水下视频图像清晰化方法

doi:10.13229/j.cnki.jdxbgxb20190344

摘要/Abstract

摘要：

针对一般水下视频图像清晰化方法使图像失真、噪声放大以及视频播放中相邻帧出现闪烁跳跃等现象，提出了基于颜色空间的单幅水下图像清晰化方法和基于时空信息融合的水下视频清晰化方法。在研究单幅水下图像清晰化过程中，修正了传统方法中透射率估计偏大的现象，并采用暗通道先验与颜色饱和度相结合的方式对背景光进行修正。针对视频播放中相邻帧出现闪烁跳跃现象，将时间、空间信息融合，通过插值和取平均的方式估计出更稳定的透射率。实验结果表明：本文方法在对水下视频图像清晰化处理方面效果良好，同时视频的连续性和平滑性也得到了较好的处理。

关键词: 信息处理技术, 水下视频图像, 颜色空间, 时空信息, 背景光, 透射率

Abstract:

The general method to sharp the underwater video image is vulnerable to provoking issues such as image distortion, noise amplification, and the flicker and jump of adjacent frames in video playback. To overcome these disadvantages, a single color method based on color space as well as the underwater image sharpening method and an underwater video sharpening method based on spatiotemporal information fusion are proposed. In the process of clearing the single underwater image, the phenomenon that the transmittance is estimated to be large in traditional method is polished, and the combination of dark channel prior and color saturation is adopted in order to polish the background light. For the phenomenon of flicker and jump in adjacent frames in video playback, the time and space information are fused, and the more stable transmittance is estimated by interpolation and averaging. The experimental results show that the proposed underwater video sharpening method works effectively in the clearing of underwater video images, resulting in well-processed continuity and smoothness of the video.

Key words: information processing technology, underwater video image, color space, space-time information, background light, transmittance

中图分类号:

TP391.4

程艳芬,姚丽娟,袁巧,陈先桥. 水下视频图像清晰化方法[J]. 吉林大学学报(工学版), 2020, 50(2): 668-677.

Yan-fen CHENG,Li-juan YAO,Qiao YUAN,Xian-qiao CHEN. Clearing strategy of underwater video image[J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(2): 668-677.

图/表 10

图1

图2

图3

图4

图5

图6

图7

图8

表1

水下单幅图像客观评价"

组别	评价指标	算法
组别	评价指标	He	CLAHE	MSRCR	本文
1	$g ˉ$	2.304 8	6.768 3	3.067 8	3.859 9
	$H$	14.635 0	17.723 0	14.059 2	16.644 7
	$E A V$	8.746 4	25.462 4	12.248 6	14.522 4
2	$g ˉ$	3.732 4	10.973 7	5.076 0	5.739 3
	$H$	15.132 0	17.053 3	14.955 3	15.689 7
	$E A V$	18.503 8	55.968 3	24.998 4	33.231 6

表1

表2

水下视频客观评价"

组别	方法	评价指标	第99帧	第103帧	第105帧
1	文献[5]	$g ˉ$	1.407 0	2.740 2	3.272 8
		$H$	14.059 4	14.310 6	14.246 5
		$E A V$	6.524 0	11.089 6	13.080 5
	逐帧处理	$g ˉ$	1.247 7	1.605 7	1.848 5
		$H$	14.519 9	14.990 1	15.212 2
		$E A V$	6.278 7	7.483 7	8.508 5
	本文算法	$g ˉ$	1.207 5	1.538 2	1.776 6
		$H$	14.352 9	14.846 0	15.046 3
		$E A V$	6.217 0	7.417 2	8.391 8
	方法	评价指标	第50帧	第59帧	第70帧
2	文献[5]	$g ˉ$	4.628 5	4.467 7	4.266 9
		$H$	13.282 8	13.291 2	13.584 1
		$E A V$	16.887 7	15.506 4	14.705 5
	逐帧处理	$g ˉ$	7.182 6	6.643 3	6.714 7
		$H$	14.768 2	14.747 9	14.850 5
		$E A V$	26.987 0	23.520 6	23.724 2
	本文算法	$g ˉ$	6.635 8	6.294 8	6.261 9
		$H$	14.459 6	14.355 0	14.442 4
		$E A V$	24.070 9	21.966 8	21.806 7

