Mapping agricultural crops with EO-1 Hyperion data

• Main Authors: Ntouros, K.D., Gitas, I.Z., Silleos, G.N.
• Format: Conference Proceeding
• Language: English
• Subjects: agriculture; Cotton; Crops; Crops classification; Earth; Hyperion; Hyperspectral imaging; Hyperspectral Remote Sensing; Image converters; Instruments; Minimum Noise Fraction; Reflectivity; Remote sensing; Satellites; Spectral analysis
• ISSN: 2158-6268; 2158-6276
• DOI: 10.1109/WHISPERS.2009.5289057

Hyperspectral data acquired by the Hyperion instrument, on board the Earth Observing - 1 Satellite (EO-1), were evaluated for the classification of five agricultural crops (maize (Zea mays), cotton (Gossypium hirsutum L.), rice (Oryza Sativa), tobacco (Nicotiana Tabacum) and tomato (Lycopersicon esculentum)) in Greece and the results were compared to classification of Landsat 5 TM data. In addition, was investigated the contribution of Hyperion SWIR bands on crops classification. The research was conducted in an agricultural area located in the North-Eastern Greece. The Hyperion radiance values, from the 196 bands, were converted into surface reflectance values using the Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes (FLAASH) model which is embedded in ENVI software. The data dimensionality reduction of Hyperion's image bands was achieved by using MNF transformation whereas the Maximum Likelihood algorithm was used in order to perform image data classification. The results showed that an overall accuracy of 91% was obtained from the classification of Hyperion image, while the overall accuracy resulted from the classification of Landsat 5 TM image was 81%. Also the Hyperion SWIR bands provide additional information on crops classification, not available by VNIR bands.