8 March 2017 Hyperspectral feature mapping classification based on mathematical morphology
Author Affiliations +
Proceedings Volume 10255, Selected Papers of the Chinese Society for Optical Engineering Conferences held October and November 2016; 102552A (2017) https://doi.org/10.1117/12.2268431
Event: Selected Papers of the Chinese Society for Optical Engineering Conferences held October and November 2016, 2016, Jinhua, Suzhou, Chengdu, Xi'an, Wuxi, China
Abstract
This paper proposed a hyperspectral feature mapping classification algorithm based on mathematical morphology. Without the priori information such as spectral library etc., the spectral and spatial information can be used to realize the hyperspectral feature mapping classification. The mathematical morphological erosion and dilation operations are performed respectively to extract endmembers. The spectral feature mapping algorithm is used to carry on hyperspectral image classification. The hyperspectral image collected by AVIRIS is applied to evaluate the proposed algorithm. The proposed algorithm is compared with minimum Euclidean distance mapping algorithm, minimum Mahalanobis distance mapping algorithm, SAM algorithm and binary encoding mapping algorithm. From the results of the experiments, it is illuminated that the proposed algorithm’s performance is better than that of the other algorithms under the same condition and has higher classification accuracy.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Chang Liu, Junwei Li, Guangping Wang, Jingli Wu, "Hyperspectral feature mapping classification based on mathematical morphology", Proc. SPIE 10255, Selected Papers of the Chinese Society for Optical Engineering Conferences held October and November 2016, 102552A (8 March 2017); doi: 10.1117/12.2268431; https://doi.org/10.1117/12.2268431
PROCEEDINGS
8 PAGES


SHARE
Back to Top