Resolution Enhancement by Prediction of the High-Frequency Image Based on the Laplacian Pyramid
- PDF / 2,377,201 Bytes
- 11 Pages / 600.03 x 792 pts Page_size
- 18 Downloads / 166 Views
Resolution Enhancement by Prediction of the High-Frequency Image Based on the Laplacian Pyramid Bo-Won Jeon,1 Rae-Hong Park,1 and Seungjoon Yang2 1 Department 2 Digital
of Electronic Engineering, School of Engineering, Sogang University, C.P.O. Box 1142, Seoul 100-611, Korea Media Research and Development Center, Samsung Electronics Corporation, Ltd., Suwon 442-742, Korea
Received 30 November 2004; Revised 20 March 2005; Accepted 7 April 2005 According to recent advances in digital image processing techniques, interest in high-quality images has been increased. This paper presents a resolution enhancement (RE) algorithm based on the pyramid structure, in which Laplacian histogram matching is utilized for high-frequency image prediction. The conventional RE algorithms yield blurring near-edge boundaries, degrading image details. In order to overcome this drawback, we estimate an HF image that is needed for RE by utilizing the characteristics of the Laplacian images, in which the normalized histogram of the Laplacian image is fitted to the Laplacian probability density function (pdf), and the parameter of the Laplacian pdf is estimated based on the Laplacian image pyramid. Also, we employ a control function to remove overshoot artifacts in reconstructed images. Experiments with several test images show the effectiveness of the proposed algorithm. Copyright © 2006 Hindawi Publishing Corporation. All rights reserved.
1.
INTRODUCTION
In most electronic imaging applications, resolution enhancement (RE) of images containing high-density pixels with high detail is desired and often required [1]. Most of all, fast advance in multimedia technology requires RE more and more. One application is to reconstruct a higher-quality digital image from a low-resolution (LR) image that is obtained with an inexpensive camera/camcorder for printing or frame-freeze purpose. Another application is conversion from a National Television System Committee (NTSC) video signal to a high-definition television (HDTV) signal to display a standard video signal on HDTV with less visual artifacts. Also, synthetic zooming of the region of interest (ROI) is another important application in TV home shopping. Generally, there are several issues to be considered in RE: unavoidable blurring artifacts, reconstruction of HF details with annoying artifacts, and high computational cost. When a portion of a digital image acquired once is enlarged for display, RE or enlargement is an indispensable digital image processing technique. RE of digital images corresponds to reduction of the spatial sampling interval, in which HF components of the resolution-enhanced images become larger. Therefore, for effective RE of digital images, it is necessary to estimate by some means the HF components that are lost in image data acquisition.
Conventional linear interpolation schemes (e.g., bilinear and bicubic) based on space-invariant models produce interpolated images with blurred edges and annoying artifacts [2, 3]. Linear interpolation is generally preferred for the compu
Data Loading...
