American Society of Agricultural and Biological Engineers



Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Using digital image classification methods and hyperspectral imaging to detect leaf blast in rice

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  Paper number  131595564,  2013 Kansas City, Missouri, July 21 - July 24, 2013. (doi: http://dx.doi.org/10.13031/aim.20131595564) @2013
Authors:   Lulu WU, Long QI, Xu MA, Zhixiong ZHENG, Guorui CHEN, Zehua LI, Junfeng XIE
Keywords:   Leaf blast Spectral Angle Mapper Minimum Distance Classification NDVI Classification

Abstract. A hyperspectral imaging system has been applied within the wavelength range of 370–987 nm for detecting leaf blast in rice. Hyperspectral image was measured to determine the spectral of leaf blast infection. The spectra of lesion edge and lesion center were collected by region pixels of image. The reflection of rice blast lesion was higher in red wavelength from 550nm to 720nm than the healthy leaf. The lesion center was higher in red wavelength from 600nm to 680nm than the lesion edge. Leaf blast was identified from images by applying Spectral Angle Mapper (SAM), Minimum Distance Classification (MDC) and Normalized Difference Vegetation Index (NDVI) methods. In SAM, the maximum angle was set in the range of 0.10 to 0.16. MDC took the same reference samples as SAM method. Reflection at 680nm and 820nm was used to calculate NDVI. By density segmentation, blast lesion can be separated. The compared results showed that SAM would be more precise to detect the lesion of blast on leaf.

(Download PDF)    (Export to EndNotes)