Time filter

Source Type

Hyderabad andhra Pradesh, India

Das P.K.,Agricultural science and Applications Group | Laxman B.,Agricultural science and Applications Group | Rao S.V.C.K.,Agricultural science and Applications Group | Seshasai M.V.R.,Agricultural science and Applications Group | Dadhwal V.K.,Indian National Remote Sensing Centre
International Journal of Pest Management | Year: 2015

Bacterial leaf blight (BLB) is one of the most common diseases of rice in India, which may lead to partial or total crop loss based on the time of infestation. Hence, on-time identification of disease and its distribution over the affected region could provide useful information for minimizing the crop loss. In the present study, a comprehensive approach has been adopted to monitor the BLB-affected rice crop using hyperspectral data at filed scale and to upscale the observations at village-level using multispectral satellite data. The spectro-radiometer data at 350–2500 nm wavelength range was collected, along with other crop parameters, viz. chlorophyll and moisture content. The step-wise discriminate analysis (SDA) revealed that only four wavebands, i.e. 760, 990, 680 and 540 nm, could significantly discriminate diseased crop from healthy one. The selected wavebands were used to compute 12 narrowband vegetation indices, whereas according to SDA plant senescence index (PSRI), pigment-specific simple ratio (PSSRb) and red-edge position were only found to be effective. PSRI and PSSRb could be successfully re-computed using resolution simulation of hyperspectral data corresponding to linear imaging self-scanner (LISS) IV sensor. The equations were deployed on LISS IV satellite data to generate geospatial maps of PSRI and PPSRb. The geospatial maps could differentiate different degrees of stressed crop very effectively. Hence, the proposed approach can be adopted for in-season monitoring and assessment of diseased crops at regional level for better agricultural planning and management. © 2015 Taylor & Francis. Source

Discover hidden collaborations