1 Content Introduction
Remote sensing image data is large and needs to be compressed using low-complexity algorithms for spaceborne devices. The binary tree coding with adaptive scanning order (BTCA) is an effective algorithm for this purpose. However, for large-scale remote sensing images, BTCA requires a large amount of memory and does not provide random access properties. In this paper, we propose a method based on BTCA and optimized truncation. The wavelet image is first divided into several independently encoded blocks provided by BTCA. Based on the nature of BTCA, we carefully select effective truncation points for each block to optimize the rate-distortion ratio, achieving a higher compression ratio, lower memory requirements, and obtaining random access properties. Without any entropy coding, the proposed method is simple and fast, making it very suitable for spaceborne devices. Experiments were conducted on three remote sensing image datasets, and the results show that it can significantly improve PSNR, SSIM, and VIF, as well as the subjective visual experience.
2 Simulation Code
%% matlab code for BTOT(Binary Tree and Optimized Truncation)% unoptimized, without head information, without entropy coding.% % Reference:% % Ke-Kun Huang, Hui Liu, Chuan-Xian Ren, Yu-Feng Yu and Zhao-Rong Lai. % Remote sensing image compression based on binary tree and optimized truncation. % Digital Signal Processing, vol. 64, pp. 96-106, 2017.% http://dx.doi.org/10.1016/j.dsp.2017.02.008% % Email: [email protected]% Homepage: http://www.scholat.com/huangkekun
clc;clear;close all%% ----------- Input ----------------imname = 'SanDiego.bmp';I_Orig = double(imread(imname));
[row, col] = size(I_Orig);blksize = 64;
%% ----------- Wavelet Decomposition -------------n_log = log2(row); level = floor(n_log);I_Dec = wavecdf97(I_Orig, level); n_min = 1;brates = [0.0625, 0.125, 0.25, 0.5, 1];
%% ----------- Coding ----------------[out_code, blklen, n_max, n_min, out_S,out_R,out_N] = encode(I_Dec, blksize, n_min);
%% ----------- Decoding ----------------disp([ 'aa_BTOT_' imname(1:end-4) '=[']);for rate=brates I_DecR = decode(out_code, blklen, n_max, n_min, blksize, row, rate, out_S,out_R,out_N); I_Rec = wavecdf97(I_DecR, -level); MSE = sum(sum((I_Rec - I_Orig).^2))/(row*row); PSNR = 10*log10(255*255/MSE); disp([sprintf('%.4f',rate) ' ' sprintf('%.2f',PSNR)]); enddisp('];');figuresubplot(121)imshow(I_Orig,[]);title('Original Image')subplot(122)imshow(I_Rec,[]);title('Compressed Image')3 Results
4 References
[1] A. Gunes, M.B. Guldogan, Joint underwater target detection and tracking with
the Bernoulli filter using an acoustic vector sensor, Digit. Signal Process. 48 (1)
(2016) 246–258.
[2] S. U
ˇ
gur,
O. Arıkan, A.C. Gürbüz, SAR image reconstruction by expectation max
imization based matching pursuit, Digit. Signal Process. 37 (2) (2015) 75–84.
[3] M.M. Nielsen, Remote sensing for urban planning and management: the use of
window-independent context segmentation to extract urban features in Stock
holm, Comput. Environ. Urban Syst. 52 (7) (2015) 1–9.
[4] J.M. Shapiro, Embedded image coding using zerotrees of wavelet coefficients,
IEEE Trans. Signal Process. 41 (12) (1993) 3445–3462.
[5] A. Said, W.A. Pearlman, A new, fast, and efficient image codec based on set
partitioning in hierarchical trees, IEEE Trans. Circuits Syst. Video Technol. 6 (3)
(1996) 243–250.
[6] W.A. Pearlman, A. Islam, N. Nagaraj, A. Said, Efficient, low-complexity image
coding with a set-partitioning embedded block coder, IEEE Trans. Circuits Syst.
Video Technol. 14 (11) (2004) 1219–1235.
[7] A. Munteanu, J. Cornelis, G.V.D. Auwera, P. Cristea, Wavelet image compression
– the quadtree coding approach, IEEE Trans. Inf. Technol. Biomed. 3 (3) (1999)
176–185.
[8] D. Taubman, High performance scalable image compression with EBCOT, IEEE
Trans. Image Process. 9 (7) (2000) 1158–1170