شماره ركورد كنفرانس :
3297
عنوان مقاله :
An Improved Cartoon+Texture Decomposition Based Pansharpening Method
عنوان به زبان ديگر :
An Improved Cartoon+Texture Decomposition Based Pansharpening Method
پديدآورندگان :
Lotfi Meysam Image Processing and Information Analysis Lab. Faculty of Electrical and Computer Engineering Tarbiat Modares University - Tehran - Iran , Ghassemian Hassan Image Processing and Information Analysis Lab. Faculty of Electrical and Computer Engineering Tarbiat Modares University - Tehran - Iran
كليدواژه :
low pass filter , texture , cartoon , image decomposition , Pansharpening
عنوان كنفرانس :
نوزدهمين سمپوزيوم بين المللي هوش مصنوعي و پردازش سيگنال
چكيده لاتين :
Pansharpening is the most widely used fusion method,
in the field of remote sensing, to increase spatial information of the
multispectral image while preserving spectral signatures. Based
on the nature of spatial and spectral information, there is a lack of
correlation between them. Therefore, separation of them can be
considered as an image decomposition to uncorrelated
components. Recently, the cartoon+texture decomposition was
used in the pansharpening and decrease spectral distortion.
However, details have not been strengthened enough. Therefore,
in this paper we aim to use a filter based detail extraction to
improve spatial information while mitigate spectral distortion.
