000121191 001__ 121191
000121191 005__ 20240319081019.0
000121191 0247_ $$2doi$$a10.1109/TGRS.2022.3205119
000121191 0248_ $$2sideral$$a130613
000121191 037__ $$aART-2022-130613
000121191 041__ $$aeng
000121191 100__ $$0(orcid)0000-0002-7057-4283$$aAlcolea, Adrián$$uUniversidad de Zaragoza
000121191 245__ $$aBayesian neural networks to analyze hyperspectral datasets using uncertainty metrics
000121191 260__ $$c2022
000121191 5060_ $$aAccess copy available to the general public$$fUnrestricted
000121191 5203_ $$aMachine learning techniques, and specifically neural networks, have proved to be very useful tools for image classification tasks. Nevertheless, measuring the reliability of these networks and calibrating them accurately are very complex. This is even more complex in a field like hyperspectral imaging, where labeled data are scarce and difficult to generate. Bayesian neural networks (BNNs) allow to obtain uncertainty metrics related to the data processed (aleatoric), and to the uncertainty generated by the model selected (epistemic). On this work, we will demonstrate the utility of BNNs by analyzing the uncertainty metrics obtained by a BNN over five of the most used hyperspectral images datasets. In addition, we will illustrate how these metrics can be used for several practical applications such as identifying predictions that do not reach the required level of accuracy, detecting mislabeling in the dataset, or identifying when the predictions are affected by the increase of the level of noise in the input data.
000121191 536__ $$9info:eu-repo/grantAgreement/ES/AEI/PID2019-105660RB-C21-AEI-10.13039-501100011033$$9info:eu-repo/grantAgreement/ES/DGA-ESF/T58-20R
000121191 540__ $$9info:eu-repo/semantics/openAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/
000121191 590__ $$a8.2$$b2022
000121191 592__ $$a2.4$$b2022
000121191 591__ $$aENGINEERING, ELECTRICAL & ELECTRONIC$$b27 / 274 = 0.099$$c2022$$dQ1$$eT1
000121191 591__ $$aIMAGING SCIENCE & PHOTOGRAPHIC TECHNOLOGY$$b5 / 28 = 0.179$$c2022$$dQ1$$eT1
000121191 591__ $$aREMOTE SENSING$$b4 / 34 = 0.118$$c2022$$dQ1$$eT1
000121191 591__ $$aGEOCHEMISTRY & GEOPHYSICS$$b5 / 87 = 0.057$$c2022$$dQ1$$eT1
000121191 593__ $$aElectrical and Electronic Engineering$$c2022$$dQ1
000121191 593__ $$aEarth and Planetary Sciences (miscellaneous)$$c2022$$dQ1
000121191 594__ $$a10.9$$b2022
000121191 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/acceptedVersion
000121191 700__ $$0(orcid)0000-0002-7532-2720$$aResano, Javier$$uUniversidad de Zaragoza
000121191 7102_ $$15007$$2035$$aUniversidad de Zaragoza$$bDpto. Informát.Ingenie.Sistms.$$cÁrea Arquit.Tecnología Comput.
000121191 773__ $$g60 (2022), 5537810 [10 pp.]$$pIEEE trans. geosci. remote sens.$$tIEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING$$x0196-2892
000121191 8564_ $$s13188744$$uhttps://zaguan.unizar.es/record/121191/files/texto_completo.pdf$$yPostprint
000121191 8564_ $$s4009682$$uhttps://zaguan.unizar.es/record/121191/files/texto_completo.jpg?subformat=icon$$xicon$$yPostprint
000121191 909CO $$ooai:zaguan.unizar.es:121191$$particulos$$pdriver
000121191 951__ $$a2024-03-18-15:58:12
000121191 980__ $$aARTICLE