2D经验模态分解与非下采样方向滤波器组的红外与可见光图像融合算法

熊芳芳; 肖宁

光学技术, 2019, 45 (3): 355, 网络出版: 2019-08-07

2D经验模态分解与非下采样方向滤波器组的红外与可见光图像融合算法

Infrared and visible image fusion algorithm based on 2D empirical mode decomposition and non-subsampled directional filter banks

论文大纲

熊芳芳 ^1,*肖宁 ²

作者单位

¹ 南通理工学院, 计算机与信息工程学院, 江苏南通 226002

² 山西财经大学, 信息工程学院, 山西太原 030006

摘要

针对当前红外(IR)与可见光(VI)图像融合中细节保留能力不足及目标配准精度不高的问题, 设计了一种多尺度2D经验模态分解耦合非下采样方向滤波器组(NSDFB)的红外与可见光图像融合算法。分别计算红外与可见光图像的熵值, 并比较二者阈值的大小, 计算阈值较大图像的残差。通过2D经验模态分解(2D-EMD)和NSDFB机制, 构建了多尺度方向分解模型, 将熵值较大图像的残差和熵值较小的图像变换为高频方向系数与低频系数, 以获得源图像的细节和特征信息。对于低频系数, 引入加权平均作为低频系数的融合准则; 根据区域能量对比度与清晰度来定义融合规则, 完成高频系数的融合。利用2D-EMD多尺度分解逆变换将获取的低频与高频系数生成新图像。实验表明: 与当前常用红外与可见光图像融合对比, 所提算法具有更高的融合质量, 所输出的图像具有更好的对比度与丰富的细节信息。

Abstract

Aiming at the problem of insufficient detail preservation and low registration accuracy in the fusion of infrared (IR) and visible (VI) images at present, a multi-scale 2D empirical mode decomposition and (Non-subsampled directional filter banks, NSDFB) was designed for the fusion of IR and VI images. The entropy values of infrared and visible images were calculated to compare the thresholds, and the residual values of images with larger thresholds were calculated. A multi-scale directional decomposition model of 2D-EMD was constructed by means of 2D empirical mode decomposition (2D-EMD) and NSDFB, Which is used to transform the residual error of image with larger entropy value and the smaller entropy value into the high frequency direction coefficient and the low frequency coefficient, which can effectively obtain the details and feature information of the source image. For low frequency coefficients, the weighted average is introduced as the fusion criterion for low frequency coefficients, and the high frequency coefficients were fused by comparing the regional energy contrast with the definition scheme. The 2D-EMD multi-scale decomposition is used to transform the low frequency and high frequency coefficients to generate new images. The experiment shows that the proposed algorithm has a higher fusion quality compared with the current infrared and visible image fusion, and the output image has a better contrast and rich details.

参考文献

[1] 唐启永, 徐凯. 基于小波变换和区域分割的红外可见光彩色图像融合算法［J］. 光学技术,2015,41(1):68-71.

Tang Qiyong, Xu Kai. Infrared visible light color image fusion algorithm based on wavelet transform and region segmentation ［J］. Optical Technique,2015,41(1):68-71.

[2] Lu Xiaoqi, Zhang Baohua, Zhao Ying. The infrared and visible image fusion algorithm based on target separation and sparse representation［J］. Infrared Phys Technol,2014,67(6):397-407.

[3] 邓立暖, 尧新峰. 基于NSST的红外与可见光图像融合算法［J］. 电子学报,2017,45(12):2965-2970.

Deng Linuan, Yao Xinfeng. Research on the fusion algorithm of infrared and visible images based on non-subsampled shearlet transform ［J］. Acta Electronica Sinica,2017,45(12):2965-2970.

[4] 张生伟, 李伟, 赵雪景. 一种基于稀疏表示的可见光与红外图像融合方法［J］.电光与控制,2017,24(06):47-52.

Zhang Shengwei, Li Wei, Zhao Xuejing. A method for fusion of visible and infrared images based on sparse representation ［J］. Electronics Optic & Control,2017,24(06):47-52.

[5] 吴一全, 王志来. 基于目标提取与引导滤波增强的红外与可见光图像融合［J］. 光学学报,2017,37(08):98-108.

Wu Yiquan, Wang Zhilai. Infrared and visible image fusion based on target extraction and guided filtering enhancement ［J］. Acta Optical Sinica,2017,37(08):98-108.

