On the use of statistical tests on homomorphic filtered SAS images

Author

Mandhouj, Imen ; Maussang, Frederic ; Solaiman, Basel ; Amiri, Hamid

Author_Institution

CID Image & Inf. Process. Dept., Inst. Mines-Telecom, Brest, France

fYear

2013

fDate

10-14 June 2013

Firstpage

1

Lastpage

5

Abstract

The principal objective of this paper is the statically study of pre-processed Synthetic Aperture Sonar (SAS) images. Different statistical tests are applied testing for distributional assumption in general and for normality in particular. Results of our statistical study are tested on a sample of SAS image filtered using a specific homomorphic filter depicted in our previous work [1]. Accordingly, hypothesis tests and graphical testing methods has been exploited to select the best representative of a filtered sample image SAS in order to perform a follow detection process.

Keywords

filtering theory; sonar imaging; statistical testing; graphical testing methods; homomorphic filtered SAS images; hypothesis tests; preprocessed synthetic aperture sonar images; statistical tests; Distribution functions; Filtering theory; Gaussian distribution; Hafnium; Synthetic aperture sonar; Testing; Homomorphic filter; detection; hypothesis test; sonar image; statistical image study; testing for normality;

fLanguage

English

Publisher

ieee

Conference_Titel

OCEANS - Bergen, 2013 MTS/IEEE

Conference_Location

Bergen

Print_ISBN

978-1-4799-0000-8

Type

conf

DOI

10.1109/OCEANS-Bergen.2013.6608112

Filename

6608112