Remote-sensing image enhancement based on tensor decomposition and nonsubsampled Contourlet transform
-
摘要:
图像质量低、特征信息不明显是遥感图像获取过程中的常见问题。传统的图像增强方法常常因为不能有效地整合全局信息，从而不能高精度、高效率地凸显有用信息。本文通过结合张量分解和非下采样Contourlet变换，提出一种改进的遥感图像增强方法。使用优化的非下采样Contourlet变换对原始图像进行分解，将各尺度和方向的高频细节图像组合成高阶张量。通过贝叶斯概率张量补全，从不完全张量中识别潜在因子，以预测图像缺失的细节信息。实验结果表明：所提出方法能在有效恢复样张缺失信息的同时突出图像的特征信息，与不同图像增强方法相比，样张处理后在信噪比、结构相似度以及均方根误差方面最大提升分别为27.9%、37.6%和45.4%。改进的遥感图像增强方法在可视化比较和定量评价方面优于常用的图像增强方法。
-
关键词:
- 图像增强 /
- Contourlet变换 /
- 张量分解 /
- 贝叶斯概率张量补全
Abstract:In the process of remote sensing image acquisition, low quality and lack of important information of image are common problems as the existence of interference information. Traditional image enhancement methods often cannot highlight useful information with high precision and high efficiency because they cannot integrate global information effectively. In order to solve these problems, a remote-sensing image enhancement method based on tensor decomposition and nonsubsampled Contourlet transform is proposed. The optimized nonsubsampled Contourlet transform is used to decompose the original image, and the high-order tensor is composed of high-frequency detail images in all directions on all scales. Through Bayesian probability tensor completion, the potential factors recognized from the incomplete tensor are used to predict the missing details of the image. Experimental results indicate that the proposed method can recover the missing information more effectively and highlight the feature information of the image. Compared with different image enhancement methods, the maximum improvement of signal-to-noise ratio, structure similarity and root mean square error are 27.9%, 37.6% and 45.4%, respectively. The proposed method is superior to the common image enhancement methods in quantitative evaluation and visual comparison.
-
图 5 模拟数据集实验结果。 (a) TDIE；(b) CTIE；(c) TDWT；(d) WTIE；(e) CTTE
Figure 5. Experimental results of simulated data set. (a) TDIE; (b) CTIE; (c) TDWT; (d) WTIE; (e) CTTE
图 6 处理后图像像素统计图。(a) TDIE；(b) CTIE；(c) TDWT；(d) WTIE；(e) CTTE
Figure 6. The processed image pixel statistics. (a) TDIE; (b) CTIE; (c) TDWT; (d) WTIE; (e) CTTE
图 8 实验结果。(a) TDIE；(b) CTIE；(c) TDWT；(d) WTIE；(e) CTTE
Figure 8. Simulation experiment results. (a) TDIE; (b) CTIE; (c) TDWT; (d) WTIE; (e) CTTE
表 1 仿真实验数据指标
Table 1. Simulation experiment data index
TDIE CTIE TDWT WTIE CTTE ALC 2.9548 4.2105 4.4852 5.0158 3.0258 PSNR 30.2847 28.2254 27.5589 25.5414 35.4478 SSIM 0.7214 0.6885 0.4986 0.5713 0.7989 RMSE 8.5980 10.5884 11.2585 9.7243 6.1452 
下载: 导出CSV
-
[1] 孙明超, 马天翔, 宋悦铭, 等. 基于相位特征的可见光和SAR遥感图像自动配准[J]. 光学 精密工程,2021,29(3):616-627. doi: 10.37188/OPE.20212903.0616SUN M CH, MA T X, SONG Y M, et al. Automatic registration of optical and SAR remote sensing image based on phase feature[J]. Optics and Precision Engineering, 2021, 29(3): 616-627. (in Chinese). doi: 10.37188/OPE.20212903.0616 [2] 保文星, 桑斯尔, 沈象飞. 基于信息熵约束和KAZE特征提取的遥感图像配准算法研究[J]. 光学 精密工程,2020,28(8):1810-1819.BAO W X, SANG S E, SHEN X F. Remote sensing image registration algorithm based on entropy constrained and KAZE feature extraction[J]. Optics and Precision Engineering, 2020, 28(8): 1810-1819. (in Chinese). [3] CHEN Y, HE W, YOKOYA N, et al. Hyperspectral image restoration using weighted group sparsity-regularized low-rank tensor decomposition[J]. IEEE Transactions on Cybernetics, 2020, 50(8): 3556-3570. doi: 10.1109/TCYB.2019.2936042 [4] 贾建鑫, 孙海彬, 蒋长辉, 等. 多源遥感数据的道路提取技术研究现状及展望[J]. 光学 精密工程,2021,29(2):430-442. doi: 10.37188/OPE.20212902.0430JIA J X, SUN H B, JIANG CH H, et al. Road extraction technology based on multi-source remote sensing data: review and prospects[J]. Optics and Precision Engineering, 2021, 29(2): 430-442. (in Chinese). doi: 10.37188/OPE.20212902.0430 [5] 程亚亚, 于化东, 于占江, 等. 微铣刀同轴全息图像增强方法[J]. 中国光学,2020,13(4):705-712. doi: 10.37188/CO.2019-0217CHENG Y Y, YU H D, YU ZH J, et al. Method of enhancing the quality of in-line holographic images for micro-milling tool[J]. Chinese Optics, 2020, 13(4): 705-712. (in Chinese). doi: 10.37188/CO.2019-0217 [6] 常志文, 王立忠, 梁晋, 等. 