Hybrid Approach for Single Image Super Resolution using ISEF/CANNY/Anisotropic Diffusion and IBP: Specific Reference to License Plate

Abstract

There are many applications which require high quality images. The applications are remote sensing,
enlarging consumer photographs, medical imaging, surveillance systems. If we look at the problem of super resolution
from the manufacturing side then increase in number of pixel per unit area by reducing sensor size or increasing the chip
size to get super resolution image. . But reducing pixel size technically increases the shot noise and increasing chip size
increases the physical size of imaging system which further increases the cost of the system. So, there is always a trade
off exits between physical size, cost and spatial resolution of the imaging system. A new approach to get high resolution
image which overcomes this above mentioned limitation is the use of image processing to get super resolved image. This
image processing technique is known as the super resolution (SR). Super resolution is the method of obtaining high
resolution image from one or more low resolution images. In this paper, a hybrid approach for single image super
resolution for vehicle license plate is presented. It is based on Iterative Back Projection(/ibp) method combined with
edge preserving Infinite Symmetrical Exponential Filter(ISEF). Though IBP can minimize the reconstruction error
significantly in iterative manner and gives good result, it suffers from ringing effect and chessboard effect because error
is back projected without edge guidance. ISEF provides edge-smoothing image by adding high frequency information
and gives more better results. Further we have used Anisotropic Diffusion with this hybrid approach.