基于暗通道先验的短波红外图像去雾

刘燕晴; 李中文; 于世孔; 刘芸邑; 姚文婷; 葛志浩; 吉莉; 张宝辉

基于暗通道先验的短波红外图像去雾

刘燕晴^1,,
李中文²,
于世孔²,
刘芸邑²,
姚文婷²,
葛志浩²,
吉莉²,
张宝辉^{1, 2, ,}

1.
昆明物理研究所, 云南昆明 650223
2.
北方夜视科技(南京)研究院有限公司, 江苏南京 211106

详细信息

作者简介:
刘燕晴（1998-），女，硕士研究生，研究方向为红外图像处理。E-mail：lyq210853291@163.com

通讯作者:
张宝辉（1984-），男，博士，正高级工程师，主要研究方向为红外探测与图像处理。E-mail：zbhmatt@163.com

中图分类号: TP391
计量
- 文章访问数: 212
- HTML全文浏览量: 53
- PDF下载量: 73
出版历程
- 收稿日期: 2023-06-14
- 修回日期: 2023-08-06
- 刊出日期: 2023-09-19

Shortwave Infrared Image Dehazing Based on Dark Channel Prior

1.
Kunming Institute of Physics, Kunming 650223, China
2.
North Night-Vision Science & Technology (Nanjing) Research Institute Co. LTD., Nanjing 211106, China

摘要

摘要: 针对短波红外成像系统在雾霾天气下存在图像质量模糊、分辨率低等问题，本文提出了一种基于暗通道先验理论的短波红外图像去雾算法。本文首先通过改进的暗通道先验得到暗通道图像数据，然后基于暗通道数据对大气光进行估计；为了避免目标局部高亮或细节模糊，采用引导滤波和多尺度Retinex（Multi-scale retinex，MSR）对透射率图进行细化和增强处理，最后结合大气散射模型来反演出去雾图像。实验结果表明，经此算法处理后的短波红外图像在主观视觉和客观指标方面均得到了较好的验证，去雾效果显著、细节特征丰富且明亮度适宜。
- 短波红外 /
- 去雾 /
- 暗通道先验 /
- MSR
Abstract: To solve the problems of blurred image quality and low-resolution weather haze in shortwave infrared imaging systems, a shortwave infrared image-defogging algorithm based on a dark channel prior is proposed. First, the algorithm obtains the dark-channel image data using an improved dark-channel prior. Then, the atmospheric light is estimated based on the dark channel data. To avoid local highlights or blurred details of the target, the transmittance map is refined and enhanced using guided filtering and multi-scale retinex (MSR). Finally, the defogged image is inverted using the atmospheric scattering model. The shortwave infrared image processed by this algorithm was verified in terms of subjective vision and objective indicators, displaying a remarkable defogging effect, rich details, and appropriate brightness.
- shortwave infrared /
- dehaze /
- dark channel prior /
- MSR

HTML全文

图 1 本文算法框图

Figure 1. The method block diagram of this paper

下载: 全尺寸图片幻灯片

图 2 暗通道对比图

Figure 2. Dark channel comparison chart

下载: 全尺寸图片幻灯片

图 3 透射率修正前(a)与透射率修正后(b)对比图

Figure 3. Comparison of transmittance before correction (a) and after transmittance correction (b)

下载: 全尺寸图片幻灯片

图 4 场景1雾霾图像对比图

Figure 4. Comparison of haze images in scene 1

下载: 全尺寸图片幻灯片

图 5 场景2雾霾图像对比图

Figure 5. Comparison of haze images in scene 2

下载: 全尺寸图片幻灯片

图 6 场景3雾霾图像对比图

Figure 6. Comparison of haze images in scene 3

下载: 全尺寸图片幻灯片

表 1 不同算法的峰值信噪比

Table 1 Peak single to noise ratio of different algorithms

Image He^[10] Liu X^[19] Ehsan^[20] Yan S^[21] Proposed

Fig.4 10.3529 8.0672 7.7717 12.4268 12.7400

Fig.5 11.4504 9.0048 9.3301 13.5508 14.2545

Fig.6 10.6362 7.5611 8.2787 12.4144 12.7427

下载: 导出CSV

表 2 不同算法的平均梯度

Table 2 Average gradient of different algorithms

Image He^[10] Liu X^[19] Ehsan^[20] Yan S^[21] Proposed

Fig.4 5.1524 6.7746 4.8771 6.0122 9.7126

Fig.5 5.9318 6.5066 5.5713 7.0088 10.7187

Fig.6 5.9800 6.7071 7.1609 7.1295 10.7249

下载: 导出CSV

表 3 不同算法的信息熵

Table 3 Information entropy of different algorithms

Image He^[10] Liu X^[19] Ehsan^[20] Yan S^[21] Proposed

Fig.4 6.6543 7.1055 6.5844 6.9944 7.2217

Fig.5 6.5985 6.6076 6.4842 7.2029 7.2868

Fig.6 6.8458 7.2238 7.0303 7.0750 7.2633

下载: 导出CSV

参考文献(22)

