Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands
High-resolution images obtained by multispectral cameras mounted on Unmanned Aerial Vehicles (UAVs) are helping to capture the heterogeneity of the environment in images that can be discretized in categories during a classification process. Currently, there is an increasing use of supervised machine...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/18/3669 |
| _version_ | 1797517368940822528 |
|---|---|
| author | Ricardo Martínez Prentice Miguel Villoslada Peciña Raymond D. Ward Thaisa F. Bergamo Chris B. Joyce Kalev Sepp |
| author_facet | Ricardo Martínez Prentice Miguel Villoslada Peciña Raymond D. Ward Thaisa F. Bergamo Chris B. Joyce Kalev Sepp |
| author_sort | Ricardo Martínez Prentice |
| collection | DOAJ |
| description | High-resolution images obtained by multispectral cameras mounted on Unmanned Aerial Vehicles (UAVs) are helping to capture the heterogeneity of the environment in images that can be discretized in categories during a classification process. Currently, there is an increasing use of supervised machine learning (ML) classifiers to retrieve accurate results using scarce datasets with samples with non-linear relationships. We compared the accuracies of two ML classifiers using a pixel and object analysis approach in six coastal wetland sites. The results show that the Random Forest (RF) performs better than K-Nearest Neighbors (KNN) algorithm in the classification of pixels and objects and the classification based on pixel analysis is slightly better than the object-based analysis. The agreement between the classifications of objects and pixels is higher in Random Forest. This is likely due to the heterogeneity of the study areas, where pixel-based classifications are most appropriate. In addition, from an ecological perspective, as these wetlands are heterogeneous, the pixel-based classification reflects a more realistic interpretation of plant community distribution. |
| first_indexed | 2024-03-10T07:15:42Z |
| format | Article |
| id | doaj.art-ad878607fbd84ea2afcf8d0f05e0fdfe |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T07:15:42Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-ad878607fbd84ea2afcf8d0f05e0fdfe2023-11-22T15:06:34ZengMDPI AGRemote Sensing2072-42922021-09-011318366910.3390/rs13183669Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal WetlandsRicardo Martínez Prentice0Miguel Villoslada Peciña1Raymond D. Ward2Thaisa F. Bergamo3Chris B. Joyce4Kalev Sepp5Institute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaCentre for Aquatic Environments, School of the Environment and Technology, University of Brighton, Cockcroft Building, Moulsecoomb, Brighton BN2 4GJ, UKInstitute of Agriculture and Environmental Sciences, Estonian University of Life Sciences, Kreutzwaldi 5, EE-51006 Tartu, EstoniaHigh-resolution images obtained by multispectral cameras mounted on Unmanned Aerial Vehicles (UAVs) are helping to capture the heterogeneity of the environment in images that can be discretized in categories during a classification process. Currently, there is an increasing use of supervised machine learning (ML) classifiers to retrieve accurate results using scarce datasets with samples with non-linear relationships. We compared the accuracies of two ML classifiers using a pixel and object analysis approach in six coastal wetland sites. The results show that the Random Forest (RF) performs better than K-Nearest Neighbors (KNN) algorithm in the classification of pixels and objects and the classification based on pixel analysis is slightly better than the object-based analysis. The agreement between the classifications of objects and pixels is higher in Random Forest. This is likely due to the heterogeneity of the study areas, where pixel-based classifications are most appropriate. In addition, from an ecological perspective, as these wetlands are heterogeneous, the pixel-based classification reflects a more realistic interpretation of plant community distribution.https://www.mdpi.com/2072-4292/13/18/3669UAVmachine learningRandom ForestKNNclassificationcomparison |
| spellingShingle | Ricardo Martínez Prentice Miguel Villoslada Peciña Raymond D. Ward Thaisa F. Bergamo Chris B. Joyce Kalev Sepp Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands Remote Sensing UAV machine learning Random Forest KNN classification comparison |
| title | Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands |
| title_full | Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands |
| title_fullStr | Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands |
| title_full_unstemmed | Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands |
| title_short | Machine Learning Classification and Accuracy Assessment from High-Resolution Images of Coastal Wetlands |
| title_sort | machine learning classification and accuracy assessment from high resolution images of coastal wetlands |
| topic | UAV machine learning Random Forest KNN classification comparison |
| url | https://www.mdpi.com/2072-4292/13/18/3669 |
| work_keys_str_mv | AT ricardomartinezprentice machinelearningclassificationandaccuracyassessmentfromhighresolutionimagesofcoastalwetlands AT miguelvillosladapecina machinelearningclassificationandaccuracyassessmentfromhighresolutionimagesofcoastalwetlands AT raymonddward machinelearningclassificationandaccuracyassessmentfromhighresolutionimagesofcoastalwetlands AT thaisafbergamo machinelearningclassificationandaccuracyassessmentfromhighresolutionimagesofcoastalwetlands AT chrisbjoyce machinelearningclassificationandaccuracyassessmentfromhighresolutionimagesofcoastalwetlands AT kalevsepp machinelearningclassificationandaccuracyassessmentfromhighresolutionimagesofcoastalwetlands |