A Survey on Fundamental Concepts and Practical Challenges of Hyperspectral images

Abstract

The Remote sensing provides a synoptic view of land by detecting the energy reflected from Earth's surface. The Hyperspectral images (HSI) use perfect sensors that extract more than a hundred of images, with more detailed information than using traditional Multispectral data. In this paper, we aim to study this aspect of communication in the case of passive reception. First, a brief overview of acquisition process and treatment of Hyperspectral images is provided. Then, we explain representation spaces and the various analysis methods of these images. Furthermore, the factors influencing this analysis are investigated and some applications, in this area, are presented. Finally, we explain the relationship between Hyperspectral images and Datamining and we outline the open issues related to this area. So we consider the case study: HSI AVIRIS 92AV3C. This study serves as map of route for integrating classification methods in the higher dimensionality data. Keywords-component: Hyperspectral images, Passive Sensing,Classification, Data mining.

Cited by in corpus (4)

• A novel information gain-based approach for classification and dimensionality reduction of hyperspectral images
• Supervised classification methods applied to airborne hyperspectral images: Comparative study using mutual information
• New wrapper method based on normalized mutual information for dimension reduction and classification of hyperspectral images
• A Novel Filter Approach for Band Selection and Classification of Hyperspectral Remotely Sensed Images Using Normalized Mutual Information and Support Vector Machines