DocumentCode :
44796
Title :
Hyperspectral Image Segmentation Using a New Spectral Unmixing-Based Binary Partition Tree Representation
Author :
Veganzones, M.A. ; Tochon, Guillaume ; Dalla-Mura, Mauro ; Plaza, Antonio J. ; Chanussot, Jocelyn
Author_Institution :
Dept. Image & Signal, Grenoble-INP, St. Martin d´Hères, France
Volume :
23
Issue :
8
fYear :
2014
fDate :
Aug. 2014
Firstpage :
3574
Lastpage :
3589
Abstract :
The binary partition tree (BPT) is a hierarchical region-based representation of an image in a tree structure. The BPT allows users to explore the image at different segmentation scales. Often, the tree is pruned to get a more compact representation and so the remaining nodes conform an optimal partition for some given task. Here, we propose a novel BPT construction approach and pruning strategy for hyperspectral images based on spectral unmixing concepts. Linear spectral unmixing consists of finding the spectral signatures of the materials present in the image (endmembers) and their fractional abundances within each pixel. The proposed methodology exploits the local unmixing of the regions to find the partition achieving a global minimum reconstruction error. Results are presented on real hyperspectral data sets with different contexts and resolutions.
Keywords :
geophysical image processing; hyperspectral imaging; image representation; image segmentation; BPT construction approach; global minimum reconstruction error; hyperspectral data set; hyperspectral image segmentation; image region-based representation; linear spectral unmixing-based binary partition tree representation; pruning strategy; tree structure; Erbium; Hyperspectral imaging; Image segmentation; Materials; Merging; Vegetation; Binary partition trees; hyperspectral images; segmentation; spectral unmixing;
fLanguage :
English
Journal_Title :
Image Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1057-7149
Type :
jour
DOI :
10.1109/TIP.2014.2329767
Filename :
6828737
Link To Document :
بازگشت