@article{fdi:010089753,
  title = {{A}dapting prediction models to bare soil fractional cover for extending topsoil clay content mapping based on {AVIRIS}-{NG} hyperspectral data},
  author = {{G}eorge, {E}. {B}. and {G}omez, {C}{\'e}cile and {K}umar, {N}. {D}.},
  editor = {},
  language = {{ENG}},
  abstract = {{T}he deployment of remote sensing platforms has facilitated the mapping of soil properties to a great extent. {H}owever, the accuracy of these soil property estimates is compromised by the presence of non-soil cover, which introduces interference with the acquired reflectance spectra over pixels. {T}herefore, current soil property estimation by remote sensing is limited to bare soil pixels, which are identified based on spectral indices of vegetation. {O}ur study proposes a composite mapping approach to extend the soil properties mapping beyond bare soil pixels, associated with an uncertainty map. {T}he proposed approach first classified the pixels based on their bare soil fractional cover by spectral unmixing. {T}hen, a specific regression model was built and applied to each bare soil fractional cover class to estimate clay content. {F}inally, the clay content maps created for each bare soil fractional cover class were mosaicked to create a composite map of clay content estimations. {A} bootstrap procedure was used to estimate the standard deviation of clay content predictions per bare soil fractional cover dataset, which represented the uncertainty of estimations. {T}his study used a hyperspectral image acquired by the {A}irborne {V}isible/{I}nfrared {I}maging {S}pectrometer-{N}ext {G}eneration ({AVIRIS}-{NG}) sensor over cultivated fields in {S}outh {I}ndia. {T}he proposed approach provided modest performances in prediction ( {R} v a l 2 ranging from 0.53 to 0.63) depending on the bare soil fractional cover class and showed a correct spatial pattern, regardless of the bare soil fraction classes. {T}he model's performance was observed to increase with the adoption of higher bare soil fractional cover thresholds. {T}he mapped area ranged from 10.4% for pixels with bare soil fractional cover >0.7 to 52.7% for pixels with bare soil fractional cover >0.3. {T}he approach thus extended the mapped surface by 42.4%, while maintaining acceptable prediction performances. {F}inally, the proposed approach could be adopted to extend the mapping capability of planned and current hyperspectral satellite missions.},
  keywords = {composite map ; clay content ; digital soil mapping ; uncertainty ; regression model},
  booktitle = {},
  journal = {{R}emote {S}ensing},
  volume = {16},
  numero = {6},
  pages = {1066 [20 p.]},
  year = {2024},
  DOI = {10.3390/rs16061066},
  URL = {https://www.documentation.ird.fr/hor/fdi:010089753},
}