A hyperspectral image reconstruction algorithm based on RGB image using multi-scale atrous residual convolution network

被引:7
|
作者
Hu, Shaoxiang [1 ]
Hou, Rong [2 ]
Ming, Luo [1 ]
Su, Meifang [1 ]
Chen, Peng [2 ]
机构
[1] Univ Elect Sci & Technol China, Sch Elect Sci & Engn, Chengdu, Peoples R China
[2] Sichuan Key Lab Conservat Biol Endangered Wildlife, Chengdu Res Base Giant Panda Breeding, Chengdu, Peoples R China
关键词
hyperspectral image reconstruction; deep learning; atrous convolution; marine science; multi-scale;
D O I
10.3389/fmars.2022.1006452
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Hyperspectral images are a valuable tool for remotely sensing important characteristics of a variety of landscapes, including water quality and the status of marine disasters. However, hyperspectral data are rare or expensive to obtain, which has spurred interest in low-cost, fast methods for reconstructing hyperspectral data from much more common RGB images. We designed a novel algorithm to achieve this goal using multi-scale atrous convolution residual network (MACRN). The algorithm includes three parts: low-level feature extraction, high-level feature extraction, and feature transformation. The high-level feature extraction module is composed of cascading multi-scale atrous convolution residual blocks (ACRB). It stacks multiple modules to form a depth network for extracting high-level features from the RGB image used as an input. The algorithm uses jump connection for residual learning, and the final high-level feature combines the output of the low-level feature extraction module and the output of the cascaded atrous convolution residual block element by element, so as to prevent gradient dispersion and gradient explosion in the deep network. Without adding too many parameters, the model can extract multi-scale features under different receptive fields, make better use of the spatial information in RGB images, and enrich the contextual information. As a proof of concept, we ran an experiment using the algorithm to reconstruct hyperspectral Sentinel-2 satellite data from the northern coast of Australia. The algorithm achieves hyperspectral spectral reconstruction in 443nm-2190nm band with less computational cost, and the results are stable. On the Realworld dataset, the reconstruction error MARE index is less than 0.0645, and the reconstruction time is less than 9.24S. Therefore, in the near infrared band, MACRN reconstruction accuracy is significantly better than other spectral reconstruction algorithms. MACRN hyperspectral reconstruction algorithm has the characteristics of low reconstruction cost and high reconstruction accuracy, and its advantages in ocean spectral reconstruction are more obvious.
引用
收藏
页数:9
相关论文
共 50 条
  • [1] A hyperspectral image classification algorithm based on atrous convolution
    Zhang, Xiaoqing
    Zheng, Yongguo
    Liu, Weike
    Wang, Zhiyong
    EURASIP JOURNAL ON WIRELESS COMMUNICATIONS AND NETWORKING, 2019, 2019 (01)
  • [2] A hyperspectral image classification algorithm based on atrous convolution
    Xiaoqing Zhang
    Yongguo Zheng
    Weike Liu
    Zhiyong Wang
    EURASIP Journal on Wireless Communications and Networking, 2019
  • [3] AMSUnet: A neural network using atrous multi-scale convolution for medical image segmentation
    Yin, Yunchou
    Han, Zhimeng
    Jian, Muwei
    Wang, Gai-Ge
    Chen, Liyan
    Wang, Rui
    COMPUTERS IN BIOLOGY AND MEDICINE, 2023, 162
  • [4] Hybrid Dilated Convolution with Multi-Scale Residual Fusion Network for Hyperspectral Image Classification
    Li, Chenming
    Qiu, Zelin
    Cao, Xueying
    Chen, Zhonghao
    Gao, Hongmin
    Hua, Zaijun
    MICROMACHINES, 2021, 12 (05)
  • [5] Hyperspectral image classification with multi-scale graph convolution network
    Zhao, Wenzhi
    Wu, Dinghui
    Liu, Yuanlin
    INTERNATIONAL JOURNAL OF REMOTE SENSING, 2021, 42 (21) : 8380 - 8397
  • [6] A Multi-scale Dilated Residual Convolution Network for Image Denoising
    Jia, Xinlei
    Peng, Yali
    Ge, Bao
    Li, Jun
    Liu, Shigang
    Wang, Wenan
    NEURAL PROCESSING LETTERS, 2023, 55 (02) : 1231 - 1246
  • [7] A Multi-scale Dilated Residual Convolution Network for Image Denoising
    Xinlei Jia
    Yali Peng
    Bao Ge
    Jun Li
    Shigang Liu
    Wenan Wang
    Neural Processing Letters, 2023, 55 : 1231 - 1246
  • [8] Hyperspectral Image Classification Based on Multi-Scale Residual Network with Attention Mechanism
    Qing, Yuhao
    Liu, Wenyi
    REMOTE SENSING, 2021, 13 (03) : 1 - 18
  • [9] Hyperspectral Image Classification Based on Multi-Scale Feature Fusion Residual Network
    Deng Ziqing
    Wang Yang
    Zhang Bing
    Ding Zhao
    Bian Lifeng
    Yang Chen
    LASER & OPTOELECTRONICS PROGRESS, 2022, 59 (18)
  • [10] Image Super-Resolution Reconstruction Algorithm Based on Enhanced Multi-Scale Residual Network
    Xu Jiao
    Yuan Sannan
    LASER & OPTOELECTRONICS PROGRESS, 2023, 60 (04)