Title :
Multiplier free co-difference matrix for image and video processing
Author :
A. Enis Cetin;Kaan Duman;Hakan Tuna;Abdulkadir Eryildirim
Author_Institution :
Elektrik ve Elektronik M?hendisli?i B?l?m?, Bilkent ?niversitesi, Turkey
fDate :
4/1/2009 12:00:00 AM
Abstract :
In this paper, we propose a new image feature extraction method. We define a matrix called co-difference matrix for a given region. This matrix can be computed without performing any multiplications. The operator that we use forms a semi-group on real numbers. The co-difference based feature extraction method can be used in other image and video processing applications such as object tracking, face detection, and licence plate detection. Experimental results for licence plate detection is presented.
Keywords :
"Face detection","Covariance matrix","Feature extraction","Object detection","Lagrangian functions"
Conference_Titel :
Signal Processing and Communications Applications Conference, 2009. SIU 2009. IEEE 17th
Print_ISBN :
978-1-4244-4435-9
DOI :
10.1109/SIU.2009.5136499