In recent years, convolutional neural networks (CNNs) have been widely used in hyperspectral images (HSIs) classification. However, 2-D CNN, 3-D CNN, and even the newly emerged hybrid CNN (HCNN) all require multiple or deep CNN layers to obtain excellent classification performance, which inevitably results in the high complexity and the need for a large number of training samples. Moreover, as a local operator, convolution is challenging to fully use global information. To solve the above two issues, we design a HCNN based on global reasoning (GloRe-HCNN) for HSI classification. On the one hand, the GloRe-HCNN uses only one layer of 3-D CNN and one layer of 2-D CNN to jointly extract the spatial–spectral features of HSI. On the other hand, we contrive a spatial–spectral global reasoning unit (SS-GloRe-Unit) to take the place of stacked multilayer 3-D CNN for extracting global features fully. We select small training samples in three standard datasets and compare them with state-of-the-art CNN methods. Numerous experiments show that our GloRe-HCNN performs advanced performance.

中文翻译：

---

基于全局推理的混合 CNN 高光谱图像分类


近年来，卷积神经网络（CNN）已广泛应用于高光谱图像（HSI）分类。然而，2-D CNN、3-D CNN，甚至新出现的混合 CNN (HCNN) 都需要多个或较深的 CNN 层才能获得优异的分类性能，这不可避免地导致复杂度较高，并且需要大量的训练样本。此外，作为局部算子，卷积对于充分利用全局信息具有挑战性。为了解决上述两个问题，我们设计了一种基于全局推理的 HCNN（GloRe-HCNN）用于 HSI 分类。一方面，GloRe-HCNN 仅使用一层 3-D CNN 和一层 2-D CNN 来联合提取 HSI 的空间光谱特征。另一方面，我们设计了一个空间光谱全局推理单元（SS-GloRe-Unit）来代替堆叠式多层 3-D CNN，以充分提取全局特征。我们在三个标准数据集中选择小型训练样本，并将它们与最先进的 CNN 方法进行比较。大量实验表明我们的 GloRe-HCNN 具有先进的性能。