DocumentCode :
3542668
Title :
Optimizing JPEG XR tile structure for fast local access
Author :
Tu, Chengjie ; Sullivan, Gary J. ; Srinivasan, Sridhar
Author_Institution :
Microsoft Corp., Redmond, WA, USA
fYear :
2010
fDate :
9-13 Jan. 2010
Firstpage :
379
Lastpage :
380
Abstract :
Compressed digital image technology is fundamental to a variety of consumer electronics applications. Coded images are often accessed in part rather than in whole, by accessing spatial regions of interest (ROIs). In many application architectures, the decoding process to access each ROI may occur separately - even when there are multiple such ROIs to be accessed within the same image. Moreover, some image coding schemes, such as JPEG XR and JPEG 2000, include some form of support for structuring the coded image data into distinct regions that can be referred to as tiles. The performance associated with decoding an ROI depends heavily on how the image is tiled. This paper provides information about the effects of image tile structuring on ROI access in the JPEG XR image coding context. Some guidelines are provided regarding how an image should be tiled to enable fast local access by aligning the ROI access with the tile structure to minimize decoding resource overhead.
Keywords :
data compression; image coding; optimisation; JPEG 2000; JPEG XR tile structuring; compressed digital image technology; consumer electronics applications; decoding process; image coding; optimization; spatial regions of interest; Consumer electronics; Decoding; Digital images; Entropy; Guidelines; ISO standards; Image coding; Image reconstruction; Tiles; Transform coding;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Consumer Electronics (ICCE), 2010 Digest of Technical Papers International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
978-1-4244-4314-7
Electronic_ISBN :
978-1-4244-4316-1
Type :
conf
DOI :
10.1109/ICCE.2010.5418690
Filename :
5418690
Link To Document :
بازگشت