MAP based resolution enhancement of video sequences using a Huber-Markov random field image prior model
نویسندگان
چکیده
In this paper, we propose two approaches for video sequence resolution enhancement using Maximum A Posteriori (MAP) estimation. Huber-Markov Random Fields (HMRF) are used as prior models. These models can better preserve image discontinuities (edges) when compared with Gaussian prior models. The two proposed approaches differ in the selection of image smoothness measure. The first approach employs a measure that is based on a discrete Laplacian kernel, while the second approach uses a finite difference approximation of second order derivatives at each pixel of the high-resolution image estimate. Experimental results are presented and conclusions are drawn.
منابع مشابه
Cluster-Based Image Segmentation Using Fuzzy Markov Random Field
Image segmentation is an important task in image processing and computer vision which attract many researchers attention. There are a couple of information sets pixels in an image: statistical and structural information which refer to the feature value of pixel data and local correlation of pixel data, respectively. Markov random field (MRF) is a tool for modeling statistical and structural inf...
متن کاملEstimation of the Regularisation Parameter in Huber-MRF for Image Resolution Enhancement
The Huber Markov Random Field (H-MRF) has been proposed for image resolution enhancement as a preferable alternative to Gaussian Random Markov Fields (G-MRF) for its ability to preserve discontinuities in the image. However, its performance relies on a good choice of a regularisation parameter. While automating this choice has been successfully tackled for G-MRF, the more sophisticated form of ...
متن کاملAdaptive Markov Random Fields for Example-Based Super-resolution of Faces
Image enhancement of low-resolution images can be done throughmethods such as interpolation, super-resolution using multiple video frames, and example-based super-resolution. Example-based super-resolution, in particular, is suited to images that have a strong prior (for those frameworks that work on only a single image, it is more like image restoration than traditional, multiframe super-resol...
متن کاملInterscale Predictive Wavelet Coding with Huber Markov Random Field
An interscale predictive wavelet coding (IPWC) scheme is proposed in this paper. The image is encoded top-down, from the coarsest scale to the finest scale. At each scale, the current scale coefficients are predicted from the coded coefficients at the coarser scale and the prediction residue is encoded. The interscale prediction is based on MAP estimation with Huber Markov random field statisti...
متن کاملHigher-order MRFs based image super resolution: why not MAP?
A trainable filter-based higher-order Markov Random Fields (MRFs) model the so called Fields of Experts (FoE), has proved a highly effective image prior model for many classic image restoration problems. Generally, two options are available to incorporate the learned FoE prior in the inference procedure: (1) sampling-based minimum mean square error (MMSE) estimate, and (2) energy minimization-b...
متن کامل