Please use this identifier to cite or link to this item:
Title: Highly - compacted DCT coefficients
Authors: Dagher, Issam 
Affiliations: Department of Computer Engineering 
Keywords: Discrete Cosine Transform (DCT)
Energy Compactness
Dimensionality reduction
Face recognition
Subjects: Image compression
Issue Date: 2010
Part of: Journal of signal image and video processing
Volume: 4
Issue: 3
Start page: 303
End page: 307
In this paper highly-compacted DCT coefficients (HDCT) are presented. This compactness is achieved by sorting in ascending order the data first, then by applying the Discrete Cosine transform (DCT) to the ordered data. Images are highly correlated. DCT exhibits excellent energy compaction. It will be shown that HDCT has much better energy compactness than the DCT. This has the effect of representing every ordered image with very small number of HDCT coefficients (dimensionality reduction). The compression capabilities of the HDCT are presented. HDCT is also applied to face recognition problem. Simulation results on different databases showed high average success rate of this algorithm compared to other algorithms.
Ezproxy URL: Link to full text
Type: Journal Article
Appears in Collections:Department of Computer Engineering

Show full item record

Record view(s)

checked on Apr 13, 2024

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.