching (EM) index. When bandpass directional subband coefficients are merged, the CCC-EM index is employed as the similarity measure and three types of regions between source images are determined for each bandpass directional subband. Then, based on some weights or salience measures, different fusion rules are designed for each type of regions. Especially, for regions with similarity in energy and positive or negative correlation relationship in relative phase, the weighted circular variance (WCV) of relative phase angles is employed. When lowpass subband coefficients are merged, the traditional structural similarity index is employed to distinguish different types of regions. For most of regions, the local energy of lowpass subband coefficients is employed as weights or salience measures. While for regions with similar intensity values but different intensity variation directions, an inter-scale based salience measure is defined by combining the local energy of the lowpass subband coefficients and the WCV of the coarsest bandpass directional subband coefficients. Several pairs of multimodality images are fused with the proposed methods. Fusion results demonstrate that the proposed fusion method can extract more salient features (not just in energy) from source images than some other complex wavelet based fusion methods. Especially, more phase information of source images can be preserved into the fused image, which makes the proposed fusion method with higher performance in spatial consistency.Pattern Recognition Letters 01/2013; 34(2):185–193. · 1.27 Impact Factor -
##### Conference Paper: Statistical image modeling using von Mises distribution in the complex directional wavelet domain

[Show abstract] [Hide abstract]

**ABSTRACT:**In this paper, a new statistical model is proposed for modeling the nature images in the transform domain. We demonstrate that the von Mises distribution (VM) fits accurately the behaviors of relative phases in the complex directional wavelet subband from different nature images. Moreover, a new image feature based on the VM model is proposed for texture image retrieval application. The VM based feature yields higher retrieval accuracy compared to the energy features and the relative phase features. In addition to magnitude information typically used in many other feature extraction methods, the VM based phase information is also incorporated to further improve the performance.Circuits and Systems, 2008. ISCAS 2008. IEEE International Symposium on; 06/2008 - [Show abstract] [Hide abstract]

**ABSTRACT:**This paper investigates the use of complex wavelets for statistical texture retrieval in a noisy environment, in which the query image is contaminated by noise. To account for the presence of noise, the feature extraction step is based on parameter estimation in noise where features are extracted from the noisy query image by modeling the magnitude and phase of complex subband coefficients of the clean image, and relating the model's parameters to the noisy coefficients. In addition to using only the magnitude or phase which is in the form of the relative phase, we incorporate both magnitude and phase information to further improve the accuracy rate. The simulation results show the retrieval rate improvement by estimating the clean parameters from the noisy query image instead of assuming that the query image is clean. Furthermore, using both magnitude and phase of complex coefficients improves the accuracy rate from using either magnitude or phase alone, and that using complex-valued wavelets yields higher rate than using real-valued wavelets.Signal Processing Image Communication 01/2013; 28(10):1494–1505. · 1.29 Impact Factor

Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.