Tree Species Identification in Mixed Baltic Forest Using LiDAR and Multispectral Data

Dinuls, R.; Erins, G.; Lorencs, Aivars; Mednieks, Ints; Sinica-Sinavskis, Juris

doi:10.1109/jstars.2012.2196978

Cited by 48 publications

(23 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By comparison, active remote sensing instruments, such as Light Detection and Ranging (LiDAR) scanners, have played a relatively small role in efforts to detect and map tree species [11,12]. Studies using LiDAR in conjunction with spectral data [13][14][15][16] have yielded promising results, indicating the potential of combining remote sensing technologies for species mapping.…”

Section: Introductionmentioning

confidence: 99%

Semi-Supervised Methods to Identify Individual Crowns of Lowland Tropical Canopy Species Using Imaging Spectroscopy and LiDAR

Féret

Asner

2012

Remote Sensing

View full text Add to dashboard Cite

Abstract:Our objective is to identify and map individuals of nine tree species in a Hawaiian lowland tropical forest by comparing the performance of a variety of semi-supervised classifiers. A method was adapted to process hyperspectral imagery, LiDAR intensity variables, and LiDAR-derived canopy height and use them to assess the identification accuracy. We found that semi-supervised Support Vector Machine classification using tensor summation kernel was superior to supervised classification, with demonstrable accuracy for at least eight out of nine species, and for all combinations of data types tested. We also found that the combination of hyperspectral imagery and LiDAR data usually improved species classification. Both LiDAR intensity and LiDAR canopy height proved useful for classification of certain species, but the improvements varied depending upon the species in question. Our results pave the way for target-species identification in tropical forests and other ecosystems.

show abstract

Section: Introductionmentioning

confidence: 99%

Semi-Supervised Methods to Identify Individual Crowns of Lowland Tropical Canopy Species Using Imaging Spectroscopy and LiDAR

Féret

Asner

2012

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Since HI records surface spectral reflectance characteristics of objects, while FWL reveals scattering properties and geometrical structure (roughness, slope, spatial distribution) of targets, their complementary nature suggests that the fusion of HI and FWL can be beneficial for applications such as land cover classification, forest inventory estimation, and obscured target detection Kanaev et al, 2011). Although previous work has addressed the issue of fusing LiDAR data and HI, for example (Dalponte et al, 2008;Erdody and Moskal, 2010;Jones et al, 2008;Dinuls et al, 2012), the work has predominantly been limited to using simple gridded discrete return LiDAR data, such as Digital Elevation Models (DEM), Digital Terrain Models (DTM), or using data from a FWL system that has been converted to discrete return point clouds (Anderson et al, 2008;Asner et al, 2007;Ranjani et al, 2014;Paris and Bruzzone, 2015 feature level fusion with HI (Sarrazin et al, 2011;Jung, 2011). However in these instances, the FWL return energy profile in the entire observed cone of diffraction of the outgoing laser pulses is discarded before the analysis of the fused FWL and HI dataset is undertaken.…”

Section: Introductionmentioning

confidence: 99%

Fusion of waveform LiDAR data and hyperspectral imagery for land cover classification

Wang

Glennie

2015

ISPRS Journal of Photogrammetry and Remote Sensing

View full text Add to dashboard Cite

“…In ideal conditions, several tree species can be identified by hyperspectral imagery with a precision up to 97-98 % (Dinuls et al, 2011;Dinuls et al, 2012). However, in practice there are several factors affecting the reflection from tree canopies.…”

Section: Introductionmentioning

confidence: 99%

“…The spectral behaviour of plants, in particular, of leaves depends on their morphological, structural and chemical properties, and the capacity to distinguish species remotely may be more successful, taking into account speciesspecific tree architectural traits (Mohammed et al, 2000;Noble and Brown, 2009;Dinuls et al, 2012).…”

Section: Introductionmentioning

confidence: 99%

Plant Reflected Spectra Depending on Biological Characteristics and Growth Conditions

Karlovska

Grīnfelde

Alsiņa

et al. 2015

Proceedings of the 7th International Scientific Conference Rural Development 2015

View full text Add to dashboard Cite

Sustainable and economically based forestry needs modern inventory and monitoring techniques. One of the most common technologies for identification of forest tree species and monitoring of forest growth conditions is the hyperspectral remote sensing. This technology gives an opportunity to economize human resources and time for data collecting and processing. The spectral behaviour of plant leaves depends on number of factors, including environmental background. The aim of this study was to assess the tree reflectance spectra in relation to the growth conditions to take into account potential differences for increasing precision of species identification in Latvian forests and for estimating of forest growth conditions. Remote sensing data were obtained using a specialized aircraft (Pilatus PC-6), which is equipped with a high-performance airborne VNIR pushbroom hyperspectral system (AisaEAGLE). The study area was flown at 1000 m altitude. Data was recorded in the 400-970 nm spectral range, spectral resolution was 3.3 nm, ground resolution 0.5 m. Data processing consisted of manually selecting trees with a recognizable tree crowns in the airborne images. Tree centres were adjusted by putting them in the accurate position according to the situation in aerial photography. All trees with a diameter at breast height DBH of more than 5 cm were measured and for each tree coordinates, its species, height, DBH, crown width and length were recorded. Differentially corrected Global Positioning System measurements were used to determine the position of each plot centre. Data from different hyperspectral bands were compared using ANOVA at confidence level 95 %. Four species: Scots pine (Pinus sylvestris L.), Norway spruce (Picea abies (L.) H. Karst), silver birch (Betula pendula Roth), and European aspen (Populus tremula L.) -were examined in distinct forest site types. The spectral response of studied species was 1) different between species and 2) different between site types within each species, correlating with soil fertility gradient and soil moisture gradient. Differences between species occurred most in the intensity of reflected electromagnetic radiation rather than distinctive locations of maximums or minimums in spectrum curve, and near infrared (NIR) region of spectrum showed more differences between species than visible light zone. Most informative wavebands for distinguishing differences between site types were 805 nm and 644 nm.

show abstract

Tree Species Identification in Mixed Baltic Forest Using LiDAR and Multispectral Data

Cited by 48 publications

References 28 publications

Semi-Supervised Methods to Identify Individual Crowns of Lowland Tropical Canopy Species Using Imaging Spectroscopy and LiDAR

Semi-Supervised Methods to Identify Individual Crowns of Lowland Tropical Canopy Species Using Imaging Spectroscopy and LiDAR

Fusion of waveform LiDAR data and hyperspectral imagery for land cover classification

Plant Reflected Spectra Depending on Biological Characteristics and Growth Conditions

Contact Info

Product

Resources

About