Author : A. S. Awati 1
Date of Publication :7th February 2017
Abstract: Image inpainting is popularly used to automatically recover the damaged or missing regions in digital images and is used to remove the unwanted objects from images. This paper presents a technique for recovering the curves of damaged regions of digital images using cubic splines. The proposed technique is divided into two steps. In the first step the curved structure of the object in the damaged region is recovered using splines. The user marks a reasonable number of points on the structure of the object around the damaged region. Using these points a spline is constructed and the isophote lines arriving at the contours of the damaged region are interpolated into the damaged regions using that spline. After recovering the structure of the object in the damaged or missing region, in the second step, a fill in process is done to fill in the color information in the damaged region. Any inpainting technique can be used in the second step to fill in the texture information into the damaged region. The proposed algorithm is tested over a large variety of images and has shown excellent results.
Reference :
-
[1] Viacheslav Voronin, Vladimir Marchuk, Dmitriy Bezuglov, Maria Butakova, “Inpainting strategies for reconstruction of missing data in images and videos: techniques, algorithms and quality assessment”, Springer Direct, April 2016.
[2] M. Shahid Farid, Hassan Khan, Arif Mahmood, “Image Inpainting Using Cubic Hermit Spline,” Proceedings of SPIE - The International Society for Optical Engineering, October 2011.
[3] Póth Miklós, “IMAGE INTERPOLATION TECHNIQUES”, Semantic Scholar 2004.
[4] Micheal Unser, “A perfect fit for signal and image processing”, IEEE Signal Processing Magzine, Nov 1999.
[5] Mhammad Sarfraz and Naelah Al-Dabbous, “Curve Representation for Outlines of Planar Images using Multilevel Coordinate Search,” WSEAS TRANSACTIONS on COMPUTERS, Issue 2, Volume 12, February 2013.
[6] Fabio Bellavia and Carlo Colombo “Colour correction for image stitching by monotone Cubic Spline interpolation”, IEEE 2015.