基于非下采樣Contourlet變換耦合近似度規(guī)則的多聚焦圖像融合算法

柏濤濤，王茜娟，譚云蘭

?

基于非下采樣Contourlet變換耦合近似度規(guī)則的多聚焦圖像融合算法

柏濤濤1，王茜娟2，*譚云蘭2

（1. 安徽廣播電視大學滁州分校，安徽，滁州 239000；2. 井岡山大學電子與信息工程學院，江西，吉安 343009）

當前多聚焦圖像融合算法主要通過單一的比值取大法來完成高頻系數(shù)的融合，忽略了不同高頻系數(shù)間的近似度，導致融合圖像存在模糊效應(yīng)與塊效應(yīng)等不足，采用非下采樣Contourlet變換耦合近似度規(guī)則對多聚焦圖像進行融合，來改善以上不足。利用非下采樣Contourlet變換對圖像進行多尺度、多方向的分解，獲取圖像的高、低頻分解系數(shù)。利用圖像的區(qū)域能量對低頻系數(shù)的信息量進行度量，構(gòu)造低頻系數(shù)融合函數(shù)，用于低頻系數(shù)融合。利用圖像的平均梯度差值對不同高頻系數(shù)的差異度進行度量，建立近似度規(guī)則，根據(jù)不同高頻系數(shù)的近似度采用不同的融合方法獲取融合高頻系數(shù)。將融合后系數(shù)通過非下采樣Contourlet逆變換獲取最后融合圖像。仿真表明，所提算法與當前多聚焦圖像融合方法相比，融合的圖像具有較好的質(zhì)量。

圖像融合；非下采樣contourlet變換；區(qū)域能量；平均梯度；近似度規(guī)則

0 引言

數(shù)字圖像因具有易于獲取、便于傳輸?shù)奶攸c，得到了廣泛的應(yīng)用[1-2]。雖然數(shù)字圖像獲取變得簡單，但是受制于成像設(shè)備的技術(shù)及環(huán)境的限制，不能一次性獲取具有多個聚焦的圖像[3-4]。因此，為了獲取具有良好視覺效果的多聚焦圖像，就需要對不同聚焦的圖像進行多聚焦圖像融合。

對此，本文將非下采樣Contourlet變換引入多聚焦圖像融合，利用其對圖像進行分解，獲取圖像系數(shù)信息。通過低頻系數(shù)的區(qū)域能量特征，構(gòu)造低頻系數(shù)融合函數(shù)，獲取融合低頻系數(shù)。通過高頻系數(shù)的平均梯度特征，構(gòu)建近似度規(guī)則，獲取融合高頻系數(shù)。結(jié)果顯示，所提算法融合的圖像具有較好的空間特性，視覺效果較好。所提算法的主要貢獻在于：1）通過低頻系數(shù)的區(qū)域能量特征，構(gòu)造低頻系數(shù)融合函數(shù)，獲取融合低頻系數(shù)，避免了比值取大法帶來的融合弊端；2）利用高頻系數(shù)的平均梯度，建立相似度規(guī)則，從平均梯度差值出發(fā)，設(shè)計不同的高頻系數(shù)融合規(guī)則，獲取融合高頻系數(shù)，使得融合高頻系數(shù)能夠包含源圖的更多細節(jié)信息。

1 本文圖像融合算法設(shè)計

本文所提多聚焦圖像融合算法的流程圖如圖1所示。從圖1可見，所提算法首先通過非下采樣Contourlet變換對源圖像進行精細的分解，使得獲取的高、低頻系數(shù)中能夠包含更多的源圖信息，以助于改善融合圖像的清晰度。通過低頻系數(shù)的區(qū)域能量特征，建立低頻系數(shù)融合函數(shù)，使得融合的低頻系數(shù)能夠包含更多的源圖能量，以提高融合圖像的質(zhì)量。最后通過高頻系數(shù)的平均梯度特征，建立近似度規(guī)則，對高頻系數(shù)進行融合，使得融合高頻系數(shù)具有豐富的源圖邊緣等信息，進一步提高了融合圖像的質(zhì)量。

圖1 本文多聚焦圖像融合算法的流程

1.1 非下采樣Contourlet變換

非下采樣Contourlet變換是依靠非下采樣塔式分解(Nonsubsampled Pyramid, NSP)與非下采樣濾波器組(Nonsubsampled Pyramid, Directional Filter Bank, NSDFB)對圖像進行分解[12]。非下采樣contourlet變換對源圖分解過程示意圖如圖2所示[13]，依圖可見，源圖經(jīng)過NSP變換進行多尺度分解獲取子圖后，再經(jīng)過NSDFB對高頻子圖進行多方向分解，獲取分解系數(shù)。

