Abstract
Recognition of tree species and geospatial information on tree species composition is essential for forest management. In this study, tree species recognition was examined using hyperspectral imagery from visible to near-infrared (VNIR) and short-wave infrared (SWIR) camera sensors in combination with a 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum with a diverse selection of 26 tree species from 14 genera was used as a test area. Aerial hyperspectral imagery and high spatial resolution photogrammetric color imagery were acquired from the test area using unmanned aerial vehicle (UAV) borne sensors. Hyperspectral imagery was processed to calibrated reflectance mosaics and was tested along with the mosaics based on original image digital number values (DN). Two alternative classifiers, a k nearest neighbor method (k-nn), combined with a genetic algorithm and a random forest method, were tested for predicting the tree species and genus, as well as for selecting an optimal set of remote sensing features for this task. The combination of VNIR, SWIR, and 3D features performed better than any of the data sets individually. Furthermore, the calibrated reflectance values performed better compared to uncorrected DN values. These trends were similar with both tested classifiers. Of the classifiers, the k-nn combined with the genetic algorithm provided consistently better results than the random forest algorithm. The best result was thus achieved using calibrated reflectance features from VNIR and SWIR imagery together with 3D point cloud features; the proportion of correctly-classified trees was 0.823 for tree species and 0.869 for tree genus.
| Original language | English |
|---|---|
| Article number | 714 |
| Journal | Remote Sensing |
| Volume | 10 |
| Issue number | 5 |
| DOIs | |
| Publication status | Published - 1 May 2018 |
| MoE publication type | A1 Journal article-refereed |
Fingerprint
Keywords
- Dense point cloud
- Genetic algorithm
- Hyperspectral imagery
- Machine learning
- Photogrammetry
- Random forest
- Reflectance calibration
- Tree species recognition
- UAV
Cite this
}
Assessment of classifiers and remote sensing features of hyperspectral imagery and stereo-photogrammetric point clouds for recognition of tree species in a forest area of high species diversity. / Tuominen, Sakari (Corresponding Author); Näsi, Roope; Honkavaara, Eija; Balazs, Andras; Hakala, Teemu; Viljanen, Niko; Pölönen, Ilkka; Saari, Heikki; Ojanen, Harri.
In: Remote Sensing, Vol. 10, No. 5, 714, 01.05.2018.Research output: Contribution to journal › Article › Scientific › peer-review
TY - JOUR
T1 - Assessment of classifiers and remote sensing features of hyperspectral imagery and stereo-photogrammetric point clouds for recognition of tree species in a forest area of high species diversity
AU - Tuominen, Sakari
AU - Näsi, Roope
AU - Honkavaara, Eija
AU - Balazs, Andras
AU - Hakala, Teemu
AU - Viljanen, Niko
AU - Pölönen, Ilkka
AU - Saari, Heikki
AU - Ojanen, Harri
PY - 2018/5/1
Y1 - 2018/5/1
N2 - Recognition of tree species and geospatial information on tree species composition is essential for forest management. In this study, tree species recognition was examined using hyperspectral imagery from visible to near-infrared (VNIR) and short-wave infrared (SWIR) camera sensors in combination with a 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum with a diverse selection of 26 tree species from 14 genera was used as a test area. Aerial hyperspectral imagery and high spatial resolution photogrammetric color imagery were acquired from the test area using unmanned aerial vehicle (UAV) borne sensors. Hyperspectral imagery was processed to calibrated reflectance mosaics and was tested along with the mosaics based on original image digital number values (DN). Two alternative classifiers, a k nearest neighbor method (k-nn), combined with a genetic algorithm and a random forest method, were tested for predicting the tree species and genus, as well as for selecting an optimal set of remote sensing features for this task. The combination of VNIR, SWIR, and 3D features performed better than any of the data sets individually. Furthermore, the calibrated reflectance values performed better compared to uncorrected DN values. These trends were similar with both tested classifiers. Of the classifiers, the k-nn combined with the genetic algorithm provided consistently better results than the random forest algorithm. The best result was thus achieved using calibrated reflectance features from VNIR and SWIR imagery together with 3D point cloud features; the proportion of correctly-classified trees was 0.823 for tree species and 0.869 for tree genus.
AB - Recognition of tree species and geospatial information on tree species composition is essential for forest management. In this study, tree species recognition was examined using hyperspectral imagery from visible to near-infrared (VNIR) and short-wave infrared (SWIR) camera sensors in combination with a 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum with a diverse selection of 26 tree species from 14 genera was used as a test area. Aerial hyperspectral imagery and high spatial resolution photogrammetric color imagery were acquired from the test area using unmanned aerial vehicle (UAV) borne sensors. Hyperspectral imagery was processed to calibrated reflectance mosaics and was tested along with the mosaics based on original image digital number values (DN). Two alternative classifiers, a k nearest neighbor method (k-nn), combined with a genetic algorithm and a random forest method, were tested for predicting the tree species and genus, as well as for selecting an optimal set of remote sensing features for this task. The combination of VNIR, SWIR, and 3D features performed better than any of the data sets individually. Furthermore, the calibrated reflectance values performed better compared to uncorrected DN values. These trends were similar with both tested classifiers. Of the classifiers, the k-nn combined with the genetic algorithm provided consistently better results than the random forest algorithm. The best result was thus achieved using calibrated reflectance features from VNIR and SWIR imagery together with 3D point cloud features; the proportion of correctly-classified trees was 0.823 for tree species and 0.869 for tree genus.
KW - Dense point cloud
KW - Genetic algorithm
KW - Hyperspectral imagery
KW - Machine learning
KW - Photogrammetry
KW - Random forest
KW - Reflectance calibration
KW - Tree species recognition
KW - UAV
UR - http://www.scopus.com/inward/record.url?scp=85047571087&partnerID=8YFLogxK
U2 - 10.3390/rs10050714
DO - 10.3390/rs10050714
M3 - Article
VL - 10
JO - Remote Sensing
JF - Remote Sensing
SN - 2072-4292
IS - 5
M1 - 714
ER -