Abstract
The paper addresses problems related to classification of images obtained by various types of remote sensing devices. Development and use of Bayes type land cover classifiers based on multidimensional Gaussian, Dirichlet and gamma distributions is analysed and compared on the basis of sample data from RGB and hyperspectral thermal sensing devices with unequal spatial resolution. Approaches to data fusion for design of the combined classifiers are presented including the cases where different families of multidimensional distributions are used to model the sensor data and classifiers are designed using combinations of their probability density functions. The best classification results are obtained when the fusion of data from both images is used together with classification based on all three considered distributions combined together. DOI: http://dx.doi.org/10.5755/j01.eie.22.4.12600
Highlights
Development of accurate and fast methods for classification of remotely sensed image pixels to predefined classes remains one of the major tasks in Earth observation from satellites and planes
As in [1], it was done on the basis of the “subset” images of the data set “grss_dfc_2014” presented for the 2014 IEEE GRSS Data Fusion Contest (DFC) [4]
Classifiers Ki, ( i 1, 3 is the index of used probability distribution; K1 is the classifier based on Gaussian distribution; K2 is the classifier based on Dirichlet distribution; K3 is the classifier based on gamma distribution) targeting differentiation of pixels within the set S between 7 categories are constructed, using only information from image I
Summary
Development of accurate and fast methods for classification of remotely sensed image pixels to predefined classes remains one of the major tasks in Earth observation from satellites and planes. In the paper [1], Bayes classification approach to this task was presented, based on the assumption that each class can be interpreted as a sample realisation from the multidimensional universe with Gaussian distribution. In this study an attempt was made to use different multidimensional distributions [2], [3], develop related classifiers and data fusion approaches, and perform comparative analysis. As in [1], it was done on the basis of the “subset” images of the data set “grss_dfc_2014” presented for the 2014 IEEE GRSS Data Fusion Contest (DFC) [4]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
