Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous trees
Tree species identification constitutes a bottleneck in remote sensing applications. Waveform LiDAR has been shown to offer potential over discrete-return observations, and we assessed if the combination of two-wavelength waveform data can lead to further improvements. A total of 2532 trees represen...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Finnish Society of Forest Science
2023-01-01
|
| Series: | Silva Fennica |
| Subjects: | |
| Online Access: | https://www.silvafennica.fi/article/22007 |
| _version_ | 1827345234086854656 |
|---|---|
| author | Ilkka Korpela Antti Polvivaara Saija Papunen Laura Jaakkola Noora Tienaho Johannes Uotila Tuomas Puputti Aleksi Flyktman |
| author_facet | Ilkka Korpela Antti Polvivaara Saija Papunen Laura Jaakkola Noora Tienaho Johannes Uotila Tuomas Puputti Aleksi Flyktman |
| author_sort | Ilkka Korpela |
| collection | DOAJ |
| description | Tree species identification constitutes a bottleneck in remote sensing applications. Waveform LiDAR has been shown to offer potential over discrete-return observations, and we assessed if the combination of two-wavelength waveform data can lead to further improvements. A total of 2532 trees representing seven living and dead conifer and deciduous species classes found in Hyytiälä forests in southern Finland were included in the experiments. LiDAR data was acquired by two single-wavelength sensors. The 1064-nm and 1550-nm data were radiometrically corrected to enable range-normalization using the radar equation. Pulses were traced through the canopy, and by applying 3D crown models, the return waveforms were assigned to individual trees. Crown models and a terrain model enabled a further split of the waveforms to strata representing the crown, understory and ground segments. Different geometric and radiometric waveform attributes were extracted per return pulse and aggregated to tree-level mean and standard deviation features. We analyzed the effect of tree size on the features, the correlation between features and the between-species differences of the waveform features. Feature importance for species classification was derived using F-test and the Random Forest algorithm. Classification tests showed significant improvement in overall accuracy (74→83% with 7 classes, 88→91% with 4 classes) when the 1064-nm and 1550-nm features were merged. Most features were not invariant to tree size, and the dependencies differed between species and LiDAR wavelength. The differences were likely driven by factors such as bark reflectance, height growth induced structural changes near the treetop as well as foliage density in old trees. |
| first_indexed | 2024-03-07T23:03:37Z |
| format | Article |
| id | doaj.art-e41b522f0d704574b1c7f089eed63a9b |
| institution | Directory Open Access Journal |
| issn | 2242-4075 |
| language | English |
| last_indexed | 2024-03-07T23:03:37Z |
| publishDate | 2023-01-01 |
| publisher | Finnish Society of Forest Science |
| record_format | Article |
| series | Silva Fennica |
| spelling | doaj.art-e41b522f0d704574b1c7f089eed63a9b2024-02-22T10:35:01ZengFinnish Society of Forest ScienceSilva Fennica2242-40752023-01-0156410.14214/sf.22007Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous treesIlkka Korpela0https://orcid.org/0000-0002-1665-3984Antti Polvivaara1Saija Papunen2https://orcid.org/0000-0001-5383-4314Laura Jaakkola3Noora Tienaho4https://orcid.org/0000-0002-6574-5797Johannes Uotila5Tuomas Puputti6https://orcid.org/0000-0003-1972-1636Aleksi Flyktman7https://orcid.org/0000-0002-5235-317XUniversity of Helsinki, Department of Forest Sciences, P.O. Box 27, FI-00014 University of Helsinki, Finland University of Helsinki, Department of Forest Sciences, P.O. Box 27, FI-00014 University of Helsinki, FinlandUniversity of Helsinki, Department of Forest Sciences, P.O. Box 27, FI-00014 University of Helsinki, FinlandUniversity of Helsinki, Department of Forest Sciences, P.O. Box 27, FI-00014 University of Helsinki, FinlandUniversity of Eastern Finland, Faculty of Science and Forestry, P.O. Box 111, FI-80101 Joensuu, FinlandUniversity of Helsinki, Department of Forest Sciences, P.O. Box 27, FI-00014 University of Helsinki, FinlandUniversity of Helsinki, Department of Forest Sciences, P.O. Box 27, FI-00014 University of Helsinki, FinlandUniversity of Helsinki, Department of Forest Sciences, P.O. Box 27, FI-00014 University of Helsinki, FinlandTree species identification constitutes a bottleneck in remote sensing applications. Waveform LiDAR has been shown to offer potential over discrete-return observations, and we assessed if the combination of two-wavelength waveform data can lead to further improvements. A total of 2532 trees representing seven living and dead conifer and deciduous species classes found in Hyytiälä forests in southern Finland were included in the experiments. LiDAR data was acquired by two single-wavelength sensors. The 1064-nm and 1550-nm data were radiometrically corrected to enable range-normalization using the radar equation. Pulses were traced through the canopy, and by applying 3D crown models, the return waveforms were assigned to individual trees. Crown models and a terrain model enabled a further split of the waveforms to strata representing the crown, understory and ground segments. Different geometric and radiometric waveform attributes were extracted per return pulse and aggregated to tree-level mean and standard deviation features. We analyzed the effect of tree size on the features, the correlation between features and the between-species differences of the waveform features. Feature importance for species classification was derived using F-test and the Random Forest algorithm. Classification tests showed significant improvement in overall accuracy (74→83% with 7 classes, 88→91% with 4 classes) when the 1064-nm and 1550-nm features were merged. Most features were not invariant to tree size, and the dependencies differed between species and LiDAR wavelength. The differences were likely driven by factors such as bark reflectance, height growth induced structural changes near the treetop as well as foliage density in old trees.https://www.silvafennica.fi/article/22007crown modeling; laser scanning; photogrammetry; individual tree detection; scandinavia |
| spellingShingle | Ilkka Korpela Antti Polvivaara Saija Papunen Laura Jaakkola Noora Tienaho Johannes Uotila Tuomas Puputti Aleksi Flyktman Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous trees Silva Fennica crown modeling; laser scanning; photogrammetry; individual tree detection; scandinavia |
| title | Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous trees |
| title_full | Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous trees |
| title_fullStr | Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous trees |
| title_full_unstemmed | Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous trees |
| title_short | Airborne dual-wavelength waveform LiDAR improves species classification accuracy of boreal broadleaved and coniferous trees |
| title_sort | airborne dual wavelength waveform lidar improves species classification accuracy of boreal broadleaved and coniferous trees |
| topic | crown modeling; laser scanning; photogrammetry; individual tree detection; scandinavia |
| url | https://www.silvafennica.fi/article/22007 |
| work_keys_str_mv | AT ilkkakorpela airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees AT anttipolvivaara airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees AT saijapapunen airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees AT laurajaakkola airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees AT nooratienaho airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees AT johannesuotila airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees AT tuomaspuputti airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees AT aleksiflyktman airbornedualwavelengthwaveformlidarimprovesspeciesclassificationaccuracyofborealbroadleavedandconiferoustrees |