[6] Zhao Jufeng, Cui Guangmang, Gong Xiaoli. Fusion of visible and infrared images using global entropy and gradient constrained regularization ［J］. Infrared Physics and Technology,2017,81(15):201-209.

[7] Singh Dilbag, Garg Deepak, Singh Pannu. Efficient landsat image fusion using fuzzy and stationary discrete wavelet transform ［J］. The Imaging Science Journal,2017,65(2):108-114.

[8] Du Jiao, Li Weisheng, Xiao Bin. Union laplacian pyramid with multiple features for image fusion ［J］. Neurocomputing,2016,194:326-339.

[9] 任晓霞, 孙秀明, 耿鹏, 等.多小波和NSDFB组合域递归滤波多聚焦图像融合［J］. 智能系统学报,2016,11(02):241-248.

Ren Xiaoxia, Sun Xiuming, Geng Peng, et al. Multifocus image fusion using a recursive filter in the combined domain of multiwavelets and NSDFB ［J］. CAAI Transactions on Intelligent Systems,2016,11(2):241-248.

[10] Tremsin A S, Lerche M, Schillinger B, et al. Bright flash neutron radiography capability of the research reactor at the McClellan Nuclear Research Center ［J］. Nuclear Inst. and Methods in Physics Research,2014,748(7):46-53.

[11] Muhammad Kaleem, Aziz Guergachi, Sridhar Krishnan. Hierarchical decomposition based on a variation of empirical mode decomposition ［J］. Signal, Image and Video Processing,2017,11(5):793-800.

[12] 杨伟, 艾廷华. 运用约束Delaunay三角网从众源轨迹线提取道路边界［J］. 测绘学报,2017,46(02):237-245.

Yang Wei, Ai Tinghua. The extraction of road boundary irom crowdsourcing trajectory using constrained delaunay triangulation ［J］. Acta Ueodaetica et Cartographica Sinica,2017,46(2):237-245.

[13] Vanmathi Cimhjtrs, Prabu Sevugan. Image steganography using fuzzy logic and chaotic for large payload and high imperceptibility ［J］. International Journal of Fuzzy Systems,2018,20(2):460- 473.

[14] 杨超, 蔡晓东, 甘凯今.基于自适应显著特征选择的动态加权平均行人识别模型［J］.计算机工程与科学,2017,39(05):936-943.

Yang Chao, Cai Xiaodong, Gan Kaijin. A dynamic weighted average pedestrian identification model based on adaptive feature selection ［J］. Computer Engineering Science,2017,39(05):936-943.

[15] Samia Neftimeziani, Mourad Oussalah, Majeed Soufian. On the use of inclusion structure in fuzzy clustering algorithm in case of Gaussian membership functions ［J］. Journal of Intelligent and Fuzzy Systems,2015,28(4):1477-1493.

[16] Tobias Pltz, Stefan Roth. Automatic registration of images to untextured geometry using average shading gradients ［J］. International Journal of Computer Vision,2017,125(3):65-81.

[17] Li Shuang, Yang Zewei, Li Hongsheng. Statistical evaluation of no-reference image quality assessment metrics for remote sensing images［J］. ISPRS International Journal of Geo- Information,2017,6(5):1-18.

[18] 樊晓婷, 马巧梅, 陈够喜.信息熵和差分激励融合的图像拼接检测［J］.计算机工程与设计,2016,37(01):37-41.

Fan Xiaoting, Ma Qiaomei, Chen Gouxi. Image splicing detection of information entropy and difference incentive fusion ［J］. Computer Engineering and Design,2016,37(01):37-41.

[19] 丰明坤, 赵生妹, 孙丽慧, 等.基于局部高斯加权融合的图像质量评价［J］. 计算机工程,2016,42(08):237-242.

Fend Mingkun, Zhao Shengmei, Sun Lihui, et al. Image quality assessment based on local caussian weighted fusion ［J］. Computer Engineering,2016,42(08):237-242.

熊芳芳, 肖宁. 2D经验模态分解与非下采样方向滤波器组的红外与可见光图像融合算法[J]. 光学技术, 2019, 45(3): 355. XIONG Fangfang, XIAO Ning. Infrared and visible image fusion algorithm based on 2D empirical mode decomposition and non-subsampled directional filter banks[J]. Optical Technique, 2019, 45(3): 355.

2D经验模态分解与非下采样方向滤波器组的红外与可见光图像融合算法

关于本站 Cookie 的使用提示

全站搜索

2D经验模态分解与非下采样方向滤波器组的红外与可见光图像融合算法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索