A segment-based approach to classify agricultural lands by using multi-temporal optical and microwave data

This research study aims to classify crop diversity in agricultural land with a segment-based approach using multi-temporal Kompsat-2 and Environmental Satellite (Envisat) advanced synthetic aperture radar (ASAR) data acquired in June, July and August on Karacabey Plain, Turkey. Analyses start with the image segmentation process applied to the fused optical images to search homogenous objects. The segmentation outputs are evaluated using multiple goodness measures, which take into consideration area and location similarities. Image classifications are performed on each multispectral (MS) single date image. In order to combine the most probable classes of the thematic maps, distance maps are generated. Evaluations of the thematic maps are performed through confusion matrices based on pixel-based and segment-based approaches. The results indicate that the highest overall accuracy of 88.71% and a kappa result of 0.86 are provided for the segment-based approach of the combined thematic map along with the microwave data, which is around 10% higher than the related pixel-based results.