Multispectral Transforms Using Convolution Neural Networks for Remote Sensing Multispectral Image Compression

被引:35
|
作者
Li, Jin [1 ]
Liu, Zilong [2 ]
机构
[1] Univ Cambridge, Dept Engn, 9 JJ Thomson Ave, Cambridge CB3 0FA, England
[2] Natl Inst Metrol, Opt Div, Beijing 100029, Peoples R China
基金
中国国家自然科学基金;
关键词
multispectral image compression; convolution neural networks; Tucker Decomposition; CLASSIFICATION; ALGORITHM;
D O I
10.3390/rs11070759
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
A multispectral image is a three-order tensor since it is a three-dimensional matrix, i.e., one spectral dimension and two spatial position dimensions. Multispectral image compression can be achieved by means of the advantages of tensor decomposition (TD), such as Nonnegative Tucker Decomposition (NTD). Unfortunately, the TD suffers from high calculation complexity and cannot be used in the on-board low-complexity case (e.g., multispectral cameras) that the hardware resources and power are limited. Here, we propose a low-complexity compression approach for multispectral images based on convolution neural networks (CNNs) with NTD. We construct a new spectral transform using CNNs, where the CNNs are able to transform the three-dimension spectral tensor from large-scale to a small-scale version. The NTD resources only allocate the small-scale three-dimension tensor to improve calculation efficiency. We obtain the optimized small-scale spectral tensor by the minimization of original and reconstructed three-dimension spectral tensor in self-learning CNNs. Then, the NTD is applied to the optimized three-dimension spectral tensor in the DCT domain to obtain the high compression performance. We experimentally confirmed the proposed method on multispectral images. Compared to the case that the new spectral tensor transform with CNNs is not applied to the original three-dimension spectral tensor at the same compression bit-rates, the reconstructed image quality could be improved. Compared with the full NTD-based method, the computation efficiency was obviously improved with only a small sacrifices of PSNR without affecting the quality of images.
引用
收藏
页数:20
相关论文
共 50 条
  • [1] Multispectral Image Analysis using Convolution Neural Networks
    Kulkarni, Arun D.
    [J]. INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS, 2023, 14 (10) : 13 - 19
  • [2] Multispectral Image Analysis using Convolution Neural Networks
    Kulkarni, Arun D.
    [J]. International Journal of Advanced Computer Science and Applications, 2023, 14 (10): : 13 - 19
  • [3] Multispectral Remote Sensing Image Change Detection Based on Twin Neural Networks
    Mo, Wenhao
    Tan, Yuanpeng
    Zhou, Yu
    Zhi, Yanli
    Cai, Yuchang
    Ma, Wanjie
    [J]. ELECTRONICS, 2023, 12 (18)
  • [4] SINGLE-IMAGE SUPER RESOLUTION FOR MULTISPECTRAL REMOTE SENSING DATA USING CONVOLUTIONAL NEURAL NETWORKS
    Liebel, L.
    Koerner, M.
    [J]. XXIII ISPRS CONGRESS, COMMISSION III, 2016, 41 (B3): : 883 - 890
  • [5] Classification of multispectral remote sensing image using an improved backpropagation neural network
    Du, HQ
    Mei, WB
    Shark, LK
    [J]. ELECTRONIC IMAGING AND MULTIMEDIA SYSTEMS II, 1998, 3561 : 403 - 408
  • [6] Multispectral image compression with band ordering and wavelet transforms
    Ahmed Hagag
    Mohamed Amin
    Fathi E. Abd El-Samie
    [J]. Signal, Image and Video Processing, 2015, 9 : 769 - 778
  • [7] Multispectral image compression with band ordering and wavelet transforms
    Hagag, Ahmed
    Amin, Mohamed
    Abd El-Samie, Fathi E.
    [J]. SIGNAL IMAGE AND VIDEO PROCESSING, 2015, 9 (04) : 769 - 778
  • [8] Integer PCA and wavelet transforms for multispectral image compression
    Kaarna, A
    [J]. IGARSS 2001: SCANNING THE PRESENT AND RESOLVING THE FUTURE, VOLS 1-7, PROCEEDINGS, 2001, : 1853 - 1855
  • [9] Progressive Recurrent Neural Network for Multispectral Remote Sensing Image Destriping
    Li, Jia
    Zhang, Junjie
    Han, Jungong
    Yan, Chenggang
    Zeng, Dan
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2023, 61
  • [10] Segmentation on multispectral remote sensing image using watershed transformation
    Zhang, Yun
    Feng, Xuezhi
    Le, Xinghua
    [J]. CISP 2008: FIRST INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING, VOL 4, PROCEEDINGS, 2008, : 773 - 777