Paper Info
Title
Novel Progressive Region of Interest Image Coding Based on Matching Pursuits
Abstract
A progressive and scalable, region of interest (ROI) image coding scheme based on matching pursuits (MP) is presented. Matching pursuit is a multi-resolutional signal analysis tool and can be employed in order to progressively refine the quality of a set of selected regions of an image up to a specific grade. The computational complexity of this analysis method can be reduced by decreasing the size of MP dictionary. Thus, the proposed method provides a trade off between complexity, rate, and quality. By the suggested scheme, regions of an image with higher receiver's priority are refined in an interactive manner. The transmitter sends an initial coarse version of the image. Then, he receiver transmits its preferred ROI parameters. Afterwards, the reconstructed image is refined according to the ROI parameters, in a progressive way. I. INTRODUCTION For a decent image browsing experience, efficient delivery of large high resolution images is essential. On the other hand, because of increasing number of Internet users and their transmitted data rate, the available bit-rate for each application has to be restricted. With a limited bit-rate, maintaining the original high visual quality for all parts of a large high resolution image is a time consuming process. To address this problem, a selection of regions of interest of an image (according to the receiver) can be transferred with higher quality, while the remainder of the image is sent at a lower quality. Matching pursuit is a greedy signal analysis algorithm in which a signal is iteratively decomposed into a linear expansion of waveforms, chosen from an over-complete dictionary (1), (2). Assume D = {gγ}γ∈Γ is an over-complete set with finite number of normalized elements (� gγ� =1 )i nL2(R) that spans a space of dimension N (Γ is a finite dictionary index set of ND elements). Each vector f of the space can be projected on a dictionary element gγ ∈D to approximate the vector in that direction. Matching pursuit algorithm is based on the following residual vector update formula
Year
DOI
Venue
2006
10.1109/ICME.2006.262404
Toronto, Ont.
Keywords
Field
DocType
computational complexity,image coding,image matching,image reconstruction,image resolution,ROI image coding,computational complexity,image reconstruction,matching pursuit,multiresolutional signal analysis tool,region of interest
Iterative reconstruction,Matching pursuit,Transmitter,Computer vision,Signal processing,Pattern recognition,Computer science,Artificial intelligence,Region of interest,Image resolution,Computational complexity theory,Scalability
Conference
ISBN
Citations 
PageRank 
1-4244-0367-7
0
0.34
References 
Authors
2
2
Name
Order
Citations
PageRank
Abbas Ebrahimi-Moghadam1274.07
Shahram Shirani225037.41