Crowd-Driven Deep Learning Tracks Amazon Deforestation
The Amazon forests act as a global reserve for carbon, have very high biodiversity, and provide a variety of additional ecosystem services. These forests are, however, under increasing pressure, coming mainly from deforestation, despite the fact that accurate satellite monitoring is in place that pr...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/21/5204 |
| _version_ | 1797631338422992896 |
|---|---|
| author | Ian McCallum Jon Walker Steffen Fritz Markus Grau Cassie Hannan I-Sah Hsieh Deanna Lape Jen Mahone Caroline McLester Steve Mellgren Nolan Piland Linda See Gerhard Svolba Murray de Villiers |
| author_facet | Ian McCallum Jon Walker Steffen Fritz Markus Grau Cassie Hannan I-Sah Hsieh Deanna Lape Jen Mahone Caroline McLester Steve Mellgren Nolan Piland Linda See Gerhard Svolba Murray de Villiers |
| author_sort | Ian McCallum |
| collection | DOAJ |
| description | The Amazon forests act as a global reserve for carbon, have very high biodiversity, and provide a variety of additional ecosystem services. These forests are, however, under increasing pressure, coming mainly from deforestation, despite the fact that accurate satellite monitoring is in place that produces annual deforestation maps and timely alerts. Here, we present a proof of concept for rapid deforestation monitoring that engages the global community directly in the monitoring process via crowdsourcing while subsequently leveraging the power of deep learning. Offering no tangible incentives, we were able to sustain participation from more than 5500 active contributors from 96 different nations over a 6-month period, resulting in the crowd classification of 43,108 satellite images (representing around 390,000 km<sup>2</sup>). Training a suite of AI models with results from the crowd, we achieved an accuracy greater than 90% in detecting new and existing deforestation. These findings demonstrate the potential of a crowd–AI approach to rapidly detect and validate deforestation events. Our method directly engages a large, enthusiastic, and increasingly digital global community who wish to participate in the stewardship of the global environment. Coupled with existing monitoring systems, this approach could offer an additional means of verification, increasing confidence in global deforestation monitoring. |
| first_indexed | 2024-03-11T11:22:20Z |
| format | Article |
| id | doaj.art-7305c79c9e0548a8881b7c905a9b9ca7 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T11:22:20Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-7305c79c9e0548a8881b7c905a9b9ca72023-11-10T15:11:22ZengMDPI AGRemote Sensing2072-42922023-11-011521520410.3390/rs15215204Crowd-Driven Deep Learning Tracks Amazon DeforestationIan McCallum0Jon Walker1Steffen Fritz2Markus Grau3Cassie Hannan4I-Sah Hsieh5Deanna Lape6Jen Mahone7Caroline McLester8Steve Mellgren9Nolan Piland10Linda See11Gerhard Svolba12Murray de Villiers13International Institute for Applied Systems Analysis, 2361 Laxenburg, AustriaSAS Campus Drive, Cary, NC 27513, USAInternational Institute for Applied Systems Analysis, 2361 Laxenburg, AustriaSAS Institute AG, 8304 Wallisellen, SwitzerlandSAS Campus Drive, Cary, NC 27513, USASAS Campus Drive, Cary, NC 27513, USASAS Campus Drive, Cary, NC 27513, USASAS Campus Drive, Cary, NC 27513, USASAS Campus Drive, Cary, NC 27513, USASAS Campus Drive, Cary, NC 27513, USASAS Campus Drive, Cary, NC 27513, USAInternational Institute for Applied Systems Analysis, 2361 Laxenburg, AustriaSAS Austria, 1020 Vienna, AustriaSAS Netherlands B.V., 1272 PC Huizen, The NetherlandsThe Amazon forests act as a global reserve for carbon, have very high biodiversity, and provide a variety of additional ecosystem services. These forests are, however, under increasing pressure, coming mainly from deforestation, despite the fact that accurate satellite monitoring is in place that produces annual deforestation maps and timely alerts. Here, we present a proof of concept for rapid deforestation monitoring that engages the global community directly in the monitoring process via crowdsourcing while subsequently leveraging the power of deep learning. Offering no tangible incentives, we were able to sustain participation from more than 5500 active contributors from 96 different nations over a 6-month period, resulting in the crowd classification of 43,108 satellite images (representing around 390,000 km<sup>2</sup>). Training a suite of AI models with results from the crowd, we achieved an accuracy greater than 90% in detecting new and existing deforestation. These findings demonstrate the potential of a crowd–AI approach to rapidly detect and validate deforestation events. Our method directly engages a large, enthusiastic, and increasingly digital global community who wish to participate in the stewardship of the global environment. Coupled with existing monitoring systems, this approach could offer an additional means of verification, increasing confidence in global deforestation monitoring.https://www.mdpi.com/2072-4292/15/21/5204deforestationcrowdsourcingmachine learning |
| spellingShingle | Ian McCallum Jon Walker Steffen Fritz Markus Grau Cassie Hannan I-Sah Hsieh Deanna Lape Jen Mahone Caroline McLester Steve Mellgren Nolan Piland Linda See Gerhard Svolba Murray de Villiers Crowd-Driven Deep Learning Tracks Amazon Deforestation Remote Sensing deforestation crowdsourcing machine learning |
| title | Crowd-Driven Deep Learning Tracks Amazon Deforestation |
| title_full | Crowd-Driven Deep Learning Tracks Amazon Deforestation |
| title_fullStr | Crowd-Driven Deep Learning Tracks Amazon Deforestation |
| title_full_unstemmed | Crowd-Driven Deep Learning Tracks Amazon Deforestation |
| title_short | Crowd-Driven Deep Learning Tracks Amazon Deforestation |
| title_sort | crowd driven deep learning tracks amazon deforestation |
| topic | deforestation crowdsourcing machine learning |
| url | https://www.mdpi.com/2072-4292/15/21/5204 |
| work_keys_str_mv | AT ianmccallum crowddrivendeeplearningtracksamazondeforestation AT jonwalker crowddrivendeeplearningtracksamazondeforestation AT steffenfritz crowddrivendeeplearningtracksamazondeforestation AT markusgrau crowddrivendeeplearningtracksamazondeforestation AT cassiehannan crowddrivendeeplearningtracksamazondeforestation AT isahhsieh crowddrivendeeplearningtracksamazondeforestation AT deannalape crowddrivendeeplearningtracksamazondeforestation AT jenmahone crowddrivendeeplearningtracksamazondeforestation AT carolinemclester crowddrivendeeplearningtracksamazondeforestation AT stevemellgren crowddrivendeeplearningtracksamazondeforestation AT nolanpiland crowddrivendeeplearningtracksamazondeforestation AT lindasee crowddrivendeeplearningtracksamazondeforestation AT gerhardsvolba crowddrivendeeplearningtracksamazondeforestation AT murraydevilliers crowddrivendeeplearningtracksamazondeforestation |