[1]	WU Q, ZHANG J, REN W, et al. Accurate transmission estimation for removing haze and noise from a single image[J]. IEEE Transactions on Image Processing, 2020, 29: 2583-2597. DOI: 10.1109/TIP.2019.2949392
[2]	WANG P, FAN Q, ZHANG Y, et al. A novel dehazing method for color fidelity and contrast enhancement on mobile devices[J]. IEEE Transactions on Consumer Electronics, 2018, 65(1): 47-56.
[3]	Yoav Y, Schechner, Srinivasa G, et al. Polarization-based vision through haze[J]. Applied Optics, 2003, 42: 511-525. DOI: 10.1364/AO.42.000511
[4]	Narasimhan S G, Nayar S K. Vision and the atmosphere[C]// International Conference on Computer Graphics and Interactive Techniques. ACM, 2008, 48: 233-254.
[5]	XU Z, LIU X, CHEN X. Fog removal from video sequences using contrast limited adaptive histogram equalization[C]//International Conference on Computational Intelligence & Software Engineering. IEEE, 2009: 1-4.
[6]	LI Jia. Application of image enhancement method for digital images based on Retinex theory[J]. Optik - International Journal for Light and Electron Optics, 2013, 124(23): 5986-5988. DOI: 10.1016/j.ijleo.2013.04.115
[7]	Seow M J, Asari V K. Ratio rule and homomorphic filter for enhancement of digital colour image[J]. Neurocomputing, 2006(7/9): 69.
[8]	BO J, ZHANG W, ZHAO J, et al. Gray-Scale image dehazing guided by scene depth information[J]. Mathematical Problems in Engineering, 2016, 2016: 1-10.
[9]	Tarel J P, N Hautière. Fast visibility restoration from a single color or gray level image[C]//IEEE International Conference on Computer Vision. IEEE, 2010: 2201-2208.
[10]	HE K. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 33: 2341-2353.
[11]	ZHU Q, MAI J, SHAO L. A fast single image haze removal algorithm using color attenuation prior[J]. IEEE Transactions on Image Processing, 2015, 24(11): 3522-3533. DOI: 10.1109/TIP.2015.2446191
[12]	LI J, ZHANG H, YUAN D, et al. Single image dehazing using the change of detail prior[J]. Neurocomputing, 2015, 156: 1-11. DOI: 10.1016/j.neucom.2015.01.026
[13]	Singh V. Single image dehazing using gradient channel prior[J]. Applied Intelligence: The International Journal of Artificial Intelligence, Neural Networks, and Complex Problem-Solving Technologies, 2019, 49(12): 4276-4293.
[14]	林森, 孙彭辉. 改进暗通道窗口与透射率修正的图像去雾[J]. 电光与控制, 2022, 29(11): 55-60. https://www.cnki.com.cn/Article/CJFDTOTAL-DGKQ202211010.htm LIN Sen, SUN Penghui. Improved image dehazing with dark channel window and transmittance correction[J]. Electrooptic and control, 2022, 29(11): 55-60. https://www.cnki.com.cn/Article/CJFDTOTAL-DGKQ202211010.htm
[15]	Mccann J. Retinex Theory[M]. New York: Springer, 2015.
[16]	李娜, 邓家先, 崔亚妮, 等. 基于暗通道先验的红外图像清晰化及FPGA实现[J]. 红外与激光工程, 2021, 50(3): 111-120. https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ202103014.htm LI Na, DENG Jiaxian, CUI Yani, et al. Infrared image clarity based on dark channel prior and FPGA implementation[J]. Infrared and Laser Engineering, 2021, 50(3): 111-120. https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ202103014.htm
[17]	HE K, SUN J, TANG X. Guided image filtering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(6): 1397-1409. DOI: 10.1109/TPAMI.2012.213
[18]	张承泓, 李范鸣, 吴滢跃. 基于自适应引导滤波的子带分解多尺度Retinex红外图像增强[J]. 红外技术, 2019, 41(4): 31-36. http://hwjs.nvir.cn/article/id/hwjs201904005 ZHANG Chenghong, LI Fanming, WU Yingyue. Multi-scale Retinex infrared image enhancement with subband decomposition based on adaptive guided filtering[J]. Infrared Technology, 2019, 41(4): 31-36. http://hwjs.nvir.cn/article/id/hwjs201904005
[19]	LIU X, ZHANG H, CHEUNG Y M, et al. Efficient single image dehazing and denosing[J]. Computer Vision and Image Understanding, 2017: 23-33. Doi: 10.1016/j.cviu.2017.08.002.
[20]	Ehsan S M, Imran M, Ullah A, et al. A single image dehazing technique using the dual transmission maps strategy and gradient-domain guided image filtering[J]. IEEE Access, 2021(99): 89055-89063. Doi: 10.1109/ACCESS.2021.3090078.
[21]	YAN S, ZHU J, YUN K, et al. An infrared image dehazing method based on modified dark channel prior[C]//International Conference on Biometrics, Microelectronic Sensors, and Artificial Intelligence (BMSAI). SPIE, 2022, 12252: 132-138.
[22]	程铁栋, 卢晓亮, 易其文, 等. 一种结合单尺度Retinex与引导滤波的红外图像增强方法[J]. 红外技术, 2021, 43(11): 1081. http://hwjs.nvir.cn/article/id/b49a0a09-e295-40e6-9736-24a58971206e CHENG Tiedong, LU Xiaoliang, YI Qiwen, et al. An infrared image enhancement method combining single-scale Retinex and guided filtering[J]. Infrared Technology, 2021, 43(11): 1081-1088. http://hwjs.nvir.cn/article/id/b49a0a09-e295-40e6-9736-24a58971206e

