Residual deep PCA-based feature extraction for hyperspectral image classification
- PDF / 1,603,828 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 100 Downloads / 307 Views
(0123456789().,-volV)(0123456789(). ,- volV)
EXTREME LEARNING MACHINE AND DEEP LEARNING NETWORKS
Residual deep PCA-based feature extraction for hyperspectral image classification Minchao Ye1
•
Chenxi Ji1 • Hong Chen1 • Ling Lei1 • Huijuan Lu1 • Yuntao Qian2
Received: 28 November 2018 / Accepted: 12 September 2019 Ó Springer-Verlag London Ltd., part of Springer Nature 2019
Abstract In hyperspectral image (HSI) classification, a big challenge is the limited sample size with a relatively high feature dimension. Therefore, effective feature extraction of data is essential, which is desired to remove the redundancy as well as improve the discrimination. A huge number of methods have been proposed for HSI feature extraction. In recent years, deep learning-based feature extraction algorithms have shown their superiorities in various classification problems. Within them, deep PCA (DPCA) is a simple but efficient algorithm, which runs fast due to the absence of back-propagation. However, DPCA fails to provide satisfactory classification accuracies on HSI datasets. In this paper, we try to combine DPCA with residual-based multi-scale feature extraction and propose a residual deep PCA (RDPCA) feature extraction algorithm for HSI classification. It is a hierarchical approach consisting of multiple layers. Within each layer, PCA is utilized for layer-wise feature extraction, and the reconstruction residual is fed into the next layer. When the feature is passed deeper into the RDPCA network, finer details are mined. The layer-wise features are concatenated to form the final output feature. Furthermore, to enhance the ability of nonlinear feature extraction, we add activation functions between adjacent layers. Experimental results on real-world HSI datasets have shown the superiority of the proposed RDPCA over DPCA and PCA. Keywords Hyperspectral image Feature extraction Residual deep PCA
1 Introduction Hyperspectral images (HSIs) are characterized in hundreds of narrow spectral bands (channels) with high spectral resolution. In recent years, HSIs have been widely adopted in remote sensing applications, including urban mapping, environmental management, crop analysis, etc. Different from ordinary optical images, the representation with high spectral resolution gives HSIs the ability of distinguishing one material from another [1]; hence, the researches and applications on HSIs are of great significance. Taking & Huijuan Lu [email protected] Minchao Ye [email protected] 1
College of Information Engineering, China Jiliang University, Hangzhou, China
2
College of Computer Science, Zhejiang University, Hangzhou, China
advantage of the rich spectral information, pixel-wise classification of HSI data has been developed for a variety of applications [2, 3]. Feature extraction plays an important role in HSI classification. Whether the extracted features are expressive or not has strong influence on the classification accuracy. Hence, feature extraction has been a hot research topic of HSI for a long period, and many dif
Data Loading...
