ISSN : 2319-7323




INTERNATIONAL JOURNAL OF COMPUTER SCIENCE ENGINEERING


Open Access

ABSTRACT

Title : A Method of DWT with Bicubic Interpolation for Image Scaling
Authors : Remimol.A.M, Sekar.K
Keywords : Image scaling; Discrete Wavelet Transform (DWT); Peak Signal to Noise Ratio (PSNR); Bicubic interpolation; Mean Square Error (MSE)
Issue Date : March 2014
Abstract : Image scaling is a technique used to scale down or scale up the pictures or video frames to fit to the application and is very important in image processing applications. In this paper, a new scaling algorithm is proposed for image scaling consisting of a Discrete Wavelet Transform (DWT) based interpolation and bicubic interpolation. A simple Haar wavelet based DWT interpolation is performed to the image to achieve higher Peak Signal to Noise Ratio (PSNR) and then bicubic interpolation is carried out to get higher visual quality. The sub band coding based DWT divides the image into four quadrants according to frequency. To reduce the artifacts, bicubic interpolation is performed to all the four frequency quadrants. This method achieves less Mean Square Error (MSE) and high average Peak Signal to Noise Ratio (PSNR). Thus this work can achieve an image quality by a factor more than 8 dB than the existing interpolation methods. The proposed method greatly reduces the image artifacts like blurring, thus this approach is better than existing methods in visual quality. The simulation of the work is performed in MATLAB R2013a.
Page(s) : 131-135
ISSN : 2319-7323
Source : Vol. 3, No.2