Segmentation of land use maps for precision agriculture

• Wilian F. Costa
• Michel J. M. Bieleveld
• Raphael G. Chinchilla
• Antonio M. Saraiva

DOI: https://doi.org/10.5753/wcama.2017.3438

Resumo

We present a multilayer image segmenter adapted to be used for Precision Agriculture (PA). PA depends strongly on the application of information technologies to divide and group geographic areas based on land use, soil data, metrological data and agricultural resources for planning and implementation of activities to increase output by using optimal strategies for each segment. We implemented a modified Baatz algorithm in the statistical language R and speed sensitive code was implemented in C++. The code will be made publicly available under the GNU Lesser Public License. We show the merit of our approach at the hand of a landscape and discuss the obtained segments generated by our tool.