Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2014, Vol. 11, No. 2, pp. 50-67
Multispectral multiresolution image synthesis using library object spectra for regularization
B.S. Zhukov
1 , M.A. Popov
2 , S.A. Stankevich
2 1 Space Research Institute RAS, Moscow, Russia
2 Scientific Centre for Aerospace Research of the Earth IGS NAS of Ukraine, Kiev, Ukraine
Comprehensive estimation of surface and atmospheric parameters from satellite observations requires synergetic interpretation of multisensor imaging data obtained in different spectral ranges with different spatial resolution. It is proposed to solve the problem of radiometrically correct synthesis of multispectral images with different spatial resolution by combining two previously developed methods: the Multisensor Multiresolution Technique (MMT) and the synthesis technique by library spectra classification. The MMT is based on classification of high-resolution (HR) images, estimation of class spectral signatures in the low-resolution (LR) bands and using them to construct an enhanced image in the LR bands with the HR pixel size. An MMT disadvantage is higher LR-signature estimation errors for small-area classes. In order to compensate for this disadvantage, it is proposed to use for regularization of the MMT algorithm the enhanced LR-images obtained by the library spectra synthesis method. This method is based on joint HR- and LR-image classification using library spectra of natural and man-made objects and on assigning weighted LR signatures to HR pixels. The combined method is applied for synthesis of images, obtained by spectroradiometer ASTER on Terra satellite in three bands of the visible and near infrared (VNIR) spectral range with a resolution of 15 m and in six bands of the shortwave infrared (SWIR) range with a resolution of 30 m. The SWIR image enhanced by the combined method shows a better sharpness than both the initial and regularization SWIR images, lower spectral signature estimation errors for small classes in comparison to the case when the MMT-method is applied without regularization, and additional spectral information that is not accounted for by the library spectra.
Keywords: sharpening techniques, multisensor data synthesis, MMT technique, synthesis technique by library spectra classification, spectroradiometer ASTER.
Full textReferences:
- Popov M.O., Stankevich S.A., Kozlova A.O., Popov M.O., Stankevich S.A., Kozlova A.O., Udoskonalena protsedura klasifіkuvannya bagatospektral'nikh aerokosmіchnikh zobrazhen' pri otsіnyuvannі bіorіznomanіttya Pіvnіchno-Prichornomors'kogo regіonu Ukraїni (Improved classification procedure of multispectral airborne and space-borne images for estimation of biodiversity in the North Black Sea region of Ukraine), Sbornik nauchnyh trudov Morskogo gidrofizicheskogo instituta, Issue 14, Sevastopol': MGI, 2006, pp. 406-410.
- Stankevich S.A., Algoritm statistichnoї klasifіkatsії ob’єktіv distantsіinogo sposterezhennya za їkh spektral'no-topologіchnimi kharakteristikami (Algorithm of statistical classification of remote observation objects by their spectro-topological characteristics), Naukovii visnik Natsional'nogo girnichogo universitetu, 2006, No. 7, pp. 38-40.
- Stankevich S.A., Sholonіk O.V., Іnstrumentarіi otsіnyuvannya ekvіvalentnoї prostorovoї rozrіznennostі bagato- ta gіperspektral'nikh tsifrovikh aerokosmіchnikh znіmkіv (Instrumentation for estimation of the equivalent spatial diversity of multi- and hyperspectral digital airborne and space-borne images), Zbіrnik naukovikh prats' Derzhavnogo naukovo-doslіdnogo іnstitutu avіatsії, Issue 3(10), Kiev: DNDIA, 2007, pp. 165-171.
- Schowengerdt R.A.,Remote sensing. Models and methods for image processing, Moscow: Tehnosfera, 2010, 560 p.
- Abrams M., Hook S., Ramachandran B., ASTER Users Handbook. Version 2, Pasadena: Jet Propulsion Laboratory, 2002, 135 p.
- Arai K., Thome K., Iwasaki A., Biggar S., ASTER VNIR and SWIR radiometric calibration and atmospheric correction, Land Remote Sensing and Global Environmental Change. NASA’s Earth Observing System and the Science of ASTER and MODIS, B. Ramachandran, C.O. Justice, M.J. Abrams (Eds), New York: Springer, 2011, pp. 83-116.
- Baldridge A.M., Hook S.J., Grove C.I., Rivera G., The ASTER spectral library version 2.0, Remote Sensing of Environment, 2009, Vol. 113, No. 4, pp. 711-715.
- Franklin S.E., Wulder M.A., Remote sensing methods in medium spatial resolution satellite data land cover classification of large areas, Progress in Physical Geography, 2002, Vol. 6, No. 2, pp. 173-205.
- King R.B., Land cover mapping principles: a return to interpretation fundamentals, International Journal of Remote Sensing, 2002, Vol. 23, No. 18, pp. 3525-3545.
- Zhukov B., Oertel D., Lanzl F., Reinhaeckel G., Unmixing-based multi-sensor multi-resolution image fusion, IEEE Trans. Geosci. Remote Sensing, 1999, Vol. 37, No. 3, pp. 1212-1226.