Mapping sugarcane yellow leaf disease affected area using remote sensing technique

Abstract

Remote sensing provides information on coverage, mapping and classification of land-cover features, such as vegetation, soil, water and forests. Satellite imageries are being used to map areas under different crops and cultivars, and to identify areas with specific characteristics like deficiency symptoms, pest and disease infestation, etc., using the spectral reflectance at canopy and pixel scales. In an attempt to identify and map sugarcane fields affected by the Yellow Leaf Disease (YLD), sugarcane growing fields in the Bhavani command areas of Erode district of Tamil Nadu State, India, were surveyed. The digital satellite imagery of IRS P6, Path 100 and Row 65 which covers the surveyed area was obtained from National Remote Sensing Center (NRSC), Hyderabad, India, and the sugarcane fields with and without YLD were demarcated. The Digital Number (DN) values were extracted using the customized programme developed in C# language of Visual Studio 2008 in the dot net platform. The DN values in the YLD affected fields were lower than the YLD free field in all the four bands. The DN values for the four bands were analysed using the bootstrap confidence intervals. The differences in DN values were compared using the difference of mean (dAVE) which is one of the measures of difference used in unpaired data, where no dependence is assumed between the two groups of data. A confidence interval for dAVE provides quantitative information, which also includes a statistical test (by looking whether it contains zero) but is not restricted to it. The healthy sugarcane field gave significantly higher DN values than YLD affected fields. The DN values in the healthy sugarcane field were higher by 7.5 units with 95% BCa confidence interval (1, 30 DN values) when compared to YLD affected fields. The results reveal the possibility of developing a subpixel classification model to the highest classification accuracies for demarcating the YLD affected sugarcane fields.