{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,6]],"date-time":"2026-02-06T04:30:04Z","timestamp":1770352204253,"version":"3.49.0"},"reference-count":41,"publisher":"MDPI AG","issue":"15","license":[{"start":{"date-parts":[[2022,7,28]],"date-time":"2022-07-28T00:00:00Z","timestamp":1658966400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Climate Land Use Alliance","award":["G-2008-57051"],"award-info":[{"award-number":["G-2008-57051"]}]},{"name":"Climate Land Use Alliance","award":["#009\/2020"],"award-info":[{"award-number":["#009\/2020"]}]},{"name":"Fundo Vale","award":["G-2008-57051"],"award-info":[{"award-number":["G-2008-57051"]}]},{"name":"Fundo Vale","award":["#009\/2020"],"award-info":[{"award-number":["#009\/2020"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>This study presents our efforts to automate the detection of unofficial roads (herein, roads) in the Brazilian Amazon using artificial intelligence (AI). In this region, roads are built by loggers, goldminers, and unauthorized land settlements from existing official roads, expanding over pristine forests and leading to new deforestation and fire hotspots. Previous research used visual interpretation, hand digitization, and vector editing techniques to create a thorough Amazon Road Dataset (ARD) from Landsat imagery. The ARD allowed assessment of the road dynamics and impacts on deforestation, landscape fragmentation, and fires and supported several scientific and societal applications. This research used the existing ARD to train and model a modified U-Net algorithm to detect rural roads in the Brazilian Amazon using Sentinel-2 imagery from 2020 in the Azure Planetary Computer platform. Moreover, we implemented a post-AI detection protocol to connect and vectorize the U-Net road detected to create a new ARD. We estimated the recall and precision accuracy using an independent ARD dataset, obtaining 65% and 71%, respectively. Visual interpretation of the road detected with the AI algorithm suggests that the accuracy is underestimated. The reference dataset does not include all roads that the AI algorithm can detect in the Sentinel-2 imagery. We found an astonishing footprint of roads in the Brazilian Legal Amazon, with 3.46 million km of roads mapped in 2020. Most roads are in private lands (~55%) and 25% are in open public lands under land grabbing pressure. The roads are also expanding over forested areas with 41% cut or within 10 km from the roads, leaving 59% of the 3.1 million km2 of the remaining original forest roadless. Our AI and post-AI models fully automated road detection in rural areas of the Brazilian Amazon, making it possible to operationalize road monitoring. We are using the AI road map to understand better rural roads\u2019 impact on new deforestation, fires, and landscape fragmentation and to support societal and policy applications for forest conservation and regional planning.<\/jats:p>","DOI":"10.3390\/rs14153625","type":"journal-article","created":{"date-parts":[[2022,7,28]],"date-time":"2022-07-28T22:43:26Z","timestamp":1659048206000},"page":"3625","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":31,"title":["Mapping Roads in the Brazilian Amazon with Artificial Intelligence and Sentinel-2"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1674-6301","authenticated-orcid":false,"suffix":"Jr.","given":"Jonas","family":"Botelho","sequence":"first","affiliation":[{"name":"IMAZON-Amazon Institute of People and the Environment, Bel\u00e9m 66055-200, PA, Brazil"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4726-640X","authenticated-orcid":false,"given":"Stefany C. P.","family":"Costa","sequence":"additional","affiliation":[{"name":"IMAZON-Amazon Institute of People and the Environment, Bel\u00e9m 66055-200, PA, Brazil"}]},{"given":"J\u00falia G.","family":"Ribeiro","sequence":"additional","affiliation":[{"name":"IMAZON-Amazon Institute of People and the Environment, Bel\u00e9m 66055-200, PA, Brazil"}]},{"suffix":"Jr.","given":"Carlos M.","family":"Souza","sequence":"additional","affiliation":[{"name":"IMAZON-Amazon Institute of People and the Environment, Bel\u00e9m 66055-200, PA, Brazil"}]}],"member":"1968","published-online":{"date-parts":[[2022,7,28]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Sarhan, E., Khalifa, E., and Nabil, A. (2011, January 3\u20135). Road Extraction Framework by Using Cellular Neural Network from Remote Sensing Images. Proceedings of the 2011 International Conference on Image Information Processing, Shimla, India.","DOI":"10.1109\/ICIIP.2011.6108892"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"749","DOI":"10.1109\/LGRS.2018.2802944","article-title":"Road Extraction by Deep Residual U-Net","volume":"15","author":"Zhang","year":"2017","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"109","DOI":"10.1111\/j.1467-9787.2007.00502.x","article-title":"Road Investments, Spatial Intensification and Deforestation in the Brazilian Amazon","volume":"47","author":"Aldrich","year":"2007","journal-title":"J. Reg. Sci."