1 April 2016 Bands selection and classification of hyperspectral images based on hybrid kernels SVM by evolutionary algorithm
Author Affiliations +
Proceedings Volume 9796, Selected Papers of the Photoelectronic Technology Committee Conferences held November 2015; 979616 (2016) https://doi.org/10.1117/12.2229875
Event: Selected Proceedings of the Chinese Society for Optical Engineering Conferences held November 2015, 2015, Various, China
Abstract
The hyperspectral images(HSI) consist of many closely spaced bands carrying the most object information. While due to its high dimensionality and high volume nature, it is hard to get satisfactory classification performance. In order to reduce HSI data dimensionality preparation for high classification accuracy, it is proposed to combine a band selection method of artificial immune systems (AIS) with a hybrid kernels support vector machine (SVM-HK) algorithm. In fact, after comparing different kernels for hyperspectral analysis, the approach mixed radial basis function kernel (RBF-K) with sigmoid kernel (Sig-K) and applied the optimized hybrid kernels in SVM classifiers. Then the SVM-HK algorithm used to induce the bands selection of an improved version of AIS. The AIS was composed of clonal selection and elite antibody mutation, including evaluation process with optional index factor (OIF). Experimental classification performance was on a San Diego Naval Base acquired by AVIRIS, the HRS dataset shows that the method is able to efficiently achieve bands redundancy removal while outperforming the traditional SVM classifier.
© (2016) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yan-Yan Hu, Yan-Yan Hu, Dong-Sheng Li, Dong-Sheng Li, } "Bands selection and classification of hyperspectral images based on hybrid kernels SVM by evolutionary algorithm", Proc. SPIE 9796, Selected Papers of the Photoelectronic Technology Committee Conferences held November 2015, 979616 (1 April 2016); doi: 10.1117/12.2229875; https://doi.org/10.1117/12.2229875
PROCEEDINGS
10 PAGES


SHARE
Back to Top