Contourlet

About: Contourlet is a research topic. Over the lifetime, 3533 publications have been published within this topic receiving 38980 citations.

Fusion of Multifocus Images Based on the Nonsubsampled Contourlet Transform

Abstract: Focusing on the deficiencies of existing wavelet based algorithm, a novel algorithm for multifocus images fusion based on the nonsubsampled contourlet transform (NSCT) is proposed. And, the concepts of the local area visibility (LAVI) and the local oriented energy (LOE) are introduced in the contourlet domain. The selection principle of the low frequency subband coefficients based on the LAVI and the selection principle of the bandpass directional subband coefficients based on the LOE are presented respectively. The experimental results demonstrate that the proposed algorithm, compared to the methods based on the discrete wavelet transform and the discrete wavelet frame transform, can effectively reduce the loss of the useful information and the introduction of the artificial information. In addition, the proposed algorithm can extract more useful information from the source images, and make the fused image with higher performance in terms of both visual quality and objective evaluation criteria.

Nonsubsampled Contourlet Transform for Texture Classifications using Support Vector Machines

Abstract: In this paper, a new texture classification method using the nonsubsampled contourlet transform (NSCT) and support vector machines (SVMs) is proposed. The NSCT provides a shift-invariant, multiscale, and multidirectional image representation that has proven to be very efficient in image analysis applications. Firstly, features are extracted from NSCT coefficients of source images. In addition, SVMs, which have been demonstrated excellent performance as classifiers in a variety of pattern recognition problems, are used as classifiers for texture classification. The algorithm is tested on texture images from Brodatz album. Experimental results demonstrate that the proposed method produces more accurate classification results than other methods.

A study on NSCT based super-resolution reconstruction for infrared image

Abstract: A non-sampled contourlet transform based super-resolution reconstruction method is proposed to improve infrared image quality, for the reason of the infrared detection image always has characteristic of low spatial resolution and contrast. the reconstruct method use the non-sampled contourlet transform to decomposed origin image into low pass sub band and band pass sub band images firstly, and then the edge detection operator and parabolic error compound interpolation are used to super resolution reconstruct for all bands image with adaptive edge preserving, simultaneously the low pass sub band image is enhanced by linear transform. At last the high resolution image is restored by non-sampled contourlet transform. An image simulation conducts to verification reconstruct algorithm for multiple infrared images. By compared with other variety of methods, the result draw a conclusion, which is that the method proposed in this paper, is not only effectively on improving the contrast of infrared image, but also the edge information preserving perfectly. This method has the important engineering value for real time reconstruct high resolution infrared image in the actual infrared imaging detection application.

Algorithm for relative radiometric consistency process of remote sensing images based on object-oriented smoothing and contourlet transforms

Abstract: Traditional relative radiometric normalization methods generally depend on global statistical linear parameters, do not consider two-dimensional radiometric distribution, and do not eliminate foreground objects in an image. Thus, we present a method for relative radiometric consistency process based on object-oriented smoothing and contourlet transforms. Object-level smoothing is applied to both the reference image and the image to be normalized so as to reduce the influence of foreground objects on background radiation extraction. Then, high-frequency and low-frequency sections of an image are separated by contourlet transforms to preserve high-frequency texture information of the image to be normalized, with low-pass filtering applied to the low-frequency sections to gather the background radiation difference. Finally, contourlet reverse transforms are used to reconstruct the radiometrically normalized images. Test results show that the proposed method is effective for radiometric normalization of images with both large-scale and small-scale radiometric characteristics. The proposed method can not only normalize linear and nonlinear radiation differences at the same time, but also maximally preserve image texture information. It can improve the visual effects of normalized images and increase change detection accuracy.

Texture image retrieval based on contourlet-2.3 and generalized Gaussian density model

Abstract: In order to improve the retrieval rate of the original contourlet transform based texture image retrieval system, a contourlet-2.3 transform based texture image retrieval system was proposed. Generalized Gaussian Density (GGD) model parameters were cascaded to form feature vectors and Kullback-Leibler distance (KLD) function was used for similarity measure. Experimental results on 640 texture images from Vistex texture image database indicate that contourlet-2.3 transform based image retrieval system is superior to that of the original contourlet transform under the same system structure with almost same length of feature vectors, retrieval time and memory needed. Furthermore, GGD combined with KLD method has higher retrieval rates than energy based features combined with Euclidean distance under comparable levels of computational complexity, decomposition parameters including the number of scale and directional subband on each scale selected in both contourlet transforms can make retrieval results quite different.