A comparison study on fusion methods using evaluation indicators

Karathanassi, V; Kolokousis, P; Ioannidou, S

dc.contributor.author	Karathanassi, V	en
dc.contributor.author	Kolokousis, P	en
dc.contributor.author	Ioannidou, S	en
dc.date.accessioned	2014-03-01T01:25:39Z
dc.date.available	2014-03-01T01:25:39Z
dc.date.issued	2007	en
dc.identifier.issn	0143-1161	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/17706
dc.subject.classification	Remote Sensing	en
dc.subject.classification	Imaging Science & Photographic Technology	en
dc.subject.other	MULTISENSOR IMAGE FUSION	en
dc.subject.other	LANDSAT TM	en
dc.subject.other	MULTIRESOLUTION	en
dc.subject.other	QUALITY	en
dc.title	A comparison study on fusion methods using evaluation indicators	en
heal.type	journalArticle	en
heal.identifier.primary	10.1080/01431160600606890	en
heal.identifier.secondary	http://dx.doi.org/10.1080/01431160600606890	en
heal.language	English	en
heal.publicationDate	2007	en
heal.abstract	Various fusion methods have been developed for improving data spatial resolution. The methods most encountered in the literature are the intensityhue-saturation (IHS) transform, the Brovey transform, the principal components algorithm (PCA) fusion method, the Gram-Schmidt fusion method, the local mean matching method, the local mean and variance matching method, the least square fusion method, the discrete wavelet fusion method including Daubechies, Symlet, Coiflet, biorthogonal spline, reverse biorthogonal spline, and Meyer wavelets, the wavelet-PCA fusion method, and the crossbred IHS and wavelet fusion method. Using various evaluation indicators such as two-dimensional correlation, relative difference of means, relative variation, deviation index, entropy difference, peak signal-to-noise ratio index and universal image quality index, as well as photo-interpretation methods and techniques, results of the above fusion methods were compared and comments on the fusion methods and potential of evaluation indicators were made. Among data fusion methods and indicators the local mean and variance matching methods proved the most efficient and the peak signal-to-noise ratio indicator proved the most appropriate for the evaluation of data fusion results.	en
heal.publisher	TAYLOR & FRANCIS LTD	en
heal.journalName	INTERNATIONAL JOURNAL OF REMOTE SENSING	en
dc.identifier.doi	10.1080/01431160600606890	en
dc.identifier.isi	ISI:000247376600017	en
dc.identifier.volume	28	en
dc.identifier.issue	10	en
dc.identifier.spage	2309	en
dc.identifier.epage	2341	en