PROJECT TITLE :

A novel MRF-based multi feature fusion for Classification of remote sensing images - 2016

ABSTRACT:

The spatial info has been proved to be effective in improving the performance of spectral-based mostly classification. However, it's tough to describe different image scenes by using monofeature owing to complexity of the geospatial scenes. During this letter, a novel framework is developed to combine the multiple spectral and spatial features based on the Markov random field (MRF). Specifically, the pixels in a picture are separated into reliable and unreliable ones in line with the decision of multifeature classifications. The labels of the reliable pixels will be conveniently determined, but the unreliable pixels are then classified by fusing the multifeature classification results and reducing the classification uncertainties based on the MRF optimization. Experiments are conducted on three multispectral high-resolution pictures to verify the effectiveness of the proposed methodology. Many state-of-the-art multifeature classification strategies are also achieved for the aim of comparison. Moreover, 3 classifiers (i.e., multinomial logistic regression, support vector machines, and random forest) are used to test the performance of the proposed framework. It is shown that the proposed technique will effectively integrate multiple features, yield promising results, and outperform alternative approaches compared.