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1047","DOI":"10.3390\/rs3051047","article-title":"Roads as Drivers of Change: Trajectories across the Tri-National Frontier in MAP, the Southwestern Amazon","volume":"3","author":"Southworth","year":"2011","journal-title":"Remote Sens."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.apgeog.2015.06.001","article-title":"Spatiotemporal Patterns of Tropical Deforestation and Forest Degradation in Response to the Operation of the Tucuru\u00ed Hydroelectric Dam in the Amazon Basin","volume":"63","author":"Chen","year":"2015","journal-title":"Appl. Geogr."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"532","DOI":"10.1071\/WF13106","article-title":"A Quantitative Study of the Proximity of Satellite Detected Active Fires to Roads and Rivers in the Brazilian Tropical Moist Forest Biome","volume":"23","author":"Kumar","year":"2014","journal-title":"Int. J. Wildland Fire"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1016\/j.biocon.2014.07.004","article-title":"Roads, Deforestation, and the Mitigating Effect of Protected Areas in the Amazon","volume":"177","author":"Barber","year":"2014","journal-title":"Biol. Conserv."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1080\/01431160500353841","article-title":"Mapping Unofficial Roads with Landsat Images: A New Tool to Improve the Monitoring of the Brazilian Amazon Rainforest","volume":"27","author":"Souza","year":"2006","journal-title":"Int. J. Remote Sens."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Nascimento, E.d.S., da Silva, S.S., Bordignon, L., de Melo, A.W.F., Brand\u00e3o, A., Souza, C.M., and Silva Junior, C.H.L. (2021). Roads in the Southwestern Amazon, State of Acre, between 2007 and 2019. Land, 10.","DOI":"10.3390\/land10020106"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Zhang, Y., Liu, J., Qian, X., Qiu, A., and Zhang, F. (2017). An Automatic Road Network Construction Method Using Massive GPS Trajectory Data. ISPRS Int. J. Geo-Inf., 6.","DOI":"10.3390\/ijgi6120400"},{"key":"ref_11","first-page":"102031","article-title":"Maintaining Accurate, Current, Rural Road Network Data: An Extraction and Updating Routine Using RapidEye, Participatory GIS and Deep Learning","volume":"87","author":"Kearney","year":"2020","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_12","first-page":"271","article-title":"A Review of Road Extraction from Remote Sensing Images","volume":"3","author":"Wang","year":"2016","journal-title":"J. Traffic Transp. Eng."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"699","DOI":"10.14358\/PERS.74.6.699","article-title":"The Fragmentation of Space in the Amazon Basin","volume":"74","author":"Arima","year":"2013","journal-title":"Photogramm. Eng. Remote Sens."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1353\/lag.0.0004","article-title":"Road Networks and Forest Fragmentation in the Amazon: Explanations for Local Differences with Implications for Conservation and Development","volume":"7","author":"Perz","year":"2008","journal-title":"J. Lat. Am. Geogr."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Rosa, I.M.D., Purves, D., Souza, C., and Ewers, R.M. (2013). Predictive Modelling of Contagious Deforestation in the Brazilian Amazon. PLoS ONE, 8.","DOI":"10.1371\/journal.pone.0077231"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"304","DOI":"10.1016\/j.spasta.2017.06.003","article-title":"A Spatiotemporal Geostatistical Hurdle Model Approach for Short-Term Deforestation Prediction","volume":"21","author":"Sales","year":"2017","journal-title":"Spat. Stat."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1889","DOI":"10.1098\/rstb.2007.0017","article-title":"Road Building, Land Use and Climate Change: Prospects for Environmental Governance in the Amazon","volume":"363","author":"Perz","year":"2008","journal-title":"Philos. Trans. R. Soc. B Biol. Sci."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"529","DOI":"10.1111\/j.1467-7660.2007.00422.x","article-title":"Unofficial Road Building in the Amazon: Socioeconomic and Biophysical Explanations","volume":"38","author":"Perz","year":"2007","journal-title":"Dev. Chang."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"20141742","DOI":"10.1098\/rspb.2014.1742","article-title":"Road Networks Predict Human Influence on Amazonian Bird Communities","volume":"281","author":"Ahmed","year":"2014","journal-title":"Proc. R. Soc. B Biol. Sci."},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Xu, Y., Xie, Z., Feng, Y., and Chen, Z. (2018). Road Extraction from High-Resolution Remote Sensing Imagery Using Deep Learning. Remote Sens., 10.","DOI":"10.3390\/rs10091461"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Stewart, C., Lazzarini, M., Luna, A., and Albani, S. (2020). Deep Learning with Open Data for Desert Road Mapping. Remote Sens., 12.","DOI":"10.3390\/rs12142274"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Gao, L., Song, W., Dai, J., and Chen, Y. (2019). Road Extraction from High-Resolution Remote Sensing Imagery Using Refined Deep Residual Convolutional Neural Network. Remote Sens., 11.","DOI":"10.3390\/rs11050552"},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Augustauskas, R., and Lipnickas, A. (2020). Improved Pixel-Level Pavement-Defect Segmentation Using a Deep Autoencoder. Sensors, 20.","DOI":"10.3390\/s20092557"},{"key":"ref_25","first-page":"257","article-title":"Explaining the Fragmentation in the Brazilian Amazonian Forest","volume":"11","author":"Arima","year":"2016","journal-title":"J. Land Use Sci."