Evaluating a Convolutional Neural Network and a Mosaic Image Database for Land Use Segmentation in the Brazilian Amazon Region

Joel Parente de Oliveira, Marly Costa, Cícero Filho

2021

Abstract

This study presents an image database and a convolutional neural network for the segmentation of land use in agriculture, forest and pasture classes. LANDSAT-8/OLI images from an area of the Brazilian Amazon region were used. The reference data were extracted from the results of the TerraClass project in 2014. The image database was generated in two versions: the first with six bands and the second with three bands. Each version of the data set has 4,000 images and size 400x400 pixels. Each image was generated using the mosaic technique. Each mosaic image is created from small agricultural, forest and grassland patches that are extracted from satellite images. The mosaic image is created with almost the same amount of agriculture, forest and pasture patches. The convolutional neural network architecture was evaluated together with three optimization methods: SGDM, ADAM and RMSProp and the dropout and L2 regularization for generalization improvement. The best model, CNN + optimization method + technique for generalization improvement, evaluated on the validation set, was used to segment some regions of the Amazon. The best results were obtained using the ADAM optimization method and L2 regularization. The accuracy values obtained for the evaluated images were above 94%.

Paper Citation

in Harvard Style

Parente de Oliveira J., Costa M. and Filho C. (2021). Evaluating a Convolutional Neural Network and a Mosaic Image Database for Land Use Segmentation in the Brazilian Amazon Region. In Proceedings of the 7th International Conference on Geographical Information Systems Theory, Applications and Management - Volume 1: GISTAM, ISBN 978-989-758-503-6, pages 165-172. DOI: 10.5220/0010509701650172

in Bibtex Style

@conference{gistam21,
author={Joel Parente de Oliveira and Marly Costa and Cícero Filho},
title={Evaluating a Convolutional Neural Network and a Mosaic Image Database for Land Use Segmentation in the Brazilian Amazon Region},
booktitle={Proceedings of the 7th International Conference on Geographical Information Systems Theory, Applications and Management - Volume 1: GISTAM,},
year={2021},
pages={165-172},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010509701650172},
isbn={978-989-758-503-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 7th International Conference on Geographical Information Systems Theory, Applications and Management - Volume 1: GISTAM,
TI - Evaluating a Convolutional Neural Network and a Mosaic Image Database for Land Use Segmentation in the Brazilian Amazon Region
SN - 978-989-758-503-6
AU - Parente de Oliveira J.
AU - Costa M.
AU - Filho C.
PY - 2021
SP - 165
EP - 172
DO - 10.5220/0010509701650172