Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble Model
Traditional methods for individual tree-crown (ITC) detection (image classification, segmentation, template matching, etc.) applied to very high-resolution remote sensing imagery have been shown to struggle in disparate landscape types or image resolutions due to scale problems and information compl...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-07-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/15/2426 |
| _version_ | 1797560985481904128 |
|---|---|
| author | Alin-Ionuț Pleșoianu Mihai-Sorin Stupariu Ionuț Șandric Ileana Pătru-Stupariu Lucian Drăguț |
| author_facet | Alin-Ionuț Pleșoianu Mihai-Sorin Stupariu Ionuț Șandric Ileana Pătru-Stupariu Lucian Drăguț |
| author_sort | Alin-Ionuț Pleșoianu |
| collection | DOAJ |
| description | Traditional methods for individual tree-crown (ITC) detection (image classification, segmentation, template matching, etc.) applied to very high-resolution remote sensing imagery have been shown to struggle in disparate landscape types or image resolutions due to scale problems and information complexity. Deep learning promised to overcome these shortcomings due to its superior performance and versatility, proven with reported detection rates of ~90%. However, such models still find their limits in transferability across study areas, because of different tree conditions (e.g., isolated trees vs. compact forests) and/or resolutions of the input data. This study introduces a highly replicable deep learning ensemble design for ITC detection and species classification based on the established single shot detector (SSD) model. The ensemble model design is based on varying the input data for the SSD models, coupled with a voting strategy for the output predictions. Very high-resolution unmanned aerial vehicles (UAV), aerial remote sensing imagery and elevation data are used in different combinations to test the performance of the ensemble models in three study sites with highly contrasting spatial patterns. The results show that ensemble models perform better than any single SSD model, regardless of the local tree conditions or image resolution. The detection performance and the accuracy rates improved by 3–18% with only as few as two participant single models, regardless of the study site. However, when more than two models were included, the performance of the ensemble models only improved slightly and even dropped. |
| first_indexed | 2024-03-10T18:07:59Z |
| format | Article |
| id | doaj.art-495a5656ef50420d820dbad83e82db28 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T18:07:59Z |
| publishDate | 2020-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-495a5656ef50420d820dbad83e82db282023-11-20T08:18:18ZengMDPI AGRemote Sensing2072-42922020-07-011215242610.3390/rs12152426Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble ModelAlin-Ionuț Pleșoianu0Mihai-Sorin Stupariu1Ionuț Șandric2Ileana Pătru-Stupariu3Lucian Drăguț4Faculty of Geography, Doctoral School Simion Mehedinți, University of Bucharest, Bd. N. Bălcescu, no.1, 010041 Bucharest, RomaniaInstitute of Research of University of Bucharest, ICUB, Transdisciplinary Research Centre Landscape- Territory-Information Systems, CeLTIS, Splaiul Independenței nr. 91–95, 050095 Bucharest, RomaniaDepartment of Regional Geography and Environment, Faculty of Geography, University of Bucharest, Bd. N. Bălcescu, 1, 010041 Bucharest, RomaniaInstitute of Research of University of Bucharest, ICUB, Transdisciplinary Research Centre Landscape- Territory-Information Systems, CeLTIS, Splaiul Independenței nr. 91–95, 050095 Bucharest, RomaniaDepartment of Geography, West University of Timișoara, Blvd. V. Parvan 4, 300223 Timisoara, RomaniaTraditional methods for individual tree-crown (ITC) detection (image classification, segmentation, template matching, etc.) applied to very high-resolution remote sensing imagery have been shown to struggle in disparate landscape types or image resolutions due to scale problems and information complexity. Deep learning promised to overcome these shortcomings due to its superior performance and versatility, proven with reported detection rates of ~90%. However, such models still find their limits in transferability across study areas, because of different tree conditions (e.g., isolated trees vs. compact forests) and/or resolutions of the input data. This study introduces a highly replicable deep learning ensemble design for ITC detection and species classification based on the established single shot detector (SSD) model. The ensemble model design is based on varying the input data for the SSD models, coupled with a voting strategy for the output predictions. Very high-resolution unmanned aerial vehicles (UAV), aerial remote sensing imagery and elevation data are used in different combinations to test the performance of the ensemble models in three study sites with highly contrasting spatial patterns. The results show that ensemble models perform better than any single SSD model, regardless of the local tree conditions or image resolution. The detection performance and the accuracy rates improved by 3–18% with only as few as two participant single models, regardless of the study site. However, when more than two models were included, the performance of the ensemble models only improved slightly and even dropped.https://www.mdpi.com/2072-4292/12/15/2426tree-crown detectiondeep learningensemble modelobject detectionsingle shot detector |
| spellingShingle | Alin-Ionuț Pleșoianu Mihai-Sorin Stupariu Ionuț Șandric Ileana Pătru-Stupariu Lucian Drăguț Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble Model Remote Sensing tree-crown detection deep learning ensemble model object detection single shot detector |
| title | Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble Model |
| title_full | Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble Model |
| title_fullStr | Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble Model |
| title_full_unstemmed | Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble Model |
| title_short | Individual Tree-Crown Detection and Species Classification in Very High-Resolution Remote Sensing Imagery Using a Deep Learning Ensemble Model |
| title_sort | individual tree crown detection and species classification in very high resolution remote sensing imagery using a deep learning ensemble model |
| topic | tree-crown detection deep learning ensemble model object detection single shot detector |
| url | https://www.mdpi.com/2072-4292/12/15/2426 |
| work_keys_str_mv | AT alinionutplesoianu individualtreecrowndetectionandspeciesclassificationinveryhighresolutionremotesensingimageryusingadeeplearningensemblemodel AT mihaisorinstupariu individualtreecrowndetectionandspeciesclassificationinveryhighresolutionremotesensingimageryusingadeeplearningensemblemodel AT ionutsandric individualtreecrowndetectionandspeciesclassificationinveryhighresolutionremotesensingimageryusingadeeplearningensemblemodel AT ileanapatrustupariu individualtreecrowndetectionandspeciesclassificationinveryhighresolutionremotesensingimageryusingadeeplearningensemblemodel AT luciandragut individualtreecrowndetectionandspeciesclassificationinveryhighresolutionremotesensingimageryusingadeeplearningensemblemodel |