-
摘要: 针对目前图像融合过程中的不足之处，结合有限离散剪切波具有高的方向敏感性和抛物尺度化特性，提出了一种有限离散剪切波变换下的图像融合算法。首先对严格配准的多传感器图像进行有限离散剪切波变换，得到低频子带系数和不同尺度不同方向的高频子带系数；然后对低频子带系数采用全局特征值和像素点之间的差异性与区域空间频率匹配度相结合的融合算法，高频方向子带系数采用方向权重对比度与相对区域平均梯度和相对区域方差相结合的方案；最后通过有限离散剪切波逆变换得到融合图像。实验结果表明，与其他的融合算法相比较，本文算法不但有良好的主观视觉效果，而且3幅图像的客观评价指标分别平均提高了0.9%、3.8%、3.1%，2.6%、3.8%、2.9%和1.5%、125%、59%，充分说明了本文融合算法的优越性。Abstract: Aiming at the deficiency of the current image fusion process, combining with good directional sensitivity and parabolic scaling properties of Finite Discrete Shearlet Transform(FDST), a new image fusion algorithm based on FDST is proposed. Firstly, the registration multi sensing images are decomposed by FDST, and the low frequency sub-band coefficients and high frequency sub-band coefficients of different scales and directions are obtained. The fusion principle of low frequency sub-band coefficients is based on the method of combining the differences between global attribute and each pixel with region spatial frequency matching degree. As for high frequency sub-band coefficients, sum of the directional weight contrast can be adopted as the fusion rule, which combines with the relative region average gradient and relative region variance. Finally, the low frequency information and high frequency information are reconstructed to image by Finite Discrete Shearlet Inverse Transform. The results indicate that the algorithm proposed in this paper has a good subjective visual effect, and its quality indexes has been increased averagely by 0.9%、3.8%、3.1%, 2.6%、3.8%、2.9% and 1.5%、125%、59% respectively compared with other fusion algorithms, which shows that the algorithm is superior to other fusion algorithms.
-
Key words:
- image fusion /
- finite discrete shearlet /
- contrast /
- regional average gradient /
- shift-invariant
-
图 3 第一层和第二层分解的方向子带
Figure 3. Directional subband of the first level and second level decomposition
图 4 不同小波域在文章融合策略下对红外与可见光图的融合结果
Figure 4. Infrared and visible image fusion results using fusion strategy of this paper in different wavelet domain
图 5 不同小波域在不同融合策略下对红外与可见光图(5)的融合结果
Figure 5. Infrared and visible image(5) fusion results using different fusion strategies in different wavelet domains
图 6 不同小波域在不同融合策略下对红外与可见光图(6)的融合结果
Figure 6. Infrared and visible image(6) fusion results using different fusion strategies in different wavelet domains
图 7 不同小波域在不同融合策略下对红外与可见光图(7)的融合结果
Figure 7. Infrared and visible image(7) fusion results using different fusion strategies in different wavelet domains
表 1 不同小波域下对红外与可见光图的融合结果比较
Table 1. Fusion results comparison of infrared and visible image in different wavelet domain
Fusion evaluation Infrared and visible image Average running time/s EN MI QAB/F SWT 5.648 1 1.618 8 0.389 4 2.246 5 DWT 5.412 3 1.521 5 0.501 8 2.487 4 NSCT 5.216 5 1.741 1 0.391 1 2.798 1 DTCWT 5.348 9 1.815 6 0.384 1 2.741 6 NSST 5.481 4 1.815 4 0.377 5 3.146 7 FDST 6.841 2 3.924 8 0.661 4 2.194 5 
下载: 导出CSV
表 2 不同小波域在不同融合策略下对红外与可见光图的融合结果比较
Table 2. Fusion results comparison of infrared and visible by using different fusion strategies in different wavelet domain
Fusion evaluation Infrared and visible image(5) Infrared and visible image(6) Infrared and visible image(7) Averagerunning time/s EN MI QAB/F EN MI QAB/F EN MI QAB/F NSST_ HSGLRV 6.825 8 3.038 8 0.447 9 7.231 7 3.152 2 0.605 3 7.432 1 2.439 2 0.548 8 10.457 9 FRWT_HRELV 6.732 3 3.749 1 0.417 8 7.132 7 2.828 2 0.423 4 7.287 8 2.686 1 0.353 9 8.407 3 FDST_LGHRS 6.924 7 2.123 0 0.474 3 7.490 1 2.511 0 0.646 9 7.413 0 3.746 4 0.686 4 9.432 7 FDST_LGVHC 7.000 8 3.870 4 0.573 2 7.349 2 3.868 5 0.708 0 7.487 2 6.460 0 0.781 7 6.228 8
下载: 导出CSV
-
[1] 杨桄, 童涛, 陆松岩, 等.基于多特征的红外与可见光图像融合[J].光学精密工程, 2014, 22(2):489-496. http://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201402034.htmYANG G, TONG T, SONG-YAN L U, et al..Fusion of infra-red and visible images based on multi-features[J].Optics & Precision Engineering, 2014, 22(2):489-496.