Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning
Measuring and monitoring the height of vegetation provides important insights into forest age and habitat quality. These are essential for the accuracy of applications that are highly reliant on up-to-date and accurate vegetation data. Current vegetation sensing practices involve ground survey, phot...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/22/3797 |
| _version_ | 1797547404337086464 |
|---|---|
| author | David Radke Daniel Radke John Radke |
| author_facet | David Radke Daniel Radke John Radke |
| author_sort | David Radke |
| collection | DOAJ |
| description | Measuring and monitoring the height of vegetation provides important insights into forest age and habitat quality. These are essential for the accuracy of applications that are highly reliant on up-to-date and accurate vegetation data. Current vegetation sensing practices involve ground survey, photogrammetry, synthetic aperture radar (SAR), and airborne light detection and ranging sensors (LiDAR). While these methods provide high resolution and accuracy, their hardware and collection effort prohibits highly recurrent and widespread collection. In response to the limitations of current methods, we designed Y-NET, a novel deep learning model to generate high resolution models of vegetation from highly recurrent multispectral aerial imagery and elevation data. Y-NET’s architecture uses convolutional layers to learn correlations between different input features and vegetation height, generating an accurate vegetation surface model (VSM) at <inline-formula><math display="inline"><semantics><mrow><mn>1</mn><mo>×</mo><mn>1</mn></mrow></semantics></math></inline-formula> m resolution. We evaluated Y-NET on 235 km<inline-formula><math display="inline"><semantics><msup><mrow></mrow><mn>2</mn></msup></semantics></math></inline-formula> of the East San Francisco Bay Area and find that Y-NET achieves low error from LiDAR when tested on new locations. Y-NET also achieves an <inline-formula><math display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula> of 0.83 and can effectively model complex vegetation through side-by-side visual comparisons. Furthermore, we show that Y-NET is able to identify instances of vegetation growth and mitigation by comparing aerial imagery and LiDAR collected at different times. |
| first_indexed | 2024-03-10T14:43:42Z |
| format | Article |
| id | doaj.art-4a8eee57549546d08385c165501ccebb |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T14:43:42Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-4a8eee57549546d08385c165501ccebb2023-11-20T21:32:22ZengMDPI AGRemote Sensing2072-42922020-11-011222379710.3390/rs12223797Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep LearningDavid Radke0Daniel Radke1John Radke2David R. Cheriton School of Computer Science, University of Waterloo, Waterloo, ON N2L 3G1, CanadaDepartment of Computer Science, Universität des Saarlandes, 66123 Saarbrücken, GermanyLandscape Architecture and Environmental Planning, City and Regional Planning, University of California, Berkeley, CA 94720-2000, USAMeasuring and monitoring the height of vegetation provides important insights into forest age and habitat quality. These are essential for the accuracy of applications that are highly reliant on up-to-date and accurate vegetation data. Current vegetation sensing practices involve ground survey, photogrammetry, synthetic aperture radar (SAR), and airborne light detection and ranging sensors (LiDAR). While these methods provide high resolution and accuracy, their hardware and collection effort prohibits highly recurrent and widespread collection. In response to the limitations of current methods, we designed Y-NET, a novel deep learning model to generate high resolution models of vegetation from highly recurrent multispectral aerial imagery and elevation data. Y-NET’s architecture uses convolutional layers to learn correlations between different input features and vegetation height, generating an accurate vegetation surface model (VSM) at <inline-formula><math display="inline"><semantics><mrow><mn>1</mn><mo>×</mo><mn>1</mn></mrow></semantics></math></inline-formula> m resolution. We evaluated Y-NET on 235 km<inline-formula><math display="inline"><semantics><msup><mrow></mrow><mn>2</mn></msup></semantics></math></inline-formula> of the East San Francisco Bay Area and find that Y-NET achieves low error from LiDAR when tested on new locations. Y-NET also achieves an <inline-formula><math display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula> of 0.83 and can effectively model complex vegetation through side-by-side visual comparisons. Furthermore, we show that Y-NET is able to identify instances of vegetation growth and mitigation by comparing aerial imagery and LiDAR collected at different times.https://www.mdpi.com/2072-4292/12/22/3797deep learningartificial intelligencevegetation surface modeling |
| spellingShingle | David Radke Daniel Radke John Radke Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning Remote Sensing deep learning artificial intelligence vegetation surface modeling |
| title | Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning |
| title_full | Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning |
| title_fullStr | Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning |
| title_full_unstemmed | Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning |
| title_short | Beyond Measurement: Extracting Vegetation Height from High Resolution Imagery with Deep Learning |
| title_sort | beyond measurement extracting vegetation height from high resolution imagery with deep learning |
| topic | deep learning artificial intelligence vegetation surface modeling |
| url | https://www.mdpi.com/2072-4292/12/22/3797 |
| work_keys_str_mv | AT davidradke beyondmeasurementextractingvegetationheightfromhighresolutionimagerywithdeeplearning AT danielradke beyondmeasurementextractingvegetationheightfromhighresolutionimagerywithdeeplearning AT johnradke beyondmeasurementextractingvegetationheightfromhighresolutionimagerywithdeeplearning |