Centre de documentation
scientifique

An assessment of algorithmic parameters affecting image classification accuracy by random forests / Dee Shi in Photogrammetric Engineering & Remote Sensing, PERS, vol 82 n° 6 (June 2016)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 82 n° 6 (June 2016) . - pp 407 - 417 Titre : An assessment of algorithmic parameters affecting image classification accuracy by random forests Type de document : Article/Communication Auteurs : Dee Shi, Auteur ; Xiaojun Yang, Auteur Année de publication : 2016 Article en page(s) : pp 407 - 417 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] algorithme d'apprentissage [Termes IGN] classification [Termes IGN] classification par forêts d'arbres décisionnels [Termes IGN] impact sur les données [Termes IGN] occupation du sol [Termes IGN] précision de la classification Résumé : (Auteur) Random forests as a promising ensemble learning algorithm have been increasingly used for remote sensor image classification, and are found to perform identical or better than some popular classifiers. With only two algorithmic parameters, they are relatively easier to implement. Existing literature suggests that the performance of random forests is insensitive to changing algorithmic parameters. However, this was largely based on the classifier's accuracy that does not necessarily represent the resulting thematic map accuracy. The current study extends beyond the classifier's accuracy assessment and investigate how the algorithmic parameters could affect the resulting thematic map accuracy by random forests. A set of random forest models with different parameter settings was carefully constructed and then used to classify a satellite image into multiple land cover categories. Both the classifier's accuracy and the map accuracy were assessed. The results reveal that these parameters can affect the map accuracy up to 9 ∼16 percent for some classes, although their impact on the classifier's accuracy was quite limited. A careful parameterization prioritizing thematic map accuracy can help improve the performance of random forests in image classification, especially for spectrally complex land cover classes. These findings can help establish practical guidance on the use of random forests in the remote sensing community. Numéro de notice : A2016-440 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.82.6.407 En ligne : http://dx.doi.org/10.14358/PERS.82.6.407 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=81345 [article]