基于图像块分解融合的水下标定图像增强[J]. 中国光学(中英文),2024,17(4):810-822. doi: 10.37188/CO.2023-0218CHANG ZH W, WANG L ZH, LIANG J, et al. Underwater calibration image enhancement based on image block decomposition and fusion[J]. Chinese Optics, 2024, 17(4): 810-822. (in Chinese). doi: 10.37188/CO.2023-0218 [7] 都元松, 董文锋, 黎波涛, 等. 基于Contourlet变换的引导滤波图像边缘增强方法[J]. 空军预警学院学报,2017,31(5):370-374.DU Y S, DONG W F, LI B T, et al. Guide filtering algorithm for image edge enhancement based on Contourlet transformation[J]. Journal of Air Force Early Warning Academy, 2017, 31(5): 370-374. (in Chinese). [8] HUANG Y H, CHEN D W. Image fuzzy enhancement algorithm based on contourlet transform domain[J]. Multimedia Tools and Applications, 2020, 79(47): 35017-35032. [9] ASMARE M H, ASIRVADAM V S, HANI A F M. Image enhancement based on contourlet transform[J]. Signal, Image and Video Processing, 2015, 9(7): 1679-1690. doi: 10.1007/s11760-014-0626-7 [10] 李彦. 小波与Contourlet变换图像增强算法优越性的探讨[J]. 安徽职业技术学院学报,2014,13(4):17-20. doi: 10.3969/j.issn.1672-9536.2014.04.005LI Y. On the superiority of the wavelet transform and Contourlet transform image enhancement algorithm[J]. Journal of Anhui Vocational and Technical College, 2014, 13(4): 17-20. (in Chinese). doi: 10.3969/j.issn.1672-9536.2014.04.005 [11] HE W, ZHANG H Y, ZHANG L P, et al. Hyperspectral image denoising via noise-adjusted iterative low-rank matrix approximation[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015, 8(6): 3050-3061. doi: 10.1109/JSTARS.2015.2398433 [12] 马友, 王强, 蔡长海. 基于张量分解的卫星遥感图像增强算法[J]. 科技与创新,2018(22):43-44,47.MA Y, WANG Q, CAI CH H. Satellite remote sensing image enhancement algorithm based on tensor decomposition[J]. Science and Technology & Innovation, 2018(22): 43-44,47. (in Chinese) [13] HAO R R, SU ZH X. A patch-based low-rank tensor approximation model for multiframe image denoising[J]. Journal of Computational and Applied Mathematics, 2018, 329: 125-133. doi: 10.1016/j.cam.2017.01.022 [14] BENGUA J A, PHIEN H N, TUAN H D, et al. Efficient tensor completion for color image and video recovery: low-rank tensor train[J]. IEEE Transactions on Image Processing, 2017, 26(5): 2466-2479. doi: 10.1109/TIP.2017.2672439 [15] VANI R, RAJAN K S. Effective satellite image enhancement based on the discrete wavelet transform[J]. International Journal of Business Information Systems, 2020, 33(4): 446-471. doi: 10.1504/IJBIS.2020.105834 [16] 栾孟杰. 一种多分辨多尺度的红外图像增强算法[J]. 激光杂志,2019,40(8):81-84.LUAN M J. A multiresolution and multiscale infrared image enhancement algorithm[J]. Laser Journal, 2019, 40(8): 81-84. (in Chinese). [17] AAMIR M, RAHMAN Z, PU Y F, et al. Satellite image enhancement using wavelet-domain based on singular value decomposition[J]. International Journal of Advanced Computer Science and Applications, 2019, 10(6): 514-519. [18] 王浩, 张叶, 沈宏海, 等. 图像增强算法综述[J]. 中国光学,2017,10(4):438-448. doi: 10.3788/co.20171004.0438WANG H, ZHANG Y, SHEN H H, et al. Review of image enhancement algorithms[J]. Chinese Optics, 2017, 10(4): 438-448. (in Chinese). doi: 10.3788/co.20171004.0438 [19] 李亮亮. 基于非下采样剪切波变换的图像增强算法研究[D]. 长春: 吉林大学, 2019.LI L L. The research of image enhancement algorithm based on nonsubsampled shearlet transform[D]. Changchun: Jilin University, 2019. (in Chinese). [20] 赵万金, 周春雷. 基于Contourlet变换的图像增强技术识别裂缝[J]. 岩性油气藏,2017,29(3):103-109. doi: 10.3969/j.issn.1673-8926.2017.03.12ZHAO W J, ZHOU CH L. Application of image enhancement technique to fracture identification based on Contourlet transform[J]. Lithologic Reservoirs, 2017, 29(3): 103-109. (in Chinese). doi: 10.3969/j.issn.1673-8926.2017.03.12 [21] 李进, 金龙旭, 李国宁. 离散小波变换域非负张量分解的高光谱遥感图像压缩[J]. 电子与信息学报,2013,35(2):489-493.LI J, JIN L X, LI G N. Hyper-spectral remote sensing image compression based on nonnegative tensor factorizations in discrete wavelet domain[J]. Journal of Electronics & Information Technology, 2013, 35(2): 489-493. (in Chinese). -
下载:
计量
- 文章访问数: 81
- HTML全文浏览量: 22
- PDF下载量: 35
- 被引次数: 0
