Phenology-Based Classification of Crop Fields Using Cross-Correlation: A Case Study

We investigated the use of phenological information extracted from satellite imagery in accurate crop classification. Vegetation indices (VI) extracted from Sentinel-2 imagery are capable to track the vegetation development through the year and from them the phenological profile can be retrieved and introduced into a multi-temporal automatic classification process to detect crop fields and to discriminate among different crop species. The matching and discrimination between phenology was evaluated by means of cross-correlation. Our case study is the Narok county located within the Great Rift Valley of Kenya.