},{"key":"ref_26","unstructured":"Funda\u00e7\u00e3o Instituto Brasileiro de Geografia e Estat\u00edstica. Biblioteca Central, and Funda\u00e7\u00e3o Instituto Brasileiro de Geografia e Estat\u00edstica (2022, March 30). Ger\u00eancia de Documenta\u00e7\u00e3o e Biblioteca. Mapas e Outros Materiais Cartogr\u00e1ficas na Biblioteca Central Do IBGE, Available online: https:\/\/biblioteca.ibge.gov.br\/biblioteca-catalogo.html?id=213675&view=detalhes."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"034057","DOI":"10.1088\/1748-9326\/ab76db","article-title":"Unmasking Secondary Vegetation Dynamics in the Brazilian Amazon","volume":"15","author":"Nunes","year":"2020","journal-title":"Environ. Res. Lett."},{"key":"ref_28","unstructured":"Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., and Isard, M. (2016, January 2\u20134). TensorFlow: A System for Large-Scale Machine Learning. Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2016, Savannah, GA, USA."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Albawi, S., Mohammed, T.A., and Al-Zawi, S. (2017, January 21\u201323). Understanding of a Convolutional Neural Network. Proceedings of the 2017 International Conference on Engineering and Technology (ICET), Antalya, Turkey.","DOI":"10.1109\/ICEngTechnol.2017.8308186"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Saraiva, M., Protas, \u00c9., Salgado, M., and Souza, C. (2020). Automatic Mapping of Center Pivot Irrigation Systems from Satellite Images Using Deep Learning. Remote Sens., 12.","DOI":"10.3390\/rs12030558"},{"key":"ref_31","unstructured":"Maas, A.L., Hannun, A.Y., and Ng, A.Y. (2013, January 16). Rectifier Nonlinearities Improve Neural Network Acoustic Models. Proceedings of the ICML Workshop on Deep Learning for Audio, Speech and Language Processing 2013, Atlanta, GA, USA."},{"key":"ref_32","first-page":"1929","article-title":"Dropout: A Simple Way to Prevent Neural Networks from Overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"240","DOI":"10.1007\/978-3-319-67558-9_28","article-title":"Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations","volume":"2017","author":"Sudre","year":"2017","journal-title":"Deep. Learn. Med. Image Anal. Multimodal. Learn. Clin. Decis. Support"},{"key":"ref_34","unstructured":"Dozat, T. (2016, January 2\u20134). Incorporating Nesterov Momentum into Adam. Proceedings of the ICLR Workshop, San Juan, Puerto Rico."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"950","DOI":"10.1111\/tgis.12068","article-title":"How to Apply the Geospatial Data Abstraction Library (GDAL) Properly to Parallel Geospatial Raster I\/O?","volume":"18","author":"Qin","year":"2014","journal-title":"Trans. GIS"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"124","DOI":"10.1016\/j.envsoft.2011.11.014","article-title":"GRASS GIS: A Multi-Purpose Open Source GIS","volume":"31","author":"Neteler","year":"2012","journal-title":"Environ. Model. Softw."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1016\/j.rse.2003.12.007","article-title":"Sources of Error in Accuracy Assessment of Thematic Land-Cover Maps in the Brazilian Amazon","volume":"90","author":"Powell","year":"2004","journal-title":"Remote Sens. Environ."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"45","DOI":"10.3389\/fenvs.2015.00045","article-title":"A Survey of Remote-Sensing Big Data","volume":"3","author":"Liu","year":"2015","journal-title":"Front. Environ. Sci."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Parente, L., Taquary, E., Silva, A.P., Souza, C., and Ferreira, L. (2019). Next Generation Mapping: Combining Deep Learning, Cloud Computing, and Big Remote Sensing Data. Remote Sens., 11.","DOI":"10.3390\/rs11232881"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Souza, C.M., Shimbo, J.Z., Rosa, M.R., Parente, L.L., Alencar, A.A., Rudorff, B.F.T., Hasenack, H., Matsumoto, M., Ferreira, L.G., and Souza-Filho, P.W.M. (2020). Reconstructing Three Decades of Land Use and Land Cover Changes in Brazilian Biomes with Landsat Archive and Earth Engine. Remote Sens., 12.","DOI":"10.3390\/rs12172735"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1016\/j.biocon.2011.08.024","article-title":"Dynamic Performance Assessment of Protected Areas","volume":"149","author":"Barber","year":"2012","journal-title":"Biol. Conserv."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/14\/15\/3625\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T23:58:45Z","timestamp":1760140725000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/14\/15\/3625"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,7,28]]},"references-count":41,"journal-issue":{"issue":"15","published-online":{"date-parts":[[2022,8]]}},"alternative-id":["rs14153625"],"URL":"https:\/\/doi.org\/10.3390\/rs14153625","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,7,28]]}}}