Hybrid Dilated Convolution Guided Feature Filtering and Enhancement Strategy for Hyperspectral Image Classification
Hybrid Dilated Convolution Guided Feature Filtering and Enhancement Strategy for Hyperspectral Image Classification
复制标题
DOI：
10.1109/lgrs.2021.3100407
复制
发表时间：
2021-08-04
影响因子：
4.8
通讯作者：
Jiang, Yizhang
中科院分区：
文献类型：
--
作者：
Liu, Runmin;Cai, Weiwei;Jiang, Yizhang
随着光学和光子学的日益成熟，高光谱技术也有了很大的发展。由数百个相邻波段组成并包含有用信息的高光谱图像可以很容易地获得。然而，与普通遥感图像不同，高光谱遥感图像中的每个样本都具有高维特征，包含丰富的空间和光谱信息，这大大增加了特征选择和挖掘的难度，增加了计算复杂度，限制了模型的识别精度。因此，本文提出了一种新的扩展卷积制导混合特征滤波和增强策略（HDCFE-Net）模型来对高光谱图像进行分类。扩张型卷积可以在不减少感受野的情况下减少空间特征的损失，并且可以获得远端特征。它也可以与传统的卷积结合而不丢失其原始信息。我们提出了一种特征过滤和增强策略，以消除冗余特征并降低计算复杂度。其核心概念是设置一个阈值特征值，就像四舍五入方法一样，对特征进行过滤和增强。在Indian Pines （IPs）、Pavia University （PU）和Salinas三个知名的高光谱数据集上进行的实验表明，在不到1% （IPs: 5%）的训练样本中，我们的方法的总体准确率（OA）分别为77%、89%和91%，优于几种知名的方法。实验证明了hdfe - net的有效性和优越性。
With the increasing maturity of optics and photonics, hyperspectral technology has also greatly advanced. Hyperspectral images composed of hundreds of adjacent bands and containing useful information can be easily obtained. However, unlike ordinary remote sensing images, each sample in hyperspectral remote sensing images has high-dimensional features and contains rich spatial and spectral information, which greatly increases the difficulty of feature selection and mining, increases the computational complexity, and limits the recognition accuracy of the model. Therefore, in this letter, a novel hybrid dilated-convolution-guided feature filtering and enhancement strategy (HDCFE-Net) model is proposed to classify hyperspectral images. Dilated convolution can reduce the spatial feature loss without reducing the receptive field and can obtain distant features. It can also be combined with the traditional convolution without losing its original information. We propose a feature filtering and enhancement strategy that eliminates redundant features and reduces computational complexity. The core concept is to set a threshold feature value, like the rounding method, to filter and enhance features. Experiments on three well-known hyperspectral datasets--Indian Pines (IPs), Pavia University (PU), and Salinas--show that in less than 1% (IPs: 5%) of the training samples, the overall accuracy (OA) of our method is 77%, 89%, and 91%, respectively, which is superior to several well-known methods. The experiments demonstrated the effectiveness and superiority of HDCFE-Net.