Image Magnification and Demagnification using Fuzzy Lattice Morphological Transformation

Image magnification has a wide range of application, including the uploading of images to a web page, display of images in mobile phones, PDAs or screens. Due to the limited memory in these devices, the need to use simple image magnification algorithms arises. Mathematical morphology based on algebraic framework endows it with strong properties and allows multiple extensions for image magnification. In specific, extensions to fuzzy sets using morphological operators namely, dilation and erosion are performed while preserving the properties of these operators. This paper utilizes this approach that helps in magnifying the images, called, Interval-valued Fuzzy Lattice Morphology-based Image Transformation (IFLM-IT). The IFLM-IT method obtains as input an original colour RGB image using Cartesian Co-ordinate system. Next, Fuzzy Lattice Morphology-based Image Transformation is applied to reduce original colour RGB image to smaller size aiming at attaining better quality image. In this work, we associate Interval-valued Fuzzy Sets to magnify transformed image to original size. Based on this set, the magnified image is compared with the original colour image. We show certain experimental results and study how the Interval-valued Fuzzy Lattice Morphology-based Image Transformation has influence on the results obtained by the algorithm. Comprehensive evaluations on a large dataset well demonstrate the better performance of IFLM-IT method over other state-of-the-arts for image magnification.