Using SPOT-5 images in rice farming for detecting BPH (Brown Plant Hopper)

Abstract

Infestation of rice plant-hopper such as Brown Plant Hopper (BPH) (Nilaparvata lugens) is one of the most notable risk in rice yield in tropical areas especially in Asia. In order to use visible and infrared images to detect stress in rice production caused by BPH infestation, several remote sensing techniques have been developed. Initial recognition of pest infestation by means of remote sensing will spreads, for precision farming practice. To address this issue, detection of sheath blight in rice farming was examined by using SPOT-5 images. Specific image indices such as Normalized decrease food production costs, limit environmental hazards, and enhance natural pest control before the problem Normalized Difference Vegetation Index (NDVI), Standard difference indices (SDI) and Ratio Vegetation Index (RVI) were used for analyses using ENVI 4.8 and SPSS software. Results showed that all the indices to recognize infected plants are significant at α = 0.01. Examination of the association between the disease indices indicated that band 3 (near infrared) and band 4 (mid infrared) have a relatively high correlation. The selected indices declared better association for detecting healthy plants from diseased ones. Consequently, these sorts of indices especially NDVI could be valued as indicators for developing techniques for detecting the sheath blight of rice by using remote sensing. This infers that they are useful for crop disease detection but the spectral resolution is probably not sufficient to distinguish plants with light infections (low severity level). Using the index as an indicator can clarify the threshold for zoning the outbreaks. Quick assessment information is very useful in precision farming to practice site specific management such as pesticide application.