Certain investigation on features based object classification under cluttered background and mild occlusions

Abstract

Object classification is one of the most challenging problems in newlinecomputer vision An object classification system must be able to detect the newlinepresence or absence of objects eg Car Category Airplane Category newlineNatural Scene Category under different illumination background clutter newlineocclusions and pose changes Many different approaches and feature newlineextraction methodologies concerning object classification have been proposed newlineand applied in the last decades Most of the literature surveys discuss object newlineclassification without cluttered background or without occlusions and discuss newlineonly classification performances Barely few almost none have proposed a newlinesuitable framework for object classification problem under the cluttered newlinebackground with and without occlusions by analyzing the computational time newlinefor classification newline newline