Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation

Abstract: 

We propose a semi-supervised algorithm for processing and classification of hyperspectral imagery. For initialization, we keep 20% of the data intact, and use Principal Component Analysis to discard voxels from noisier bands and pixels. Then, we use either an Accelerated Proximal Gradient algorithm (APGL), or a modified APGL algorithm with a penalty term for distance between inpainted pixels and endmembers (APGL Hyp), on the initialized datacube to inpaint the missing data. APGL and APGL Hyp are distinguished by performance on datasets with full pixels removed or extreme noise. This inpainting technique results in band-by-band datacube sharpening and removal of noise from individual spectral signatures. We can also classify the inpainted cube by assigning each pixel to its nearest endmember via Euclidean distance. We demonstrate improved accuracy in classification over data-mining techniques like k-means, unmixing techniques like Hierarchical Non-Negative Matrix Factorization, and graph-based methods like Non-Local Total Variation.

up
0 users have voted:

Paper Details

Authors:
Victoria Chayes, Kevin Miller, Rasika Bhalerao, Jiajie Luo, Wei Zhu, Andrea L. Bertozzi, Wenzhi Liao, Stanley Osher
Submitted On:
6 March 2017 - 2:08pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Victoria Chayes
Paper Code:
SS-L4.4
Document Year:
2017
Cite

Document Files

ICASSP Presentation.pdf

(114 downloads)

Subscribe

[1] Victoria Chayes, Kevin Miller, Rasika Bhalerao, Jiajie Luo, Wei Zhu, Andrea L. Bertozzi, Wenzhi Liao, Stanley Osher, "Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1659. Accessed: Dec. 16, 2017.
@article{1659-17,
url = {http://sigport.org/1659},
author = {Victoria Chayes; Kevin Miller; Rasika Bhalerao; Jiajie Luo; Wei Zhu; Andrea L. Bertozzi; Wenzhi Liao; Stanley Osher },
publisher = {IEEE SigPort},
title = {Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation},
year = {2017} }
TY - EJOUR
T1 - Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation
AU - Victoria Chayes; Kevin Miller; Rasika Bhalerao; Jiajie Luo; Wei Zhu; Andrea L. Bertozzi; Wenzhi Liao; Stanley Osher
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1659
ER -
Victoria Chayes, Kevin Miller, Rasika Bhalerao, Jiajie Luo, Wei Zhu, Andrea L. Bertozzi, Wenzhi Liao, Stanley Osher. (2017). Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation. IEEE SigPort. http://sigport.org/1659
Victoria Chayes, Kevin Miller, Rasika Bhalerao, Jiajie Luo, Wei Zhu, Andrea L. Bertozzi, Wenzhi Liao, Stanley Osher, 2017. Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation. Available at: http://sigport.org/1659.
Victoria Chayes, Kevin Miller, Rasika Bhalerao, Jiajie Luo, Wei Zhu, Andrea L. Bertozzi, Wenzhi Liao, Stanley Osher. (2017). "Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation." Web.
1. Victoria Chayes, Kevin Miller, Rasika Bhalerao, Jiajie Luo, Wei Zhu, Andrea L. Bertozzi, Wenzhi Liao, Stanley Osher. Pre-processing And Classification Of Hyperspectral Imagery Via Selective Inpainting Presentation [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1659