Abstract
Hyperspectral image classification is a hot issue in the field of remote sensing. It is possible to achieve high accuracy and strong generalization through a good classification method that is used to process image data. In this paper, an efficient hyperspectral image classification method based on improved Rotation Forest (ROF) is proposed. It is named ROF-KELM. Firstly, Non-negative matrix factorization( NMF) is used to do feature segmentation in order to get more effective data. Secondly, kernel extreme learning machine (KELM) is chosen as base classifier to improve the classification efficiency. The proposed method inherits the advantages of KELM and has an analytic solution to directly implement the multiclass classification. Then, Q-statistic is used to select base classifiers. Finally, the results are obtained by using the voting method. Three simulation examples, classification of AVIRIS image, ROSIS image and the UCI public data sets respectively, are conducted to demonstrate the effectiveness of the proposed method.
Highlights
Remote sensing technology is a non-contact and long distance detection technology
To solve the problem of hyperspectral remote sensing images data classification, this paper proposes a classification algorithm based on improved Rotation Forest, namely ROF-kernel extreme learning machine (KELM)
To verify the performance of ROF-KELM algorithm, we did an experiment using hyperspectral remote sensing data called AVIRIS obtained from the airborne visible infrared imaging spectroradiometer of NASA
Summary
Remote sensing technology is a non-contact and long distance detection technology. With the development of Internet of Things (IoT) technology [1,2], the field of remote sensing shows new vitality, more and more remote sensing information can be obtained, such as low-resolution remote sensing images, hyperspectral remote sensing images and so on. Hyperspectral remote sensing data has a great connection with adjacent bands, so all bands are not guaranteed high accuracy at the same time [7,8] According to these limitations, some new methods are needed to improve the algorithm performance. On the basis of summarizing hyperspectral remote sensing classification technology and ensemble algorithm, this paper discusses the classification problem of hyperspectral image data based on ensemble method. In hyperspectral image processing field, Pal et al [16] applied ELM based on kernel to classify remote sensing image, and it gives a better result than support vector machine (SVM) and some other neural network frameworks [17]. To solve the problem of hyperspectral remote sensing images data classification, this paper proposes a classification algorithm based on improved Rotation Forest, namely ROF-KELM.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
