DocumentCode
703552
Title
Information criteria based edge detection
Author
Jouzel, Frederic ; Olivier, Christian ; El Matouat, Abdelaziz
Author_Institution
PSI-La3i, Univ. of Rouen, Rouen, France
fYear
1998
fDate
8-11 Sept. 1998
Firstpage
1
Lastpage
4
Abstract
We propose an algorithm to detect ruptures in a signal by means of information criteria along with its application to edge detection on grey-level images. Information criteria are made of two terms: a likelihood term and an original penalization factor. They enable us to find in an optimal way autoregressive models and their number to model the signal. The model changes represent the edges of the objects in the image.
Keywords
autoregressive processes; edge detection; image colour analysis; image representation; edge detection; grey-level imaging; image representation; information criteria; optimal way autoregressive model; original penalization factor; rupture detection; Biological system modeling; Image edge detection; Integrated circuit modeling; Maximum likelihood estimation; Probability density function;
fLanguage
English
Publisher
ieee
Conference_Titel
Signal Processing Conference (EUSIPCO 1998), 9th European
Conference_Location
Rhodes
Print_ISBN
978-960-7620-06-4
Type
conf
Filename
7090023
Link To Document