圖2 非下采樣Contourlet變換的示意圖

1.2 低頻系數(shù)融合

低頻系數(shù)包含了源圖的主要能量，反應(yīng)了源圖的主要圖像信息[14]。圖像的區(qū)域能量能夠較好地表述不同低頻系數(shù)所含圖像信息的大小，區(qū)域能量越大表示低頻系數(shù)所含圖像信息越多，其越有優(yōu)勢。在此，將借助低頻系數(shù)的區(qū)域能量來完成低頻系數(shù)的融合。

1.3 高頻系數(shù)融合

通過近似度規(guī)則，可利用（5）式和（6）完成高頻系數(shù)的融合，從而輸出融合圖像。

2 實驗結(jié)果

圖3 不同方法的融合結(jié)果

圖4 不同方法的融合結(jié)果

圖5 不同方法對亮度差異圖像的融合結(jié)果

（8）式中，p表示灰度級為的像素點所占的幾率[21]。

表1 圖3中不同方法融合圖像的量化測試結(jié)果

表2 圖4中不同方法融合圖像的量化測試結(jié)果

3 結(jié)論

本文將非下采樣Contourlet變換引入多聚焦圖像的融合，設(shè)計了非下采樣Contourlet變換耦合近似度規(guī)則的多聚焦圖像融合算法。利用具有平移不變性的非下采樣Contourlet變換，對源圖進行多尺度、多方向的精細分解，獲取包含豐富源圖信息的高、低頻系數(shù)，提高了融合圖像所包含源圖信息的豐富度。通過區(qū)域能量特征獲取低頻融合系數(shù)，使得融合低頻系數(shù)能夠包含更多的源圖能量。利用高頻系數(shù)的平均梯度，建立相似度規(guī)則，從平均梯度差值出發(fā)，設(shè)計不同的高頻系數(shù)融合規(guī)則，獲取融合高頻系數(shù)，使得融合高頻系數(shù)能夠包含源圖的更多細節(jié)信息，提高了融合圖像的視覺效果。

[1] Zhao J F, Cui G M, Gong X L. Fusion of Visible and Infrared Images Using Global Entropy and Gradient Constrained Regularization[J]. Infrared Physics and Technology, 2017, 81(3) :201-209.

[2] Chen Z Y, Muramatsu S. Multi-Focus Image Fusion Based on Multiple Directional LOTs[J].ieice transactions on Fundamentals of Electronics, Communications and Computer Sciences, 2015, E98-A(11): 2360-2365.

[3] 古麗米熱·米吉提, 吐爾洪江·阿布都克力木. 基于二進小波變換的遙感圖像融合方法[J].新疆師范大學學報:自然科學版,2017,36(4):83-85.

[4] 孔祥陽. 分塊噪聲自適應(yīng)高光譜圖像去噪算法研究[J]. 井岡山大學學報:自然科學版, 2016, 8(1): 69-74.

[5] Tang S Z, Shen C M, Zhang G X. Adaptive Regularized Scheme for Remote Sensing Image Fusion[J]. Frontiers of Earth Science, 2016, 10(2):236-244.

[6] Zhang Y X, Chen L, Zhao Z H. Multi-Focus Image Fusion Based on Cartoon-texture Image Decomposition[J]. Optik-International Journal for Light and Electron Optics, 2016, 127(3): 1291-1296.

[7] Qin X Q, Zheng J Y, Gang H. Multi-Focus Image Fusion Based on Window Empirical Mode Decomposition[J]. Infrared Physics and Technology,2017,85(1): 251-260.

[8] 王榮榮,楊亞東. 基于復Shearlet變換耦合改進引導濾波的圖像融合算法[J].電子測量與儀器學報,2018,32(5): 126-133.

[9] Vikrant B, Aisha M, Anuja S. Multispectral Medical Image Fusion in Contourlet Domain for Computer Based Diagnosis of Alzheimer's disease[J]. Rev Sci Instrum, 2016, 87(7):1-8.

[10] Adu J H, Xie S H, Gan J H. Image Fusion Based on Visual Salient Features and the Cross-Contrast[J]. Journal of Visual Communication and Image Representation,2016, 40(A):218-224.

[11] 程鋼,李瑋琳,李麗. 基于非下采樣Contourlet變換的多模態(tài)醫(yī)學圖像融合[J].長春工業(yè)大學學報,2018,39(3): 248-252.