(in Chinese) http://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201402034.htm [2] MEHRA I, NISHCHAL N K.Wavelet-based image fusion for securing multiple images through asymmetric keys[J].Optics Communications, 2015, 335:153-160. doi: 10.1016/j.optcom.2014.09.040 [3] ZHANG H, CAO X.A way of image fusion based on wavelet transform[C].2013 IEEE 9th International Conference on Mobile Ad-hoc and Sensor Networks, IEEE Computer Society, 2013:498-501. [4] CHENG J, LIU H, LIU T, et al..Remote sensing image fusion via wavelet transform and sparse represent-tation[J].Isprs J.Photogrammetry & Remote Sensing, 2015, 104:158 173. [5] YANG Y.Multimodal medical image fusion through a new DWT based technique[C].2010 4th International Conference on Bioinformatics and Biomedical Engineering(iCBBE), IEEE, 2010:1-4. [6] WANG X, LIU Q, WANG R, et al..Natural image statistics based 3D reduced reference image quality assessment in contourlet domain[J].Neurocomputing, 2015, 151:683 691. doi: 10.1016/j.neucom.2014.05.090 [7] 杨粤涛, 朱明, 贺柏根, 等.采用改进投影梯度非负矩阵分解和非采样Contourlet变换的图像融合方法[J].光学精密工程, 2011, 19(5):1143-1150. doi: 10.3788/OPE.YANG Y T, ZHU M, HE B G, et al..Fusion algori-thm based on improved projected gradient NMF and NSCT[J].Opt.Precision Eng., 2011, 19(5):1143-1150.(in Chinese) doi: 10.3788/OPE. [8] 陈小林, 王延杰.非下采样变换的红外与可见光图像融合[J].中国光学, 2011, 4(5):489-496. http://www.chineseoptics.net.cn/CN/abstract/abstract8724.shtmlCHEN X L, WANG Y J.Infrared and visible image fusion based on nonsubsampled Contourlet transform[J].Chinese Optics, 2011, 04(5):489-496.(in Chinese) http://www.chineseoptics.net.cn/CN/abstract/abstract8724.shtml [9] 张蕾, 金龙旭, 韩双丽, 等.采用非采样Contourlet变换与区域分类的红外和可见光图像融合[J].光学精密工程, 2015, 23(3):810-818. http://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201503027.htmZHANG L, JIN L X, HAN SH L, et al..Fusion of infrared and visual images based on non-sampled contourlet transform and region classification[J].Optics Communications, 2015, 23(3):810-818.(in Chinese) http://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201503027.htm [10] RANGASWAMY Y, RAJA K B, VENUGOPAL K R.FRDF:face recognition using fusion of DTCWT and FFT features[J].Procedia Computer Science, 2015, 54:809-817. doi: 10.1016/j.procs.2015.06.095 [11] 陈贞, 邢笑雪.基于非下采样Shearlet变换与压缩感知的图像融合[J].液晶与显示, 2015(6):1024-1031. http://www.cnki.com.cn/Article/CJFDTOTAL-YJYS201506020.htmCHEN ZH, XING X X.Image fusion algorithm based on non-subsampled Shearlet transform and compressed sensing[J].2015, 30(6):1024-1031.(in Chinese) http://www.cnki.com.cn/Article/CJFDTOTAL-YJYS201506020.htm [12] HAUSER S, STEIDL G.Convex multiclass segmentation with shearlet regularization[J].International J.Computer Mathematics, 2013, 90(1/2):62-81. http://www.researchgate.net/publication/51962975_Convex_Multiclass_Segmentation_with_Shearlet_Regularization [13] 蒲恬, 方庆喆, 倪国强.基于对比度的多分辨率图像融合[J].电子学报, 2000, 12:116-118. http://www.cnki.com.cn/article/cjfdtotal-dzxu200012035.htmPU T, FANG Q ZH, NI G Q.Contrast-based multiresolution image fusion[J].Acta Electronica Sinica, 2000, 12:116-118.(in Chinese) http://www.cnki.com.cn/article/cjfdtotal-dzxu200012035.htm [14] ZHAO CH H, GUO Y T, WANG Y L.A fast fusion scheme for infrared and visible light images in NSCT domain[J].Infrared Physics and Technology, 2015, 72:266-275. doi: 10.1016/j.infrared.2015.07.026 [15] LIU X, ZHOU Y, WANG J.Image fusion based on Shearlet transform and regional features[J].AEU-International J.Electronics and Communications, 2013, 68(6):471-477. [16] 石智, 张卓, 岳彦刚.基于Shearlet变换的自适应图像融合算法[J].光子学报, 2013, 42(1):115-120. doi: 10.3788/gzxbSHI ZH, ZHANG ZH, YUE Y G.Adaptive image fusion algorithm based on shearlet transform[J].Acta Photonica Sinica, 2013, 42(1):115-120.(in Chinese) doi: 10.3788/gzxb [17] 徐小军, 王友仁, 陈帅.基于下采样分数阶小波变换的图像融合新方法[J].仪器仪表学报, 2014(9):2061-2069. http://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201409018.htmXU X J, WANG Y R, CHEN SH.Novel image fusion method based on downsampling fractional wavelet transform[J].Chinese J.Scientific Instrument, 2014(9):2061-2069.(in Chinese) http://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201409018.htm [18] 陈广秋, 高印寒, 刘广文, 等.有限离散剪切波域结合区域客观评价的图像融合[J].吉林大学学报(工学版), 2014, 44 (6):1849-18 59. http://www.cnki.com.cn/Article/CJFDTOTAL-JLGY201406049.htmCHEN G Q, GAO Y H, LIU G W, et al..Image fusion based on area objective assessment in finite discrete shearlet transform domain[J].J.Jilin University(Engineering and Techonlogy Edition), 2014, 44(6):1849-1859.(in Chinese) http://www.cnki.com.cn/Article/CJFDTOTAL-JLGY201406049.htm -
下载:
图(7) / 表(2)
计量
- 文章访问数: 1632
- HTML全文浏览量: 402
- PDF下载量: 806
- 被引次数: 0
