Horizon / Plein textes La base de ressources documentaires de l'IRD

IRD

Publications des scientifiques de l'IRD

Boissieu Florian de, Sevin B., Cudahy T., Mangeas Morgan, Chevrel S., Ong C., Rodger A., Maurizot P., Laukamp C., Lau I., Touraivane T., Cluzel D., Despinoy Marc. (2018). Regolith-geology mapping with support vector machine : a case study over weathered Ni-bearing peridotites, New Caledonia. International Journal of Applied Earth Observation and Geoinformation, 64, 377-385. ISSN 0303-2434

Accès réservé (Intranet IRD) Document en accès réservé (Intranet IRD)

Lien direct chez l'éditeur doi:10.1016/j.jag.2017.05.012

Titre
Regolith-geology mapping with support vector machine : a case study over weathered Ni-bearing peridotites, New Caledonia
Année de publication2018
Type de documentArticle référencé dans le Web of Science WOS:000413880000033
AuteursBoissieu Florian de, Sevin B., Cudahy T., Mangeas Morgan, Chevrel S., Ong C., Rodger A., Maurizot P., Laukamp C., Lau I., Touraivane T., Cluzel D., Despinoy Marc.
SourceInternational Journal of Applied Earth Observation and Geoinformation, 2018, 64, p. 377-385. ISSN 0303-2434
RésuméAccurate maps of Earth's geology, especially its regolith, are required for managing the sustainable exploration and development of mineral resources. This paper shows how airborne imaging hyperspectral data collected over weathered peridotite rocks in vegetated, mountainous terrane in New Caledonia were processed using a combination of methods to generate a regolith-geology map that could be used for more efficiently targeting Ni exploration. The image processing combined two usual methods, which are spectral feature extraction and support vector machine (SVM). This rationale being the spectral features extraction can rapidly reduce data complexity by both targeting only the diagnostic mineral absorptions and masking those pixels complicated by vegetation, cloud and deep shade. SVM is a supervised classification method able to generate an optimal non-linear classifier with these features that generalises well even with limited training data. Key minerals targeted are serpentine, which is considered as an indicator for hydrolysed peridotitic rock, and iron oxy-hydroxides (hematite and goethite), which are considered as diagnostic of laterite development. The final classified regolith map was assessed against interpreted regolith field sites, which yielded approximately 70% similarity for all unit types, as well as against a regolith-geology map interpreted using traditional datasets (not hyperspectral imagery). Importantly, the hyperspectral derived mineral map provided much greater detail enabling a more precise understanding of the regolith-geological architecture where there are exposed soils and rocks.
Plan de classementTélédétection [126] ; Géologie et formations superficielles [064] ; Cartographie / Méthodes graphiques [128]
Descr. géo.NOUVELLE CALEDONIE
LocalisationFonds IRD [F B010071320]
Identifiant IRDfdi:010071320
Lien permanenthttp://www.documentation.ird.fr/hor/fdi:010071320

Export des données

Disponibilité des documents

Télechargment fichier PDF téléchargeable

Lien sur le Web lien chez l'éditeur

Accès réservé en accès réservé

HAL en libre accès sur HAL


Accès aux documents originaux :

Accès direct

Bureau du chercheur

Site de la documentation

Espace intranet IST (accès réservé)

Suivi des publications IRD (accès réservé)

Mentions légales

Services Horizon

Poser une question

Consulter l'aide en ligne

Déposer une publication (accès réservé)

S'abonner au flux RSS

Voir les tableaux chronologiques et thématiques

Centres de documentation

Bondy

Montpellier (centre IRD)

Montpellier (MSE)

Nouméa

Papeete

Niamey

Ouagadougou

Tunis

La Paz

Quito