6533b86efe1ef96bd12cbe88
RESEARCH PRODUCT
Robustified smoothing for enhancement of thermal image sequences affected by clouds
Maurizio LongoPaolo AddessoAntonino MalteseGemine VivoneRita MontoneRocco Restainosubject
Cloud MaskingEarth observationComputer scienceSharpeningBayesian SmoothingRobustness (computer science)Multitemporal AnalysiThermalRobustneSettore AGR/08 - Idraulica Agraria E Sistemazioni Idraulico-ForestaliComputer visionRobustnessImage resolutionMultitemporal AnalysisPixelbusiness.industrySettore ING-INF/03 - TelecomunicazioniSettore ICAR/02 - Costruzioni Idrauliche E Marittime E IdrologiaComputer Science Applications1707 Computer Vision and Pattern RecognitionBayesian Smoothing; Cloud Masking; Multitemporal Analysis; Robustness; Thermal Sharpening; Earth and Planetary Sciences (all); Computer Science Applications1707 Computer Vision and Pattern RecognitionThermal SharpeningArtificial intelligencebusinessEarth and Planetary Sciences (all)SmoothingSettore ICAR/06 - Topografia E CartografiaInterpolationdescription
Obtaining radiometric surface temperature information with both high acquisition rate and high spatial resolution is still not possible through a single sensor. However, in several earth observation applications, the fusion of data acquired by different sensors is a viable solution for so called image sharpening. A related issue is the presence of clouds, which may impair the performance of the data fusion algorithms. In this paper we propose a robustified setup for the sharpening of thermal images in a non real-time scenario, capable to deal with missing thermal data due to cloudy pixels, and robust with respect to cloud mask misclassifications. The effectiveness of the presented technique is assessed via numerical simulations based on SEVIRI data.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2015-07-01 |