A Replacement Based Video De-Interlacing Technique by Feathering Effect Detection and Artifact Agglomeration Index

Several video de-interlacing techniques have been already developed and each one presents better performance in certain conditions. Occasionally even the most modern de-interlacing techniques create frames with worse quality than primitive de-interlacing processes. This paper validates that the fina...

Full description

Saved in:
Bibliographic Details
Published inWVC 2017 : proceedings, XIII Workshop of Computer Vision : Natal, Rio Grande do Norte, Brazil, 30 October-1 November 2017 pp. 49 - 54
Main Authors Martins, Andre Luis, Linhari Rodrigues, Evandro Luis, Veludo De Paiva, Maria Stela
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2017
Subjects
Bandwidth
De-interlacing
Detectors
ELA
Feathering Effect
Image edge detection
Image quality
Indexes
Online AccessGet full text

Cover

More Information
Summary:Several video de-interlacing techniques have been already developed and each one presents better performance in certain conditions. Occasionally even the most modern de-interlacing techniques create frames with worse quality than primitive de-interlacing processes. This paper validates that the final image quality can be improved if the frames are processed locally and selectively, combining different types of de-interlacing techniques. The proposed strategy is able to identify and locally correct defects on de-interlaced frames with the support of a feathering effect detector. Starting from a de-interlaced frame produced by "inter-field average" method, the defective areas are identified and fully replaced by pixels generated through "edge-based line average" method (ELA). Test results have proven that the proposed technique is able to produce video frames with higher quality than applying a single de-interlacing technique through getting what is good from intra- and inter- field methods.
DOI:10.1109/WVC.2017.00016