Abstract
Discrete Wavelet Transform (DWT) has gained much limelight in the past years. Wavelet Transform has precedence over Discrete Fourier Transform and Discrete Cosine Transform because they capture the frequency as well as spatial information of a signal. In this paper DWT has been used for image scaling purpose. To achieve higher visual quality image, DWT is applied on the gray scale image using a downscaling technique. The original image is recovered using IDWT by employing different interpolation techniques for upscaling. In this paper interpolation techniques used are: Nearest Neighbor, Bilinear and Bicubic. Peak Signal to Noise Ratio (PSNR) and Mean Square Error (MSE) is calculated for quantifying interpolated image effectiveness. Results show that the reconstructed image is better when using a combination of DWT and Bicubic interpolation
Highlights
In digital imaging, image scaling technique has wide range of applications from consumer electronics to medical field [1]
In bicubic interpolation data points are interpolated on a two dimensional regular grid and the scaled image has much smoother results as compared to nearest neighbor and bilinear interpolation
Due to the admirable approximation ability of wavelet transform, image interpolation based on wavelets approach, performs well at the non-edge regions [28]
Summary
Image scaling technique has wide range of applications from consumer electronics to medical field [1]. Bilinear interpolation and bicubic interpolation are commonly used interpolation techniques. In bicubic interpolation data points are interpolated on a two dimensional regular grid and the scaled image has much smoother results as compared to nearest neighbor and bilinear interpolation. Unlike windowed Fourier analysis, a mother wavelet can change the size of widow by stretching or compressing. Due to this outstanding feature of wavelets, approximate and detail image can be attained, where approximate image of the signal is attained by large wavelets, while smaller wavelets give the detail information of the image. In the results it is shown that in order to achieve image of lower resolution, by employing diverse interpolation techniques, quality of image is affected
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
