Compressive sensing for reconstruction, classification, and detection of hyperspectral images

Hyperspectral imagery (HSI) acquired by remote-sensing systems (e.g., a spaceborne or an airborne sensor) typically records reflectance values over a wide region of the electromagnetic spectrum [1,2]. Each pixel in an HSI represents many contiguous and narrow spectral bands (e.g., a spectral range of 0.4 to 2.4 µm). This enables HSI to potentially provide rich information about the materials in the image scene. HSI has become a widely available modality used in a wide variety of applications, including urban-growth analysis, biological and chemical detection, environmental monitoring, mineral exploration, etc.