施引文献(24)

期刊类型引用(10)

1.	林碧洪，冯雪兰. 基于遗传算法的预压缩图像破损数据重构. 信息技术. 2024(06): 100-104 . 百度学术
2.	吴宗翔，刘立群. 基于自注意力机制和RSA加密的图像融合算法. 软件工程. 2024(09): 20-25 . 百度学术
3.	杨艳春，王可，闫岩. 基于FRGF和改进PCNN的红外可见光图像融合. 激光与红外. 2023(10): 1593-1601 . 百度学术
4.	廖宁，陈怡然. 红外和可见光图像高效融合的人工智能技术研究. 激光杂志. 2022(03): 109-113 . 百度学术
5.	冯青，郭昊麟. 数字化视域下的宋代缂丝设计元素提取与应用. 包装工程. 2022(12): 250-256+272 . 百度学术
6.	李威，李忠民. 一种基于EASSF的红外与可见光图像视觉保真度融合. 红外技术. 2022(07): 686-692 . 本站查看
7.	王欧，罗小波. 基于细节信息提取的全色与多光谱图像融合方法. 红外技术. 2022(09): 920-928 . 本站查看
8.	张洲宇，曹云峰. 采用局部约束线性编码的像素级舰船目标图像融合. 导航定位与授时. 2021(02): 120-129 . 百度学术
9.	李艳. 基于立体双目视觉的模糊图像视觉优化算法. 吉林大学学报(信息科学版). 2021(06): 656-661 . 百度学术
10.	杨蕾蕾，吐尔洪江·阿布都克力木. 基于滚动导向滤波的单幅图像去雨算法. 河南科技. 2021(20): 24-26 . 百度学术

其他类型引用(14)

资源附件(0)

图(6) / 表(3)

计量

文章访问数: 212
HTML全文浏览量: 53
PDF下载量: 73
被引次数: 24

Image	He^[10]	Liu X^[19]	Ehsan^[20]	Yan S^[21]	Proposed
Fig.4	10.3529	8.0672	7.7717	12.4268	12.7400
Fig.5	11.4504	9.0048	9.3301	13.5508	14.2545
Fig.6	10.6362	7.5611	8.2787	12.4144	12.7427

Image	He^[10]	Liu X^[19]	Ehsan^[20]	Yan S^[21]	Proposed
Fig.4	5.1524	6.7746	4.8771	6.0122	9.7126
Fig.5	5.9318	6.5066	5.5713	7.0088	10.7187
Fig.6	5.9800	6.7071	7.1609	7.1295	10.7249

Image	He^[10]	Liu X^[19]	Ehsan^[20]	Yan S^[21]	Proposed
Fig.4	6.6543	7.1055	6.5844	6.9944	7.2217
Fig.5	6.5985	6.6076	6.4842	7.2029	7.2868
Fig.6	6.8458	7.2238	7.0303	7.0750	7.2633

基于暗通道先验的短波红外图像去雾

作者简介: 刘燕晴（1998-），女，硕士研究生，研究方向为红外图像处理。E-mail：lyq210853291@163.com

通讯作者: 张宝辉（1984-），男，博士，正高级工程师，主要研究方向为红外探测与图像处理。E-mail：zbhmatt@163.com

计量

出版历程

Shortwave Infrared Image Dehazing Based on Dark Channel Prior

期刊类型引用(10)

其他类型引用(14)

计量

出版历程

目录

作者简介:
刘燕晴（1998-），女，硕士研究生，研究方向为红外图像处理。E-mail：lyq210853291@163.com

通讯作者:
张宝辉（1984-），男，博士，正高级工程师，主要研究方向为红外探测与图像处理。E-mail：zbhmatt@163.com