PREMIER GRADE LOSSY IMAGE COMPRESSION USING GREEN ERROR INFLUENCED DEMOSAICKING AND INTERPOLATION TECHNIQUES

Abstract

Recent literature of lossy image compression states the dominant role of the demosaicking and interpolation concepts. This paper develops a new method which is guided by the Green Channel Error of interpolation technique. This method is influenced by three novel methods viz. i) Hexagon pattern and Edge Direction guided Demosaicking (HPEDD), ii) Error signal constrained Neighbour based Image Upsampling (ESNIU) and iii) Green channel based Error signal generation using Histogram, Interpolation and XOR (GEHIX). The HPEDD method is constructed by the new hexagon pattern and Green error constrained rules. The ESNIU upsampling is designed based on the neighbor based interpolation aided by Green Channel Error. The GEHIX method brilliantly handles the Green Error and Basic error coefficients. This approach is abbreviated as HPEDD-ESNIU method. The Criteria formation based on Green Error with empowered HPEDD demosaicking and ESNIU upsampling, induces the compression performance to a significant level in terms of the state-of-the-art metrics like PSNR, Compression ratio, etc. This method is an apt one for web oriented applications which is the real time need of the current digital world.