Estimation of cropped area and grain yield of rice using remote sensing data

Abstract

Studies for estimating cultivated area and yield of rice using remote sensing data acquired from the Landsat Thematic Mapper (TM) system with six bands (1, 2, 3,4, 5,7) were conducted. Three images taken on different days were visually and digitally analysed for the estimation of rice cultivated area. This was obtained with over 90 per cent classification accuracy. False colour composite of 5R, 4G, 3B was selected for the identification of rice cultivated area at maturity stage. Band combination of 1, 3, 4 and 5 was selected as an appropriate subset in the digital analysis for the estimation of rice areas. An attempt was made to develop a relationship between reflectance values and actual grain yield using an image of the fully ripened stage of the crop. High correlation was observed with reflectance values of some bands and yields. A computerized plant process model was adopted for the simulation of rice growth and yield. This was used for estimating yield per unit area.