Se-hyeok Park

Self-synthesis based super-resolution for a natural image

In this paper, we propose self-synthesis based super-resolution algorithm for natural images. The proposed method analyzes the property of the natural image and focuses on the self-similarity of natural objects. Structure and texture components are extracted from the self-image and synthesized on the interpolated image with matching weights to restore the self-similarity of interpolated image. Then, the synthesized image is compensated with MAP estimator to prevent artifacts. Experimental results show that the proposed method gives more natural texture in terms of human perception.

Contrast-preserved chroma enhancement technique using YCbCr color space

The YCbCr color space is widely used in image processing techniques. The preference for YCbCr is due to the rough correspondence between the YCbCr components and visual attributes. However, image contrast decreases as the chroma value increases owing to the nonorthogonality between YCbCr components. In this study, a contrast-preserved chroma enhancement algorithm using YCbCr signals is proposed by predicting the amounts of luma change needed to compensate for the lightness change induced by chroma enhancement. Numerical and psychophysical experimental results show that the proposed algorithm can effectively enhance the perceived image chroma and details.

Horizontal scrolling text processing for frame rate conversion

In this paper, we propose a novel method reducing motion artifacts in horizontal scrolling text (HST) area which has been frequently occurred during frame rate conversion (FRC) processing. The proposed method takes place in the following two steps: HST area detection and motion vector (MV) correction. The most dominant MV in each horizontal line is referred to as a block line motion vector (BLMV) and its direction and magnitude are used to detect HST regions. MVs in the HST regions are corrected in order to suppress the motion artifacts. It is advantageous to improve the performance of various motion-based image processing applications such as FRC. Its merits and performance in motion artifact reduction could be verified through a set of visual experiments.

An integrated framework for both compression noise reduction and super-resolution

Even though super-resolution is a promising technique, it can cause unwanted increase of compression noises, if it is applied to an image coded with a low bit-rate. To solve this problem, we propose an integrated framework to reduce compression noises and to recover high-frequency components of a coded low-resolution image. The framework effectively combines two patch-based algorithms, a patch-matching-based 3-D filtering algorithm and an example-based super-resolution algorithm, based on the combination of 3-D transform coefficients. Experimental results demonstrate that the proposed framework successfully improves the resolution while alleviating compression noises in the images coded with low bit-rates.

Advanced local fallback processing for motion-compensated frame rate up-conversion

We introduce an advanced local fallback algorithm that is sensible of an object boundary. In previous studies, local vector smoothness (LVS) was used to determine the local fallback ratio. However, LVS in a properly interpolated object boundary results in unnecessary motion. Our local fallback algorithm application is sensible of the moving object boundary and occlusion though spatial and temporal analysis of motion vector reliability.

Edge enhancement of a sub-pixel rendered image for printing system

Recently, sub-pixel rendered filter is often used as a post processing of graphic contents to reduce the jagging artifacts in displays. However, in a printing system, sub-pixel filtered image yields blurring and jagging artifacts due to the difference of displaying method. In order to prevent artifacts, new edge enhancement method is proposed. Edge region of an image is processed by an adaptive intensity mapping function to reduce the mid-range of intensities that make jagging artifacts in half-toning. Then, a color replacement function is applied to reduce the color fringe artifacts. Experimental results show that the proposed algorithm improves the quality of printed sub-pixel rendered image.