Infrared Image Segmentation via Fast Fuzzy C-Means with Spatial Information

Author

Wu, Jin ; Li, Juan ; Liu, Jian ; Tian, Jinwen

Author_Institution

Coll. of Inf. Sci. & Engineer., Wuhan Univ. of Sci. & Technol.

fYear

2004

fDate

22-26 Aug. 2004

Firstpage

742

Lastpage

745

Abstract

Forward looking infrared (FLIR) image segmentation is crucial for automatic target recognition (ATR). This paper presents a new clustering algorithm for FLIR image segmentation. We combine the standard fuzzy c-means algorithm (FCM) with two-dimensional histogram of the image, and modify the membership function of the FCM for taking into account the spatial information of image data. Since the FCM is computationally cumbersome, we also present a fast and efficient implementation of the proposed method. Experiments show that the proposed method can segment the infrared image properly and fast, also has the better robust performance to noise

Keywords

fuzzy set theory; image segmentation; infrared imaging; pattern clustering; 2D histogram; FLIR image segmentation; automatic target recognition; clustering algorithm; forward looking infrared image segmentation; fuzzy c-means algorithm; image data; membership function; spatial information; Clustering algorithms; Educational institutions; Fuzzy control; Histograms; Image segmentation; Infrared detectors; Infrared imaging; Noise robustness; Pixel; Target recognition; clustering; fuzzy c means; infrared image segmentation; membership function; robust;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics and Biomimetics, 2004. ROBIO 2004. IEEE International Conference on

Conference_Location

Shenyang

Print_ISBN

0-7803-8614-8

Type

conf

DOI

10.1109/ROBIO.2004.1521874

Filename

1521874