Centre de documentation
scientifique

Dépouillements

Factors influencing pulse width of small footprint, full wave form airborne laser scanning data / Y.C. Lin in Photogrammetric Engineering & Remote Sensing, PERS, vol 76 n° 1 (January 2010)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 76 n° 1 (January 2010) . - pp 49 - 59 Titre : Factors influencing pulse width of small footprint, full wave form airborne laser scanning data Type de document : Article/Communication Auteurs : Y.C. Lin, Auteur ; Jon P. Mills, Auteur Année de publication : 2010 Article en page(s) : pp 49 - 59 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] balayage laser [Termes IGN] classification automatique [Termes IGN] impulsion laser [Termes IGN] lidar à retour d'onde complète [Termes IGN] longueur d'onde [Termes IGN] télémétrie laser aéroporté Résumé : (Auteur) Small footprint, full waveform airborne laser scanning provides the opportunity to derive high-resolution geometric and physical information simultaneously from a single scanner system. This study evaluates the influence of various factors on the shape of the returned waveform and investigates the possibility of improving terrain classification by applying waveform-derived information. The factors discussed are surface roughness, slope angle, scan angle, amplitude, and footprint size. It is statistically demonstrated that roughness is the most significant factor affecting pulse width, and that, over relatively smooth surfaces, there is no significant variation in pulse width behavior resulting from different footprint sizes. Pulse width also exhibits a relatively stable behavior when amplitude, range distance, or scan angle vary substantially. The overall accuracy of classification achieved by applying pulse width information over all the different land-cover types examined in this study (including scrub, hillside, single trees, and forest areas) was greater than 85 percent, with > 94 percent achieved for open vegetation areas. Physical surface information provided by small footprint waveform data is considered to be at the microscale, therefore it is recommended to combine such information with geometry (e.g., filtering algorithms) for the optimal identification of terrain points. Copyright ASPRS Numéro de notice : A2010-013 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.76.1.49 En ligne : https://doi.org/10.14358/PERS.76.1.49 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=30209 [article]

A network representation of raster land-cover patches / L. De Cola in Photogrammetric Engineering & Remote Sensing, PERS, vol 76 n° 1 (January 2010)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 76 n° 1 (January 2010) . - pp 61 - 72 Titre : A network representation of raster land-cover patches Type de document : Article/Communication Auteurs : L. De Cola, Auteur Année de publication : 2010 Article en page(s) : pp 61 - 72 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] analyse comparative [Termes IGN] autocorrélation spatiale [Termes IGN] classification automatique [Termes IGN] données maillées [Termes IGN] structure de données localisées [Termes IGN] système d'information géographique Résumé : (Auteur) Network models, based on mathematical graph theory, are used in many fields, from chemistry and biology to ecology and sociology. A nominal-valued land-cover raster can be represented as a network in which patches are nodes connected by links if two patches are adjacent. Built with existing GIS technology, the system is applied to the largest patches of a raster for the Washington, D.C. area in order to provide a data structure that visualizes the gross spatial structure of the raster; provides information about the overall arrangement of the patches and their adjacencies; and generates new metrics about land-cover patterns, connectivity and spatial autocorrelation. The system facilitates the evaluation of land-cover classifications, the measurement of landscape change, comparisons of data from different regions, and the analysis of land/environmental interactions. Copyright ASPRS Numéro de notice : A2010-014 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.76.1.61 En ligne : https://doi.org/10.14358/PERS.76.1.61 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=30210 [article]