Intelligent fusion method of infrared polarization image based on fireworks algorithm

Wei CHEN; Xiao-Bing SUN; Yan-Li QIAO; Fei-Nan CHEN; Yu-Long YIN

doi:10.11972/j.issn.1001-9014.2020.04.017

Journals >Journal of Infrared and Millimeter Waves >Volume 39 >Issue 4 >Page 523 > Article

Journal of Infrared and Millimeter Waves
Vol. 39, Issue 4, 523 (2020)

Intelligent fusion method of infrared polarization image based on fireworks algorithm

Wei CHEN¹, Xiao-Bing SUN^2、*, Yan-Li QIAO², Fei-Nan CHEN², and Yu-Long YIN³

Author Affiliations

¹Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei23003,China

²Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei23003,China

³Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei23003,China

show less

DOI: 10.11972/j.issn.1001-9014.2020.04.017 Cite this Article

Wei CHEN, Xiao-Bing SUN, Yan-Li QIAO, Fei-Nan CHEN, Yu-Long YIN. Intelligent fusion method of infrared polarization image based on fireworks algorithm[J]. Journal of Infrared and Millimeter Waves, 2020, 39(4): 523 Copy Citation Text

show less

Fig. 1. Flow chart of FWA

Download full size | View in the Article

Fig. 2. Convergence process of FWA （a） fitness, （b） explosion amplitude

Download full size | View in the Article

Fig. 3. Fusion results of “ground” infrared-polarization image （a） infrared-intensity image, （b） infrared-polarization image, （c） FWA, （d） LP, （e） GP, （f） PCA, （g） MP, （h） DWT, （i） SIDWT

Download full size | View in the Article

Fig. 4. Fusion results of “truck” infrared-polarization image （a） infrared-intensity image, （b） infrared-polarization image, （c） FWA, （d） LP, （e） GP, （f） PCA, （g） MP, （h） DWT, （i） SIDWT

Download full size | View in the Article

Fig. 5. Fusion results of “car” infrared-polarization image （a） infrared-intensity image, （b） infrared-polarization image, （c） FWA, （d） LP, （e） GP, （f） PCA, （g） MP, （h） DWT, （i） SIDWT

Download full size | View in the Article

火花生成算法伪代码
1: for $j = 1$ to $λ_{i}$ do
2: $e \leftarrow A_{i} \cdot r a n d (- 1,1)$
3: 随机选择 $z$ 个维度， $z \leftarrow r o u n d (d \cdot r a n d (0,1))$
4: $s_{i j} \leftarrow ω_{i}$
5: for $k = 1$ to $z$ do
6: $s_{i j k} \leftarrow s_{i j k} + e$
7: end for
8: end for

Table 1. 规则火花生成伪代码

View in the Article

火花生成算法伪代码
1: for $j = 1$ to $\dot{λ}$ do
2: $e \leftarrow r a n d n (1,1)$
3: 随机选择 $z$ 个维度， $z \leftarrow r o u n d (d \cdot r a n d (0,1))$
4: $g_{j} \leftarrow ω_{i}$
5: for $k = 1$ to $z$ do
6: $g_{j k} \leftarrow g_{j k} \cdot e$
7: end for
8: end for

Table 2. 变异火花生成伪代码

View in the Article

Algorithm	Algorithm setting
FWA	Dim=4, Seed Num=10, Son Num=30, Maximum Iterations=500
LP	Level=4, Highpass combination: choose max, Lowpass combination: average
GP	Level=4, Highpass combination: choose max, Lowpass combination: average
PCA
MP	Level=4, Highpass combination: choose max, Lowpass combination: average
DWT	Level=4, Highpass combination: choose max, Lowpass combination: average
SIDWT	Level=4, Highpass combination: choose max, Lowpass combination: average

Table 3. 实验所用融合算法设置

View in the Article

Metrics	RE	FSSIM	FMI
FWA	0.7774	1.4149	1.5462
LP	1.2245	1.3532	1.0964
GP	1.7645	1.4090	1.2368
PCA	1.6119	1.1061	1.7559
MP	0.9401	1.2618	1.0168
DWT	1.2921	1.3124	1.0137
SIDWT	1.2953	1.3649	1.0908

Table 4. “ground”的红外强度和偏振度图融合结果客观评价

View in the Article

Metrics	RE	FSSIM	FMI
FWA	0.3253	1.7744	3.3937
LP	0.3798	1.7211	2.1439
GP	0.4080	1.7701	2.2947
PCA	0.3385	1.7955	2.7763
MP	0.3947	1.6492	1.9312
DWT	0.3915	1.6893	2.1045
SIDWT	0.3871	1.7409	2.1856

Table 5. “truck”的红外强度和偏振度图融合结果客观评价

View in the Article

Metrics	RE	FSSIM	FMI
FWA	2.2053	1.2559	2.3473
LP	2.9673	1.1467	1.3457
GP	3.0748	1.2267	1.5746
PCA	2.6545	1.1711	3.3017
MP	2.9885	1.0695	1.2487
DWT	2.9697	1.1477	1.2587
SIDWT	2.9647	1.1764	1.3705

Table 6. “car”的红外强度和偏振度图融合结果客观评价

View in the Article

Metrics

（2a,2f）

（2b,2f）

（3a,3f）

（3b,3f）

（4a,4f）

（4b,4f）

0.2669

1.4890

1.9571

0.8192

2.9449

0.3568

Table 7. 互信息计算结果

View in the Article

	ground	truck	car
RE	FWA>MP>LP>DWT>SIDWT>GP	FWA>LP>SIDWT>DWT>MP>GP	FWA>SIDWT>LP>DWT>MP>GP
FSSIM	FWA>GP>SIDWT>LP>DWT>MP	FWA>GP>SIDWT>LP>DWT>MP	FWA>GP>SIDWT>DWT>LP>MP
FMI	FWA>GP>LP>SIDWT>MP>DWT	FWA>GP>SIDWT>LP>DWT>MP	FWA>GP>SIDWT>LP>DWT>MP

Table 8. 算法性能排序

Download Citation

Save the article for my favorites

Paper Information