Spectral-Spatial Joint Classification of Hyperspectral Image Based on Broad Learning System

Abstract

At present many researchers pay attention to a combination of spectral features and spatial features to enhance hyperspectral image (HSI) classification accuracy. However, the spatial features in some methods are utilized insufficiently. In order to further improve the performance of HSI classification, the spectral-spatial joint classification of HSI based on the broad learning system (BLS) (SSBLS) method was proposed in this paper; it consists of three parts. Firstly, the Gaussian filter is adopted to smooth each band of the original spectra based on the spatial information to remove the noise. Secondly, the test sample’s labels can be obtained using the optimal BLS classification model trained with the spectral features smoothed by the Gaussian filter. At last, the guided filter is performed to correct the BLS classification results based on the spatial contextual information for improving the classification accuracy. Experiment results on the three real HSI datasets demonstrate that the mean overall accuracies (OAs) of ten experiments are 99.83% on the Indian Pines dataset, 99.96% on the Salinas dataset, and 99.49% on the Pavia University dataset. Compared with other methods, the proposed method in the paper has the best performance.

An Assimilating Model Using Broad Learning System for Incorporating Multi‐Source Precipitation Data With Environmental Factors Over Southeast China

“…BLS has been successfully applied in the fields of facial expression recognition, fault diagnosis, and classification of the hyperspectral image, etc (Zhang et al, 2018;X. Zhao et al, 2019;G. Zhao et al, 2021).…”

“…Compared to deep learning, the broad learning system does not need to train the stacks of hierarchical layers which leads to expensive computational cost (Schmidhuber, 2015), because BLS can be trained incrementally based on the information from the trained architecture (Kuok & Yuen, 2020). BLS has been successfully applied in the fields of facial expression recognition, fault diagnosis, and classification of the hyperspectral image, etc (Zhang et al., 2018; X. Zhao et al., 2019; G. Zhao et al., 2021). Nevertheless, the performance of this method in fusing multi‐source precipitation data has not been tested.…”

Hierarchical broad learning system for hyperspectral image classification

“…It combines wavelet, BLS and Gabor filters, and can provide more effective spectral-spatial features by using hierarchical structure. Moreover, different from [47] and [48], the The structure of the rest of this paper is as follows. In the following Section 2, the proposed HBLS will be described in detail.…”

“…However, BLS cannot directly extract the spectral-spatial feature of HSI [41]. To take full advantage of spatial information, some spectral-spatial classification methods based on BLS have been proposed [47,48]. In [47], a new method based on the discriminative locality preserving broad learning system was proposed for hyperspectral image classification by exploiting the manifold structure between neighbouring pixels of hyperspectral image.…”

“…In [47], a new method based on the discriminative locality preserving broad learning system was proposed for hyperspectral image classification by exploiting the manifold structure between neighbouring pixels of hyperspectral image. In [48], the Gaussian filter was adopted to remove the noise, and the guided filter was performed to correct the BLS classification results based on the spatial contextual information for improving the classification accuracy.…”