Diffused Convolutional Neural Network for Hyperspectral Image Super-Resolution

With the rapid development of deep convolutional neural networks (CNNs), super-resolution (SR) in hyperspectral image (HSI) has achieved good results. Current methods generally use 2-D convolution for feature extraction, but they cannot effectively extract spectral information. Although 3-D convolution can better characterize feature structure of HSI, it will lead to parameter redundancy, model complexity, and severe memory shortage. To address the above problems, we propose a new HSI SR method, named diffused CNN (DCNN). Specifically, spectral convolutions have been added into the enhanced convolutional neural (ECN) block, and a series of spectral convolutions are introduced in the residual network to learn features in the channel direction of different depths. Furthermore, histogram of oriented gradient (HOG) and local binary pattern (LBP) are used to retain the shape and texture information of the image, respectively, which can well represent the spatial structure of the object. To effectively make use of the extracted shallow and deep features, a feature fusion strategy is used to reinforce the reconstruction efficiency. Besides, an image enhancement module has been developed to diffuse the SR image into the image space. Extensive evaluations and comparisons show that our DCNN approach can not only recover the HSI data with richer details but also achieve superiority over several state-of-the-art methods.

I. Introduction

Hyperspectral image (HSI) has the unique advantage of “map-spectrum integration” [1], [2], which can accurately characterize the intrinsic structure and material properties of features and provide the possibility to distinguish and measure the composition of objects with high accuracy. Therefore, HSI is widely used in remote sensing applications such as mineral exploration [3], [4], environmental monitoring [5], climate prediction [6], and image classification [7], [8], [9], [10]. However, the long exposure time of hyperspectral systems is necessary for adequate signal-to-noise ratio (SNR), which results in low spatial resolution of HSI and limits the analysis and applications of HSI. Thus, it is meaningful to reconstruct high-resolution HSI (HR-HSI) with both spectral and spatial information from low-resolution HSI (LR-HSI).