PERPENDICULAR CROP ENHANCEMENT INDEX: A NEW APPROACH TO SOYBEAN MONITORING USING TIME-SERIES

In Brazil, despite the improvements with respect the technological knowledge, agricultural areas are often estimated in loco. Here, soybean areas in Paraná, Brazil, using MODIS imagery were mapped. We applied the vegetation index PCEI (Perpendicular Crop Enhancement Index) and threshold determination for the automation of soybean area discrimination by geo-object (GEOBIA). For this, vegetation indices (NDVI, EVI and CEI) and the development of the PCEI were used with the aid of time-series images from the TERRA/MODIS. By geo-objects and decision tree based on data mining support analysis, the new vegetation index was determined. Kappa and Overall Accuracy statistics were applied to evaluate classification precision. Regarding the ground line, R and R² were above 0.92 and 0.84, respectively (p<0.01). The test results indicate that the proposed methodology is efficient for mapping soybean distribution. Thus, this study allows automated mapping of with soybean crops areas at large scales.