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.

There is a rapidly growing interest in methods for automatic plant identification in agricultural research. Cotton (Gossypium spp.) is a crop well-suited to precision agriculture and its inherent goals of increasing yields while minimizing environmental impacts. Ten cotton (G. hirsutum and G. barbadense) cultivars with differing leaf characteristics were evaluated in a greenhouse environment. Hyperspectral data collected with a handheld spectroradiometer were used to distinguish among the cultivars. The features extracted by principal component analysis and stepwise selection approaches were used for discriminant analysis. The best discrimination accuracy by selected wavelengths was 90.4% for G. hirsutum cultivars, 100% for G. barbadense cultivars, and 91.6% for pooled cultivars of the two species. Spectral wavelengths at 550 and 760 nm were most relevant to the discrimination between these two cotton species. Two vegetation indices, NDVI and PRI, were also investigated for any significant differences across cotton cultivars. The results demonstrated that hyperspectral radiometry has good potential for discrimination of G. hirsutum and G. barbadense cotton cultivars in early stages of growth.