表2

参考文献 13

1	徐岩, 孙美双. 基于卷积神经网络的水下图像增强方法[J]. 吉林大学学报：工学版, 2018, 48(6): 1895-1903.
	Xu Yan, Sun Mei-shuang. Enhancing underwater image based on convolutional neural networks[J]. Journal of Jilin University (Engineering and Technology Edition), 2018, 48(6): 1895-1903.
2	李一兵, 付强, 张静. 水下模糊图像参数估计复原方法[J]. 吉林大学学报：工学版, 2013, 43(4): 1133-1138.
	Li Yi-bing, Fu Qiang, Zhang Jing. Underwater blurred image restoration based on parameter estimation[J]. Journal of Jilin University (Engineering and Technology Edition), 2013, 43(4): 1133-1138.
3	John J, Wilscy M. Enhancement of weather degraded video sequences using wavelet fusion[C]∥IEEE International Conference on Cybernetic Intelligent Systems, London, UK， 2008: 1-6.
4	Yoon I, Kim S, Kim D, et al. Adaptive defogging with color correction in the HSV color space for consumer surveillance system[J]. IEEE Transactions on Consumer Electronics, 2012, 58(1): 111-116.
5	Qing C, Yu F, Xu X, et al. Underwater video dehazing based on spatial-temporal information fusion[J]. Multidimensional Systems & Signal Processing, 2016, 27(4): 909-924.
6	Mcglamery B L. A computer model for underwater camera systems[J]. Proc Spie, 1980(208): 221-231.
7	杨淼, 纪志成. 基于模糊形态筛和四元数的水下彩色图像增强[J]. 仪器仪表学报, 2012, 33(7): 1601-1605.
	Yang Miao, Ji Zhi-cheng. Underwater color image enhancement based on fuzzy morphological sieve and quaternion[J]. Chinese Journal of Scientific Instrument, 2012, 33(7): 1601-1605.
8	黄冬梅, 宋巍, 王龑, 等. 一种基于相对全局直方图拉伸的浅海水下图像增强方法[P]. 中国CN107067386 A, 2017⁃04⁃24.
9	Berman D, Treibitz T, Avidan S. Non-local image dehazing[C]∥Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 2016: 1674-1682.
10	Wang M, Mai J, Liang Y, et al. Component-based distributed framework for coherent and real-time video dehazing[C]∥IEEE International Conference on Computational Science and Engineering, Guangzhou, China, 2017: 321-324.
11	窦易文, 周鸣争, 唐肝翌, 等. 焦点引导的带颜色恢复的多尺度Retinex算法[J]. 计算机工程与应用,2013, 49(2): 207-210.
	Dou Yi-wen, Zhou Ming-zheng, Tang Gan-yi, et al. Focus-guided multi-scale Retinex algorithm with color restoration[J]. Computer Engineering and Applications, 2013, 49(2): 207-210.
12	He K, Sun J, Tang X. Single image haze removal using dark channel prior[C]∥Computer Vision and Pattern Recognition, 2009, Miami, Florida, USA, 2009: 1956-1963.
13	Hitam M S, Awalludin E A, Wan N J H W Y, et al. Mixture contrast limited adaptive histogram equalization for underwater image enhancement[C]∥International Conference on Computer Applications Technology, Bali, Indonesia, 2013: 1-5.

相关文章 15

[1]	刘富, 权美静, 王柯, 刘云, 康冰, 韩志武, 侯涛. 仿蝎子振源定位机理的位置指纹室内定位方法[J]. 吉林大学学报(工学版), 2019, 49(6): 2076-2082.
[2]	于晓辉,张志成,李新波,孙晓东. 基于状态空间模型的指数衰减正弦信号参数估计[J]. 吉林大学学报(工学版), 2019, 49(6): 2083-2088.
[3]	马子骥,卢浩,董艳茹. 双通道单图像超分辨率卷积神经网络[J]. 吉林大学学报(工学版), 2019, 49(6): 2089-2097.
[4]	郭继昌,吴洁,郭春乐,朱明辉. 基于残差连接卷积神经网络的图像超分辨率重构[J]. 吉林大学学报(工学版), 2019, 49(5): 1726-1734.
[5]	曹运合,曾丽,王宇. 基于特征空间的子阵级自适应和差波束测角方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1735-1744.
[6]	卢洋,王世刚,赵文婷,赵岩. 基于离散Shearlet类别可分性测度的人脸表情识别方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1715-1725.
[7]	董超,刘晶红,徐芳,王仁浩. 光学遥感图像舰船目标快速检测方法[J]. 吉林大学学报(工学版), 2019, 49(4): 1369-1376.
[8]	王柯俨,胡妍,王怀,李云松. 结合天空分割和超像素级暗通道的图像去雾算法[J]. 吉林大学学报(工学版), 2019, 49(4): 1377-1384.
[9]	托乎提努尔,张海龙,王杰,王娜,冶鑫晨,王万琼. 基于图形处理器的高速中值滤波算法[J]. 吉林大学学报(工学版), 2019, 49(3): 979-985.
[10]	付银娟,李勇,徐丽琴,张昆辉. NLFM⁃Costas射频隐身雷达信号设计及分析[J]. 吉林大学学报(工学版), 2019, 49(3): 994-999.
[11]	苏寒松,代志涛,刘高华,张倩芳. 结合吸收Markov链和流行排序的显著性区域检测[J]. 吉林大学学报(工学版), 2018, 48(6): 1887-1894.
[12]	应欢,刘松华,唐博文,韩丽芳,周亮. 基于自适应释放策略的低开销确定性重放方法[J]. 吉林大学学报(工学版), 2018, 48(6): 1917-1924.
[13]	陆智俊,钟超,吴敬玉. 星载合成孔径雷达图像小特征的准确分割方法[J]. 吉林大学学报(工学版), 2018, 48(6): 1925-1930.
[14]	刘仲民,王阳,李战明,胡文瑾. 基于简单线性迭代聚类和快速最近邻区域合并的图像分割算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1931-1937.
[15]	单泽彪,刘小松,史红伟,王春阳,石要武. 动态压缩感知波达方向跟踪算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1938-1944.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed