This paper presents an edge-directed image interpolation algorithm. In the proposed algorithm, the edge directions are implicitly estimated with a statistical-based approach. In opposite to explicit edge directions, the local edge directions are indicated by length-16 weighting vectors. Implicitly, the weighting vectors are used to formulate geometric regularity (GR) constraint (smoothness along edges and sharpness across edges) and the GR constraint is imposed on the interpolated image through the Markov random field (MRF) model. Furthermore, under the maximum a posteriori-MRF framework, the desired interpolated image corresponds to the minimal energy state of a 2-D random field given the low-resolution image. Simulated annealing methods are used to search for the minimal energy state from the state space. To lower the computational complexity of MRF, a single-pass implementation is designed, which performs nearly as well as the iterative optimization. Simulation results show that the proposed MRF model-based edge-directed interpolation method produces edges with strong geometric regularity. Compared to traditional methods and other edge-directed interpolation methods, the proposed method improves the subjective quality of the interpolated edges while maintaining a high PSNR level.

Markov Random Field Model-Based Edge-Directed Image Interpolation

A modification of the new edge-directed interpolation method is presented. The modification eliminates the prediction error accumulation problem with adopting a modified training window structure, and further extends the covariance matching into multiple directions for suppressing the covariance mis-match problem. Simulation results show that the proposed method achieves remarkable subjective performance in preserving the edge smoothness and sharpness among other methods in literature. It also demonstrates consistent objective performance among a variety of images.

/pdf/a-modified-edge-directed-interpolation-for-images-3937n7i590.pdf

A Modified Edge Directed Interpolation for images

In this paper, we present a novel edge-directed upsampling method based on radial basis function (RBF) interpolation. In order to remove artifacts such as blurred edges or blocking effects, we suggest a nonlinear method capable of taking edge information into account. The resampling evaluation is determined according to the edge orientation. The proposed scheme is as simple to implement as linear methods but it demonstrates improved visual quality by preserving the edge features better than the classical linear interpolation methods. The algorithm is compared with some well-known linear schemes as well as recently developed nonlinear schemes. The resulting images demonstrate the new algorithm's ability to magnify an image while preserving the edge features.

Nonlinear Image Upsampling Method Based on Radial Basis Function Interpolation

Image zooming method using edge-directed moving least squares interpolation based on exponential polynomials

High quality video interpolation is always desirable, since the definition of video display devices is improving. It is always necessary to port a video of lower quality to higher quality displays, such as the conversion of SDTV videos to HDTV videos. In this paper we will present a further improved Edge-Directed Interpolation (EDI) by proposing a new sampling pattern and subsequently give a fast approach for the enlargement of a SDTV (704×480) video into a HDTV (1280×720) video (1.5 time enlargement). Experimental results show that our approach has a better performance in terms of edge quality compared to that of the New Edge-Directed Interpolation (NEDI) in the literature. Moreover, the fast approach is 50% faster and the visual quality is also better than the NEDI.

https://www.eie.polyu.edu.hk/~wcsiu/paper_store/Conference/2010_Conf/2010_EUSIPCO_Wong-n-Siu_1569292991.pdf

Further improved Edge-Directed Interpolation and fast EDI for SDTV to HDTV conversion

Recently many nonlinear interpolation schemes for images have produced images of higher resolution than those produced by linear techniques But smoothness still exists, especially on edges The smoothing process can be described as a traditional heat diffusion equation In this paper, this equation is combined with an improved edge-directed interpolation method to improve visual quality in terms of sharpened edges Experimental results indicate that the combination of the two methods is capable of achieving images with noticeably sharper edges

Improvement of a nonlinear image interpolation method based on heat diffusion equation

A local-to-holistic face recognition approach is proposed, which is an extension to elastic graph matching. In the first step, the facial features such as eyes, mouth and nose, are detected by an eigenface method, so we can set up the local feature model (eye, mouth and nose) and the holistic feature model (inner face). In the second step, an elastic graph-matching (EGM) algorithm is performed not only on holistic features, but also on local features. An experiment on the Yale face database shows the performance of the approach.

Le-Nan Wu

Papers

Improvement of a nonlinear image interpolation method based on heat diffusion equation

A local-to-holistic face recognition approach using elastic graph matching