Clustering Based Band Selection
Classification of Hyperspectral Images
Versandkostenfrei!
Versandfertig in 6-10 Tagen
40,99 €
inkl. MwSt.
PAYBACK Punkte
20 °P sammeln!
Hyperspectral Images (HSIs) are popular in diversified applications, such as; Geo-sciences, Biomedical imaging, Agriculture, and physics-related research. The rich spatial and spectral information of HSI are the key factors for robust representation of class-specific objects, in remote sensing applications. But these images often suffer from the Hughes effect. This demands a dimensionality reduction using feature selection. The feature selection process is commonly called Band Selection (BS) for the HS dataset. This Book is mainly focused on three proposed models, where, mostly the clustering ...
Hyperspectral Images (HSIs) are popular in diversified applications, such as; Geo-sciences, Biomedical imaging, Agriculture, and physics-related research. The rich spatial and spectral information of HSI are the key factors for robust representation of class-specific objects, in remote sensing applications. But these images often suffer from the Hughes effect. This demands a dimensionality reduction using feature selection. The feature selection process is commonly called Band Selection (BS) for the HS dataset. This Book is mainly focused on three proposed models, where, mostly the clustering based unsupervised strategies are adopted for BS. First, Derivative-based band clustering and multi-agent PSO optimization for optimal band selection (DBC_MAPSO) is proposed. But they are time consuming and the selected bands are not persistent for each evaluation, due to the random nature of the optimizers. To overcome this, Spatial residual clustering and entropy-based ranking (SRC_EBR) and Featured clustering and ranking based bad cluster removal (FC_RBCR) are proposed.
Andere Kunden interessierten sich für
