Spectral Inpainting for the Restoration of Missing Data from Multispectral Satellite Sensors: Case study on Aqua MODIS Band 6

Marouan Bouali (1) , Saïd Ladjal (1, 2)

1 LTCI - Laboratoire Traitement et Communication de l'Information
2 TSI - Département Traitement du Signal et des Images

In this letter, we introduce an algorithm for the restoration of missing data from multi- spectral satellite imagery. The proposed approach combines two simple principles; non local or neighborhood filters used in the context of still image denoising/inpainting and spectral matching techniques based on spectral similarity measures and required for the classification of hyperspectral images. The resulting semi-physical approach, refered to as spectral inpaint- ing, is applied to the particular issue of Aqua Moderate Resolution Imaging Spectroradiometer (MODIS) band 6 (1628-1652 µm) non functional detectors and is shown to provide very good results compared to standard interpolation techniques.

Mots clés: Multispectral; image restoration; inpainting

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Marouan Bouali, Saïd Ladjal. Spectral Inpainting for the Restoration of Missing Data from Multispectral Satellite Sensors: Case study on Aqua MODIS Band 6. 2011. ⟨hal-00639083v2⟩