A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Multichannel Texture Analysis Using Localized Spatial Filters
IEEE Transactions on Pattern Analysis and Machine Intelligence
Characterization of Signals from Multiscale Edges
IEEE Transactions on Pattern Analysis and Machine Intelligence
Texture Features for Browsing and Retrieval of Image Data
IEEE Transactions on Pattern Analysis and Machine Intelligence
Texture Classification by Wavelet Packet Signatures
IEEE Transactions on Pattern Analysis and Machine Intelligence
Multiple Resolution Texture Analysis and Classification
IEEE Transactions on Pattern Analysis and Machine Intelligence
Wavelet-based texture retrieval using generalized Gaussian density and Kullback-Leibler distance
IEEE Transactions on Image Processing
Texture classification and segmentation using wavelet frames
IEEE Transactions on Image Processing
Hi-index | 0.00 |
In the paper, we proposed a novel feature descriptor using over-complete wavelet transform and wavelet domain based fractal signature for texture image analysis and retrieval. Traditionally, discrete wavelet frame took the first order derivative of smoothing function into account, which is equivalent to Canny edge detection, with the specific case using Gaussian function as smoothing function. The second order derivative Spline Wavelet has more stronger ability to distinguish the variation of the edge width than the first order one. The over-complete B-Spline wavelet scheme is discussed and the finite impulse response of over-complete wavelet transform is also represented in the paper. In feature extraction phase, 56 dimensional statistical features, including means and variances in positive and negative parts of wavelet coefficients, are extracted respectively. At the same time, the fractal signature based on the fractal surface area function in a Besov space is very accurate and robust for gray scale texture classification so that 24 dimensional over-complete wavelet based fractal feature is extracted. Experimental results have shown that the proposed method is reasonable to describe the characteristics of the texture in temporal-frequent and fractal domains and can achieve the highest retrieval rate comparing with Gabor filter, first order derivative over-complete wavelet transformation, and some other pyramid-structured wavelet transformation considered.