[12] Yan T, Liu F X, Chen B. Microscopy Image Fusion Algorithm Based on Saliency Analysis and Adaptive M-Pulse-Coupled Neural Network in Non-Subsampled Contourlet Transform Domain[J]. International Journal of Distributed Sensor Networks, 2017,13(6):1-8.

[13] Shailesh P, Rajesh A T. Contourlet Transform with Sparse Representation-Based Integrated Approach for Image Pansharpening[J]. IETE Journal of Research, 2017, 63(6): 823-833.

[14] Liu S Q, Shi M Z, Zhu Z H. Image Fusion Based on Complex-Shearlet Domain with Guided Filtering[J]. Multidimensional Systems and Signal Processing, 2017, 28(1):207-224.

[15] 趙丹,戴文戰(zhàn),李俊峰. 基于NSST和改進PCNN的醫(yī)學圖像融合[J].光電子·激光,2018, 29(1): 95-104.

[16] Sonam M K. Discrete Wavelet Transform and Cross Bilateral Filter based Image Fusion[J]. International Journal of Intelligent Systems and Applications, 2017,9(1):37-45.

[17] Wang M, He L X, Cheng Y F. Panchromatic and Multi-spectral Fusion Method Combined with Adaptive Gaussian Filter and SFIM Model[J]. Acta Geodaetica et Cartographica Sinica,2018,47(1):82-90.

[18] 劉斌,付忠旺. 基于四通道不可分提升小波的多聚焦圖像融合[J].系統(tǒng)工程與電子技術(shù),2018,40(2): 463-471.

[19] Wang Z B, Wang S, Zhu Y. Multi-Focus Image Fusion Based on the Improved PCNN and Guided Filter[J]. Neural Processing Letters, 2017,45(1):75-94.

[20] Li X F, Wang L L, Wang J. Multi-Focus Image Fusion Algorithm Based on Multilevel Morphological Component Analysis and Support Vector Machine[J]. IET Image Processing, 2017,11(10): 919-926.

[21] 李艷瑋,鄭偉勇. 基于目標提取和信息逼真度的圖像融合[J]. 計算機工程與設(shè)計,2017,38(4):965-969+988.

Multi Focus Image Fusion Algorithm Based on Nonsubsampled Contourlet Transform Coupling Approximation Rule

BAI Tao-tao1, WANG Xi-juan2

(1. Anhui radio and Television University, Chuzhou branch, Chuzhou, Anhui, 239000,China ; 2. Jinggangshan University, Ji’an Jiangxi, 343009, China)

Currently, most of the focus image fusion algorithms mainly use a single ratio to maximize the fusion of high-frequency coefficients, which ignores the approximation among different high-frequency coefficients, resulting in fuzziness and blocking effects in the fusion image. In this paper, a multi-focus image fusion algorithm based on the coupling approximation rule of non-downward contourlet transform is proposed. The non-downsampling contourlet transform is used to decompose the image in multi-scale and multi-direction to obtain the high and low frequency decomposition coefficients. The information of low-frequency coefficients is measured by the region energy of the image, and the fusion function of low-frequency coefficients is constructed for the fusion of low-frequency coefficients. The difference of different high frequency coefficients is measured by the mean gradient difference of the image, and the approximation rule is established. Different fusion methods are used according to the approximation of different high frequency coefficients to complete the fusion of high frequency coefficients. The fusion coefficients are acquired by the non - down sampling inverse transform to get the final fused image. Experimental results show that the proposed method has better image quality than the current multi-focus image fusion method.

image fusion;nonsubsampled contourlet transform; region energy; average gradient; approximation rule

1674-8085(2019)01-0039-06

TP391

A

10.3969/j.issn.1674-8085.2019.01.009

2018-11-07；

2018-12-19

江西省教育廳科學技術(shù)研究項目(GJJ160750)

柏濤濤(1982-)，男，安徽鳳陽人，講師，碩士，主要從事圖像處理、計算機應(yīng)用技術(shù)、大數(shù)據(jù)技術(shù)等方面的研究(E-mail:AnhuiddBotaotao@126.com);

王茜娟(1976-)，女，江西吉安人，副教授，碩士，主要從事圖像處理、模式識別等方面的研究( E-mail: wxj1976@sina.com);

*譚云蘭(1972-)，女，江西新干人，副教授，博士，主要從事圖像處理、模式識別等方面的研究( E-mail: tanyunlan@163.com).