A Hyperspectral Image Classification Method Based on Multi-Discriminator Generative Adversarial Networks

Abstract

Hyperspectral remote sensing images (HSIs) have great research and application value. At present, deep learning has become an important method for studying image processing. The Generative Adversarial Network (GAN) model is a typical network of deep learning developed in recent years and the GAN model can also be used to classify HSIs. However, there are still some problems in the classification of HSIs. On the one hand, due to the existence of different objects with the same spectrum phenomenon, if only according to the original GAN model to generate samples from spectral samples, it will produce the wrong detailed characteristic information. On the other hand, the gradient disappears in the original GAN model and the scoring ability of a single discriminator limits the quality of the generated samples. In order to solve the above problems, we introduce the scoring mechanism of multi-discriminator collaboration and complete semi-supervised classification on three hyperspectral data sets. Compared with the original GAN model with a single discriminator, the adjusted criterion is more rigorous and accurate and the generated samples can show more accurate characteristics. Aiming at the pattern collapse and diversity deficiency of the original GAN generated by single discriminator, this paper proposes a multi-discriminator generative adversarial networks (MDGANs) and studies the influence of the number of discriminators on the classification results. The experimental results show that the introduction of multi-discriminator improves the judgment ability of the model, ensures the effect of generating samples, solves the problem of noise in generating spectral samples and can improve the classification effect of HSIs. At the same time, the number of discriminators has different effects on different data sets.

Structure Aware Generative Adversarial Networks for Hyperspectral Image Classification

“…In the recent study, authors suggested to use collaborating learning and attention mechanism to assists the generator to provide real VTS [34]. Gao et al [35] proposed using multidiscriminator with scoring mechanism rather than single discriminator to overcome the problem of insufficient diversity of generated samples. A training sample generated by the generator might partially be real and partially be fake.…”

Hyperspectral image spectral-spatial classification via weighted Laplacian smoothing constraint-based sparse representation

“…As a powerful image-processing tool, the support vector machine (SVM) [11][12][13][14] and sparse representation (SR) model [15,16] and its derivative model have attracted much attention for HSI classification [17][18][19][20]. However, the noise and mixed spectral information in HSI cause several theoretical and practical challenges for pixel-wise classification [21][22][23].…”

Generative Adversarial Networks Based on Collaborative Learning and Attention Mechanism for Hyperspectral Image Classification

“…Recently, several researchers have tried to use GAN for HSI classification. GAN-based HSI classification methods focus on semi-supervised GANs [51][52][53][54][55][56][57] and spatial-spectral GANs [58,59]. In semi-supervised GAN methods, some methods were proposed by combining GAN with the traditional techniques, such as conditional random fields [51] and 3D bilateral filter [52].…”

“…Later, improved HSGAN methods [54,55] were proposed by adding the majority voting or the dynamic neighborhood voting strategies for classification. Gao et al [56] proposed a semi-supervised multi-discriminator GANs (MDGANs) to improve the judgment ability by averaging the results of multiple discriminators. In spatial-spectral GAN methods, Zhu et al [57] proposed a 3D-GAN method to use both the spatial and spectral information of HSIs.…”