Premium
Survey of progressive image transmission methods
Author(s) -
Chee Y.Kheong
Publication year - 1999
Publication title -
international journal of imaging systems and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.359
H-Index - 47
eISSN - 1098-1098
pISSN - 0899-9457
DOI - 10.1002/(sici)1098-1098(1999)10:1<3::aid-ima2>3.0.co;2-e
Subject(s) - computer science , image (mathematics) , transmission (telecommunications) , residual , image compression , quantization (signal processing) , artificial intelligence , algorithm , coding (social sciences) , computer vision , image processing , pattern recognition (psychology) , mathematics , telecommunications , statistics
Progressive image transmission (PIT) is an elegant method for making effective use of communication bandwidth. Unlike conventional sequential transmission, an approximate image is transmitted first, which is then progressively improved over a number of transmission passes. PIT allows the user to quickly recognize an image and is essential for databases with large images and image transmission over low‐bandwidth connections. This article presents a review of PIT techniques. A classification scheme based on the method used to progressively update the image is proposed. Four different classes of PIT methods are identified: successive approximation, transmission sequence‐based, multistage residual quantization, and multiresolutional or hierarchical coding methods. Subclasses are defined based on the image compression method used. Using this classification, a comprehensive survey and comparison of these methods is performed. © 1999 John Wiley & Sons, Inc. Int J Imaging Syst Technol 10: 3–19, 1999