Fig 1 - uploaded by W.Z Shao
Content may be subject to copyright.
Six test images for synthetic super-resolution experiments. Left to right and top to bottom: Cameraman, Lena, Barbara, Peppers, Boat, House  

Six test images for synthetic super-resolution experiments. Left to right and top to bottom: Cameraman, Lena, Barbara, Peppers, Boat, House  

Source publication
Article
Full-text available
This paper proposes to adaptively combine the known total variation model and more recent Frobenius norm regularization for multi-frame image super-resolution (SR). In contrast to existing literature, in this paper both the composite prior modeling and posterior variational optimization are achieved in the Bayesian framework by utilizing the Kullba...

Context in source publication

Context 1
... validate the performance of our proposed method, six HR images are used to generate synthetic LR frames, which are respectively Cameraman, Lena, Barbara, Peppers, Boat, and House, as shown in Fig. 1. All the images are of size 256 × 256. In specific, the synthetic LR frames are generated in the following manner. As for each HR image, 16 LR frames are acquired via translational and rotational motions, a 5 × 5 Gaussian point spread function (PSF) with standard deviation 1 and downsampling by a factor of 4; and amounts of translation ...

Similar publications

Article
Full-text available
The present case study aims to contribute to the knowledge of music composition as a meaningful activity in music education. An extensive amount of literature on music composition is available; however, music composition as a regular classroom activity is rarely seen in elementary schools. The effects of closely guided music composition, in which e...

Citations

... We mention an approach combining sparse priors (TV, 1 ) with non-sparse Gaussian prior [14], yet the estimation procedure assumes fixed weights between the priors. Other examples are combining 1:1 TV prior with a prior based on Frobenius norm of Hessian [15], or composing priors with the first or second order derivatives in different directions [16]. More recent papers propose to integrate prior terms with an adaptively estimated norm inside the model [17], [18]. ...
Article
Image restoration is typically defined as an ill-posed problem which has to be regularized to obtain an acceptable solution. In Bayesian interpretation, regularization is equivalent to prior model of the image. An added value of Bayesian point of view is the ability to form a hierarchical model and estimate the hyper-parameters of the prior from the data. Many prior models are available, usually based on automatic relevance determination principle applied to the transformed image. However, the transformation (the most common is a differential operator) is assumed to be known. In this paper, we propose to relax this assumption and estimate the image transformation from the data. The resulting algorithm is analytically tractable using the Variational Bayes method. Properties of the new prior are demonstrated on the problem of image super-resolution.
Article
The multiframe super-resolution (SR) technique aims to obtain a high-resolution (HR) image by using a set of observed low-resolution (LR) images. In the reconstruction process, artifacts may be possibly produced due to the noise, especially in presence of stronger noise. In order to suppress artifacts while preserving discontinuities of images, in this paper a multiframe SR method is proposed by involving the reconstruction properties of the half-quadratic prior model together with the quadratic prior model using a convex combination. Moreover, by analyzing local features of the underlined HR image, these two prior models are combined by using an automatically calculated weight function, making both smooth and discontinuous pixels handled properly. A variational Bayesian inference (VBF) based algorithm is designed to efficiently and effectively seek the solution of the proposed method. With the VBF framework, motion parameters and hyper-parameters are all determined automatically, leading to an unsupervised SR method. The efficiency of the hybrid prior model is demonstrated theoretically and practically, which shows that our SR method can obtain better results from LR images even with stronger noise. Extensive experiments on several visual data have demonstrated the efficacy and superior performance of the proposed algorithm, which can not only preserve image details but also suppress artifacts.
Conference Paper
Anisotropic Total variation (ATV) based super resolution method is a state-of-the-art method, which is popularly used. However, the traditional ATV based method is a kind of local method, which is just considered its 4 or 8 neighborhood of each pixel, without considering a larger region. In this paper, we proposed a non-local based ATV method (NLATV), which considers a larger region of each pixel. The experimental results show that the proposed method is much better than the original ATV method in peak signal-to-noise radio and visual quality.