AC-WGAN-GP: Generating Labeled Samples for Improving Hyperspectral Image Classification with Small-Samples
AC-WGAN-GP: Generating Labeled Samples for Improving Hyperspectral Image Classification with Small-Samples
复制标题
AC-WGAN-GP：生成标记样本以改进小样本的高光谱图像分类
DOI：
10.3390/rs14194910
复制
发表时间：
2022-10
期刊：
影响因子：
5
通讯作者：
Jinhua Zhang
中科院分区：
文献类型：
--
作者：
Caihao Sun;Xiaohua Zhang;Hongyun Meng;Xianghai Cao;Jinhua Zhang
标签样本的缺乏严重制约了深度学习在高光谱图像分类中的分类性能。为了解决这一问题，生成性对抗网络(GAN)通常被用于数据增强。然而，GaN在这项任务中存在一些问题，如生成的样本质量较差，训练过程不稳定。因此，如何构造GAN来生成高质量的高光谱训练样本，对于高光谱数据的小样本分类任务具有重要意义。提出了一种基于辅助分类器的梯度惩罚Wasserstein GAN(AC-WGAN-GP)。该框架包括AC-WGAN-GP、在线生成机制和样本选择算法。提出的方法具有以下显著优点。首先，生成器的输入由先验知识指导，并在AC-WGAN-GP的体系结构中引入单独的分类器来产生可靠的标签。第二，在线生成机制确保生成的样本的多样性。第三，选择与真实数据相似的生成样本。在三个公开的高光谱数据集上的实验表明，生成的样本与实际样本的分布相同，并具有足够的多样性，有效地扩展了训练集。与其他有竞争力的分类方法相比，该框架利用少量的标记样本获得了更好的分类精度。
The lack of labeled samples severely restricts the classification performance of deep learning on hyperspectral image classification. To solve this problem, Generative Adversarial Networks (GAN) are usually used for data augmentation. However, GAN have several problems with this task, such as the poor quality of the generated samples and an unstable training process. Thereby, knowing how to construct a GAN to generate high-quality hyperspectral training samples is meaningful for the small-sample classification task of hyperspectral data. In this paper, an Auxiliary Classifier based Wasserstein GAN with Gradient Penalty (AC-WGAN-GP) was proposed. The framework includes AC-WGAN-GP, an online generation mechanism, and a sample selection algorithm. The proposed method has the following distinctive advantages. First, the input of the generator is guided by prior knowledge and a separate classifier is introduced to the architecture of AC-WGAN-GP to produce reliable labels. Second, an online generation mechanism ensures the diversity of generated samples. Third, generated samples that are similar to real data are selected. Experiments on three public hyperspectral datasets show that the generated samples follow the same distribution as the real samples and have enough diversity, which effectively expands the training set. Compared to other competitive methods, the proposed framework achieved better classification accuracy with a small number of labeled samples.
登录
查看更多内容
影响因子：
4.8
作者：
Lin Zhao-;Wenqiang Luo;Qiming Liao;Siyuan Chen;Jianhui Wu
通讯作者：
Lin Zhao-;Wenqiang Luo;Qiming Liao;Siyuan Chen;Jianhui Wu
影响因子：
8.2
作者：
Sikang Hou;Hongye Shi;Xianghai Cao;Xiaohua Zhang;Licheng Jiao
通讯作者：
Licheng Jiao
影响因子：
3.9
作者：
Wang, Zesong;Zou, Cui;Cai, Weiwei
通讯作者：
Cai, Weiwei
影响因子：
14.9
作者：
N. Nasrabadi
通讯作者：
N. Nasrabadi
DOI：
10.1109/icccnt56998.2023.10306417
发表时间：
2022-02
期刊：
2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT)
影响因子：
--
作者：
Gilad Cohen;Raja Giryes
通讯作者：
Gilad Cohen;Raja Giryes