Publication: Object-based dynamic time warping for crop mapping
A new paper is now published in Remote Sensing journal (open-access). It uses Sentinel-2 time series of vegetation indices to extract temporal pattern for crops in highly managed agricultural areas in Texas and California. It uses the dynamic time warping algorithm to match the highly heterogeneous patterns of crops and test multiple time constrained on DTW. We applied the analysis on objects derived using the multiresolution segmentation and implemented the entire workflow as an easy-to-use eCognition customized algorithm.
You can find the open-access paper here:
Csillik, O.; Belgiu, M.; Asner, G.P.; Kelly, M. Object-Based Time-Constrained Dynamic Time Warping Classification of Crops Using Sentinel-2. Remote Sensing 2019, 11, 1257.
Have a look at the tool and demo data to perform dynamic time warping classification of crops using objects as spatial analysis units and let me know your valuable feedback!