Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-Validation
Mixture tuned matched filtering (MTMF) image classification capitalizes on the increasing spectral and spatial resolutions of available hyperspectral image data to identify the presence, and potentially the abundance, of a given cover type or endmember. Previous studies using MTMF have relied on ext...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/10/11/1675 |
| id |
doaj-037b40553aef42e6881d0c5fafed3161 |
|---|---|
| record_format |
Article |
| spelling |
doaj-037b40553aef42e6881d0c5fafed31612020-11-24T21:59:56ZengMDPI AGRemote Sensing2072-42922018-10-011011167510.3390/rs10111675rs10111675Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-ValidationDevin Routh0Lindsi Seegmiller1Charlie Bettigole2Catherine Kuhn3Chadwick D. Oliver4Henry B. Glick5Department of Environmental Systems Science, ETH Zürich, 8092 Zürich, SwitzerlandDivision of Geoinformatics, Department of Urban Planning and Environment, KTH Royal Institute of Technology, SE-114 28 Stockholm, SwedenSchool of Forestry and Environmental Studies, Yale University, New Haven, CT 06511, USASchool of Environmental and Forest Sciences, University of Washington, Seattle, WA 98195, USASchool of Forestry and Environmental Studies, Yale University, New Haven, CT 06511, USASchool of Forestry and Environmental Studies, Yale University, New Haven, CT 06511, USAMixture tuned matched filtering (MTMF) image classification capitalizes on the increasing spectral and spatial resolutions of available hyperspectral image data to identify the presence, and potentially the abundance, of a given cover type or endmember. Previous studies using MTMF have relied on extensive user input to obtain a reliable classification. In this study, we expand the traditional MTMF classification by using a selection of supervised learning algorithms with rigorous cross-validation. Our approach removes the need for subjective user input to finalize the classification, ultimately enhancing replicability and reliability of the results. We illustrate this approach with an MTMF classification case study focused on leafy spurge (<i>Euphorbia esula</i>), an invasive forb in Western North America, using free 30-m hyperspectral data from the National Aeronautics and Space Administration’s (NASA) Hyperion sensor. Our protocol shows for our data, a potential overall accuracy inflation between 18.4% and 30.8% without cross-validation and according to the supervised learning algorithm used. We propose this new protocol as a final step for the MTMF classification algorithm and suggest future researchers report a greater suite of accuracy statistics to affirm their classifications’ underlying efficacies.https://www.mdpi.com/2072-4292/10/11/1675mixture tuned matched filtering (MTMF)image classificationaccuracy assessmentpost-processing automationlinear unmixinghyperspectral remote sensingsupervised learningmachine learningleafy spurge |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Devin Routh Lindsi Seegmiller Charlie Bettigole Catherine Kuhn Chadwick D. Oliver Henry B. Glick |
| spellingShingle |
Devin Routh Lindsi Seegmiller Charlie Bettigole Catherine Kuhn Chadwick D. Oliver Henry B. Glick Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-Validation Remote Sensing mixture tuned matched filtering (MTMF) image classification accuracy assessment post-processing automation linear unmixing hyperspectral remote sensing supervised learning machine learning leafy spurge |
| author_facet |
Devin Routh Lindsi Seegmiller Charlie Bettigole Catherine Kuhn Chadwick D. Oliver Henry B. Glick |
| author_sort |
Devin Routh |
| title |
Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-Validation |
| title_short |
Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-Validation |
| title_full |
Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-Validation |
| title_fullStr |
Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-Validation |
| title_full_unstemmed |
Improving the Reliability of Mixture Tuned Matched Filtering Remote Sensing Classification Results Using Supervised Learning Algorithms and Cross-Validation |
| title_sort |
improving the reliability of mixture tuned matched filtering remote sensing classification results using supervised learning algorithms and cross-validation |
| publisher |
MDPI AG |
| series |
Remote Sensing |
| issn |
2072-4292 |
| publishDate |
2018-10-01 |
| description |
Mixture tuned matched filtering (MTMF) image classification capitalizes on the increasing spectral and spatial resolutions of available hyperspectral image data to identify the presence, and potentially the abundance, of a given cover type or endmember. Previous studies using MTMF have relied on extensive user input to obtain a reliable classification. In this study, we expand the traditional MTMF classification by using a selection of supervised learning algorithms with rigorous cross-validation. Our approach removes the need for subjective user input to finalize the classification, ultimately enhancing replicability and reliability of the results. We illustrate this approach with an MTMF classification case study focused on leafy spurge (<i>Euphorbia esula</i>), an invasive forb in Western North America, using free 30-m hyperspectral data from the National Aeronautics and Space Administration’s (NASA) Hyperion sensor. Our protocol shows for our data, a potential overall accuracy inflation between 18.4% and 30.8% without cross-validation and according to the supervised learning algorithm used. We propose this new protocol as a final step for the MTMF classification algorithm and suggest future researchers report a greater suite of accuracy statistics to affirm their classifications’ underlying efficacies. |
| topic |
mixture tuned matched filtering (MTMF) image classification accuracy assessment post-processing automation linear unmixing hyperspectral remote sensing supervised learning machine learning leafy spurge |
| url |
https://www.mdpi.com/2072-4292/10/11/1675 |
| work_keys_str_mv |
AT devinrouth improvingthereliabilityofmixturetunedmatchedfilteringremotesensingclassificationresultsusingsupervisedlearningalgorithmsandcrossvalidation AT lindsiseegmiller improvingthereliabilityofmixturetunedmatchedfilteringremotesensingclassificationresultsusingsupervisedlearningalgorithmsandcrossvalidation AT charliebettigole improvingthereliabilityofmixturetunedmatchedfilteringremotesensingclassificationresultsusingsupervisedlearningalgorithmsandcrossvalidation AT catherinekuhn improvingthereliabilityofmixturetunedmatchedfilteringremotesensingclassificationresultsusingsupervisedlearningalgorithmsandcrossvalidation AT chadwickdoliver improvingthereliabilityofmixturetunedmatchedfilteringremotesensingclassificationresultsusingsupervisedlearningalgorithmsandcrossvalidation AT henrybglick improvingthereliabilityofmixturetunedmatchedfilteringremotesensingclassificationresultsusingsupervisedlearningalgorithmsandcrossvalidation |
| _version_ |
1725846389596880896 |