Fusion of hyperspectral imaging and LiDAR for forest monitoring

Tusa, Eduardo; Laybros, Anthony; Monnet, Jean‐Matthieu; Mura, Mauro Dalla; Barré, Jean-Baptiste; Vincent, Grégoire; Dalponte, Michele; Féret, Jean‐Baptiste; Chanussot, Jocelyn

doi:10.1016/b978-0-444-63977-6.00013-4

Cited by 20 publications

(16 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This indicates that crown segmentation in a multi-layered closed canopy forest can be better improved by using 3D segmentation methods rather than relying mostly on the canopy surface model. Future improvement of current methods may stem from the fusion of lidar and spectral information [53,54]. Due to increasing occlusion levels towards the ground, segmentation of subcanopy trees in tall dense forests remains an open challenge.…”

Section: Discussionmentioning

confidence: 99%

A Comparative Assessment of the Performance of Individual Tree Crowns Delineation Algorithms from ALS Data in Tropical Forests

et al. 2019

Self Cite

View full text Add to dashboard Cite

Tropical forest canopies are comprised of tree crowns of multiple species varying in shape and height, and ground inventories do not usually reliably describe their structure. Airborne laser scanning data can be used to characterize these individual crowns, but analytical tools developed for boreal or temperate forests may require to be adjusted before they can be applied to tropical environments. Therefore, we compared results from six different segmentation methods applied to six plots (39 ha) from a study site in French Guiana. We measured the overlap of automatically segmented crowns projection with selected crowns manually delineated on high-resolution photography. We also evaluated the goodness of fit following automatic matching with field inventory data using a model linking tree diameter to tree crown width. The different methods tested in this benchmark segmented highly different numbers of crowns having different characteristics. Segmentation methods based on the point cloud (AMS3D and Graph-Cut) globally outperformed methods based on the Canopy Height Models, especially for small crowns; the AMS3D method outperformed the other methods tested for the overlap analysis, and AMS3D and Graph-Cut performed the best for the automatic matching validation. Nevertheless, other methods based on the Canopy Height Model performed better for very large emergent crowns. The dense foliage of tropical moist forests prevents sufficient point densities in the understory to segment subcanopy trees accurately, regardless of the segmentation method.

show abstract

Section: Discussionmentioning

confidence: 99%

A Comparative Assessment of the Performance of Individual Tree Crowns Delineation Algorithms from ALS Data in Tropical Forests

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Segmentation itself may take advantage of multisource data. Depending of the way data are merged, fine co-registration can be required but may be difficult to achieve [64,65].…”

Section: Discussionmentioning

confidence: 99%

Quantitative Airborne Inventories in Dense Tropical Forest Using Imaging Spectroscopy

et al. 2020

Self Cite

View full text Add to dashboard Cite

Tropical forests have exceptional floristic diversity, but their characterization remains incomplete, in part due to the resource intensity of in-situ assessments. Remote sensing technologies can provide valuable, cost-effective, large-scale insights. This study investigates the combined use of airborne LiDAR and imaging spectroscopy to map tree species at landscape scale in French Guiana. Binary classifiers were developed for each of 20 species using linear discriminant analysis (LDA), regularized discriminant analysis (RDA) and logistic regression (LR). Complementing visible and near infrared (VNIR) spectral bands with short wave infrared (SWIR) bands improved the mean average classification accuracy of the target species from 56.1% to 79.6%. Increasing the number of non-focal species decreased the success rate of target species identification. Classification performance was not significantly affected by impurity rates (confusion between assigned classes) in the non-focal class (up to 5% of bias), provided that an adequate criterion was used for adjusting threshold probability assignment. A limited number of crowns (30 crowns) in each species class was sufficient to retrieve correct labels effectively. Overall canopy area of target species was strongly correlated to their basal area over 118 ha at 1.5 ha resolution, indicating that operational application of the method is a realistic prospect (R2 = 0.75 for six major commercial tree species).

show abstract

“…Land cover classification is another application area where hyperspectral data are used extensively [21]. To improve land cover classification, the use of light detection and ranging (LiDAR) data and the fusion of LiDAR and hyperspectral data have been considered by several works [22][23][24][25][26][27]. Because some land covers differ among themselves with respect to their height, height could be a valuable piece of information.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning for Land Cover Classification Using Only a Few Bands

et al. 2020

View full text Add to dashboard Cite

There is an emerging interest in using hyperspectral data for land cover classification. The motivation behind using hyperspectral data is the notion that increasing the number of narrowband spectral channels would provide richer spectral information and thus help improve the land cover classification performance. Although hyperspectral data with hundreds of channels provide detailed spectral signatures, the curse of dimensionality might lead to degradation in the land cover classification performance. Moreover, in some practical applications, hyperspectral data may not be available due to cost, data storage, or bandwidth issues, and RGB and near infrared (NIR) could be the only image bands available for land cover classification. Light detection and ranging (LiDAR) data is another type of data to assist land cover classification especially if the land covers of interest have different heights. In this paper, we examined the performance of two Convolutional Neural Network (CNN)-based deep learning algorithms for land cover classification using only four bands (RGB+NIR) and five bands (RGB+NIR+LiDAR), where these limited number of image bands were augmented using Extended Multi-attribute Profiles (EMAP). The deep learning algorithms were applied to a well-known dataset used in the 2013 IEEE Geoscience and Remote Sensing Society (GRSS) Data Fusion Contest. With EMAP augmentation, the two deep learning algorithms were observed to achieve better land cover classification performance using only four bands as compared to that using all 144 hyperspectral bands.

show abstract

Fusion of hyperspectral imaging and LiDAR for forest monitoring

Cited by 20 publications

References 59 publications

A Comparative Assessment of the Performance of Individual Tree Crowns Delineation Algorithms from ALS Data in Tropical Forests

A Comparative Assessment of the Performance of Individual Tree Crowns Delineation Algorithms from ALS Data in Tropical Forests

Quantitative Airborne Inventories in Dense Tropical Forest Using Imaging Spectroscopy

Deep Learning for Land Cover Classification Using Only a Few Bands

Contact Info

Product

Resources

About