Mapping Mosaic Virus in Sugarcane Based on Hyperspectral Images

Author(s)

Moriya, Erika Akemi Saito ; Imai, Nilton Nobuhiro ; Tommaselli, Antonio Maria Garcia ; Miyoshi, Gabriela Takahashi

• Document type: Journal article
• Source: IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING; v. 10, n. 2, p. 740-748, FEB 2017.
• Web of Science Citations: 10

Abstract

The aim of this research was to develop a methodology involving aerial surveying using an unmanned aerial system (UAS), processing and analysis of images obtained by a hyperspectral camera, achieving results that enable discrimination and recognition of sugarcane plants infected with mosaic virus. It was necessary to characterize the spectral response of healthy and infected sugarcane plants in order to define the correct mode of operation for the hyperspectral camera, which provides many spectral band options for imaging but limits each image to 25 spectral bands. Spectral measurements of the leaves of infected and healthy sugarcane with a spectroradiometer were used to produce a spectral library. Once the most appropriate spectral bands had been selected, it was possible to configure the camera and carry out aerial surveying. The empirical line approach was adopted to obtain hemispherical conical reflectance factor values with a radiometric block adjustment to produce a mosaic suitable for the analysis. A classification based on spectral information divergence was applied and the results were evaluated by Kappa statistics. Areas of sugarcane infected with mosaic were identified from these hyperspectral images acquired by UAS and the results obtained had a high degree of accuracy. (AU)

• FAPESP's process: 13/50426-4 - Unmanned airborne vehicle-based 4D remote sensing for mapping Rain Forest biodiversity and its change in Brazil (UAV_4D_Bio)
• Grantee: Antonio Maria Garcia Tommaselli
• Support Opportunities: Regular Research Grants