Search

The applications of extreme learning machine (ELM) to the hyperspectral-image (HSI) classification have attracted a great deal of research attention because of its excellent performance and fast learning speed. However, conventional ELM is unable to achieve satisfactory accuracy since it only exploits the spectral information to conduct the HSI classification. To address the above issues, we propose a novel classification algorithm based on both spectral and multiscale spatial information, referred to as ELM with enhanced composite feature (ELM-ECF). To be specific, we adopt the original ELM, exploit a multiscale spatial weighted-mean-filtering-based approach to extract multiple spatial information, and use the majority vote method to select the final classification result. The proposed ELM-ECF significantly improves the classification accuracy of the original ELM. Experimental results on three public HSIs (i.e., Indian Pines data set, Pavia University data set, and Salinas data set) illustrate that the proposed ELM-ECF outperforms a variety of the state-of-the-art HSI classification counterparts in terms of classification accuracy.

As a new machine learning method, extreme learning machine (ELM) has received wide attention for image classification due to its good performances. Since ELM cannot catch the spatial information, the results of classification is not good while applying to hyperspectral image (HSI) classification. In view of this, this paper proposes a new method for HSI classification by combining ELM with Loopy Belief Propagation (LBP). The proposed method can not only reduce time-consuming, but also improve the accuracy of classification greatly. The experimental result in HSI data set of Indian Pines demonstrates that the proposed method outperforms several classical algorithms.