Centre de documentation
scientifique

Mapping active paddy rice area over monsoon asia using time-series Sentinel-2 images in Google earth engine : a case study over lower gangetic plain / Arabinda Maiti in Geocarto international, vol 38 n° inconnu ([01/01/2023])  

Public [article]  inGeocarto international > vol 38 n° inconnu [01/01/2023] Titre : Mapping active paddy rice area over monsoon asia using time-series Sentinel-2 images in Google earth engine : a case study over lower gangetic plain Type de document : Article/Communication Auteurs : Arabinda Maiti, Auteur ; Prasenjit Acharya, Auteur ; Srikanta Sannigrahi, Auteur ; et al., Auteur Année de publication : 2023 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications de télédétection [Termes IGN] carte thématique [Termes IGN] Gange (fleuve) [Termes IGN] Google Earth Engine [Termes IGN] image Sentinel-SAR [Termes IGN] Inde [Termes IGN] mousson [Termes IGN] plaine [Termes IGN] rizière Résumé : (auteur) We proposed a modification of the existing approach for mapping active paddy rice fields in monsoon-dominated areas. In the existing PPPM approach, LSWI higher than EVI at the transplantation stage enables the identification of rice fields. However, it fails to recognize the fields submerged later due to monsoon floods. In the proposed approach (IPPPM), the submerged fields, at the maximum greenness time, were excluded for better estimation. Sentinel–2A/2B time-series images were used for the year 2018 to map paddy rice over the Lower Gangetic Plain (LGP) using Google earth engine (GEE). The overall accuracy (OA) obtained from IPPPM was 85%. Further comparison with the statistical data reveals the IPPPM underestimates (slope (β1) = 0.77) the total reported paddy rice area, though R2 remains close to 0.9. The findings provide a basis for near real-time mapping of active paddy rice areas for addressing the issues of production and food security. Numéro de notice : A2022-924 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1080/10106049.2022.2032396 En ligne : https://doi.org/10.1080/10106049.2022.2032396 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=99963 [article]

Effect of label noise in semantic segmentation of high resolution aerial images and height data / Arabinda Maiti in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol V-2-2022 (2022 edition)  

Public [article]  inISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences > vol V-2-2022 (2022 edition) . - pp 275 - 282 Titre : Effect of label noise in semantic segmentation of high resolution aerial images and height data Type de document : Article/Communication Auteurs : Arabinda Maiti, Auteur ; Sander J. Oude Elberink, Auteur ; M. George Vosselman, Auteur Année de publication : 2022 Article en page(s) : pp 275 - 282 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage profond [Termes IGN] bruit (théorie du signal) [Termes IGN] données altimétriques [Termes IGN] données d'entrainement (apprentissage automatique) [Termes IGN] image à très haute résolution [Termes IGN] image aérienne [Termes IGN] orthoimage [Termes IGN] segmentation sémantique Résumé : (auteur) The performance of deep learning models in semantic segmentation is dependent on the availability of a large amount of labeled data. However, the influence of label noise, in the form of incorrect annotations, on the performance is significant and mostly ignored. This is a big concern in remote sensing applications, wherein acquired datasets are spatially limited, labeling is done by domain experts with possible sources of high inter-and intra-observer variability leading to erroneous predictions. In this paper, we first simulate the label noise while conducting experiments on two different datasets with very high-resolution aerial images, height data, and inaccurate labels, responsible for the training of deep learning models. We then focus on the effect of these noises on the model performance. Different classes respond differently to the label noise. The typical size of an object belonging to a class is a crucial factor regarding the class-specific performance of the model trained with erroneous labels. Errors caused by relative shifts of labels are the most influential label errors. The model is generally more tolerant of the random label noise than other label errors. It has been observed that the accuracy gets reduced by at least 3% while 5% of label pixels are erroneous. In this regard, our study provides a new perspective of evaluating and quantifying the propagation of label noise in the model performance that is indeed important for adopting reliable semantic segmentation practices. Numéro de notice : A2022-434 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE Nature : Article DOI : 10.5194/isprs-annals-V-2-2022-275-2022 Date de publication en ligne : 17/05/2022 En ligne : https://doi.org/10.5194/isprs-annals-V-2-2022-275-2022 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100741 [article]