A new segmentation approach for unimodal image histograms: Application to the detection of regions of interest in sonar images

Author

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

Author_Institution

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

fYear

2014

fDate

14-19 Sept. 2014

Firstpage

1

Lastpage

6

Abstract

The purpose of our work is to define an original approach to determine the threshold of unimodal image histograms in a robust manner. Our proposed segmentation approach refers to Shannon entropy. Threshold estimation is based on the exploitation of the curve of the entropy loss quantity. The final expression of the sigmoid estimated function of the entropy error and derivatives are used to select the image´s threshold segmentation. The proposed approach is evaluated on a set of SAS (Synthetic Aperture Sonar) images.

Keywords

entropy; image segmentation; sonar detection; sonar imaging; SAS images; Shannon entropy; entropy derivatives; entropy error; entropy loss quantity curve; image threshold segmentation; region of interest detection; sigmoid estimated function; synthetic aperture sonar images; threshold estimation; unimodal image histogram segmentation approach; Entropy; Estimation; Histograms; Image segmentation; Shape; Synthetic aperture sonar; entropy error function; logistic function; sigmoidal curve; sonar image; unimodal thresholding;

fLanguage

English

Publisher

ieee

Conference_Titel

Oceans - St. John's, 2014

Conference_Location

St. John´s, NL

Print_ISBN

978-1-4799-4920-5

Type

conf

DOI

10.1109/OCEANS.2014.7003094

Filename

7003094