Please use this identifier to cite or link to this item: http://idr.nitk.ac.in/jspui/handle/123456789/11425
Full metadata record
DC FieldValueLanguage
dc.contributor.authorShilpa, Kamath, S.-
dc.contributor.authorAparna, P.-
dc.contributor.authorAntony, A.-
dc.date.accessioned2020-03-31T08:31:21Z-
dc.date.available2020-03-31T08:31:21Z-
dc.date.issued2018-
dc.identifier.citationAEU - International Journal of Electronics and Communications, 2018, Vol.95, , pp.73-81en_US
dc.identifier.urihttp://idr.nitk.ac.in/jspui/handle/123456789/11425-
dc.description.abstractRecent advancements in the capture and display technologies motivated the ITU-T Video Coding Experts Group and ISO/IEC Moving Picture Experts Group to jointly develop the High-Efficiency Video Coding (HEVC), a state-of-the-art video coding standard for efficient compression. The compression applications that essentially require lossless compression scenarios include medical imaging, video analytics, video surveillance, video streaming etc., where the content reconstruction should be flawless. In the proposed work, we present a gradient-oriented directional prediction (GDP) strategy at the pixel level to enhance the compression efficiency of the conventional block-based planar and angular intra prediction in the HEVC lossless mode. The detailed experimental analysis demonstrates that the proposed method outperforms the lossless mode of HEVC anchor in terms of bit-rate savings by 8.29%, 1.65%, 1.94% and 2.21% for Main-AI, LD, LDP and RA configurations respectively, without impairing the computational complexity. The experimental results also illustrates that the proposed predictor performs superior to the existing state-of-the-art techniques in the literature. 2018 Elsevier GmbHen_US
dc.titleGradient-oriented directional predictor for HEVC planar and angular intra prediction modes to enhance lossless compressionen_US
dc.typeArticleen_US
Appears in Collections:1. Journal Articles

Files in This Item:
File Description SizeFormat 
5 Gradient-oriented directional.pdf1.04 MBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.