Maximizing image quality over visual sensor networks via DCT bit allocation

Sungjin Lee, Sanghoon Lee, Alan C. Bovik

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Recently, VSNs (Visual Sensor Networks) are becoming of increased interest in a variety of applications. Due to power constraints at each visual sensor node, the processing of live video is constrained. In this paper, we propose a DCT (Discrete Cosine Transform)-based bit allocation scheme to maximize image quality for a given bit rate constraint. Two major featured algorithms are proposed : "SLO" (Searching from the Lowest Order) and "SPP" (Searching the Present optimal bit allocation from the Previous optimal bit allocation). The proposed algorithms reduce the search range based on a simple IQA (Image Quality Assessment) model and DCT statistics. Thus, they demonstrate low-complexity and fast computation. In the simulations, we show the superiority of the proposed algorithms over conventional approaches.

Original languageEnglish
Title of host publication2010 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2010 - Proceedings
Pages69-72
Number of pages4
DOIs
Publication statusPublished - 2010
Event2010 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2010 - Austin, TX, United States
Duration: 2010 May 232010 May 25

Publication series

NameProceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation

Other

Other2010 IEEE Southwest Symposium on Image Analysis and Interpretation, SSIAI 2010
Country/TerritoryUnited States
CityAustin, TX
Period10/5/2310/5/25

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Maximizing image quality over visual sensor networks via DCT bit allocation'. Together they form a unique fingerprint.

Cite this