A new interpolation method based on multi-resolution technique is presented and used for medical image zooming. The
aim of this work is to focus on similarity analysis of adjacent
sub-bands provided by Discrete Wavelet Transform (DWT)
to enhance the accuracy of the interpolation. First, decompose the original image into sub-bands by the DWT; second,
consider the similarity between adjacent sub-bands to calculate the high frequency components; third, use the original
image as the low frequency component and apply the inverse DWT to obtain the final interpolation result. Experimental
results on magnetic resonance (MR) images and positron emission tomography (PET) images illustrate the effectiveness
of the proposed method.
The images obtained by different sensors differ from each other in terms of resolution and spectrum. Image fusion is a
very useful technique which focuses on forming a new image by combining the unique and complementary information
provided by different images. A new image fusion method based on multi-resolution analysis and directional gradient is
described in this paper. The key idea is to use different fusion strategies along different directions, in order to take into
account the anisotropy of the images. The directional gradient is used in describing directional property. Experimental
results show that the proposed method is effective and promising to enhance visualization in the fused image.
An airborne vehicle such as a tactical missile must avoid obstacles like towers, tree branches, mountains and building
across the flight path. So the ability to detect and locate obstacles using on-board sensors is an essential step in the
autonomous navigation of aircraft low-altitude flight. This paper describes a novel method to detect and locate obstacles
using a sequence of images from a passive sensor (TV, FLIR). We model 3D scenes in the field-of-view (FOV) as a
collection of approximately planar layers that corresponds to the background and obstacles respectively. So each pixel
within a layer can have the same 2D affine motion model which depends on the relative depth of the layer. We formulate
the prior assumptions about the layers and scene within a Bayesian decision making framework which is used to
automatically determine the assignment of individual pixels to layers. Then, a generalized expectation maximization
(EM) method is used to find the MAP solution. Finally, simulation results demonstrate that this method is successful.
Speckle noise in synthetic aperture radar (SAR) images is characterized as multiplicative random noise. To address SAR image speckle denoising, this paper proposes a new method which is based on the combination of statistical model of wavelet coefficients and modification to the coefficients according to module-maximum-based (significant coefficient) rule. In our method, wavelet coefficients of image are firstly modeled as mixture density of two Gaussian (MG) distributions with zero mean. In order to incorporate the spatial dependencies into the denoising procedure, hidden markov tree (HMT) model is explored and expectation maximization (EM) algorithm is proposed to estimate model parameters. Bayes minimum mean square error (Bayes MMSE) method is used to estimate the wavelet coefficients free of noise. The wavelet coefficients are updated according to a rule whether the coefficient is a significant one or not. 2D inverse DWT is performed on the updated coefficients to get denoised SAR image. Experimental Results using real SAR image demonstrate that the method can not only reduce the speckle but also preserve edges and radiometric scatter points. Equivalent Number of Look Enl shows that the proposed method yields very satisfactory results compared with other methods.