@article{fdi:010079308,
  title = {{Q}uantitative airborne inventories in dense tropical forest using imaging spectroscopy},
  author = {{L}aybros, {A}. and {A}ubry-{K}ientz, {M}. and {F}eret, {J}. {B}. and {B}edeau, {C}. and {B}runaux, {O}. and {D}erroire, {G}. and {V}incent, {G}r{\'e}goire},
  editor = {},
  language = {{ENG}},
  abstract = {{T}ropical forests have exceptional floristic diversity, but their characterization remains incomplete, in part due to the resource intensity of in-situ assessments. {R}emote sensing technologies can provide valuable, cost-effective, large-scale insights. {T}his study investigates the combined use of airborne {L}i{DAR} and imaging spectroscopy to map tree species at landscape scale in {F}rench {G}uiana. {B}inary classifiers were developed for each of 20 species using linear discriminant analysis ({LDA}), regularized discriminant analysis ({RDA}) and logistic regression ({LR}). {C}omplementing 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%. {I}ncreasing the number of non-focal species decreased the success rate of target species identification. {C}lassification 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. {O}verall 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 ({R}-2 = 0.75 for six major commercial tree species).},
  keywords = {tropical forest ; species diversity ; hyperspectral ; {L}i{DAR} ; {GUYANE} {FRNCAISE} ; {PARACOU}},
  booktitle = {},
  journal = {{R}emote {S}ensing},
  volume = {12},
  numero = {10},
  pages = {art. 1577 [27 p.]},
  year = {2020},
  DOI = {10.3390/rs12101577},
  URL = {https://www.documentation.ird.fr/hor/fdi:010079308},
}