Underwater Image Restoration Method Combining Improved Red Channel Prior and Power Law Correction-based CLAHE Algorithm
-
摘要: 针对水下降质图像复原过程中,存在背景光预估偏差及对比度失衡的问题,提出一种水下图像复原方法。首先根据超像素图像分割方法确定背景光区域及取值,然后采用红通道先验理论求取预估透射率,获得初步复原图像;最终通过归一化幂律校正的限制对比度自适应直方图均衡化(Contrast Limited Adaptive Histogram Equalization,CLAHE)算法增强复原图像的颜色。使用3种图像质量评价标准对实验结果进行客观分析,结果表明,该方法可以有效均衡对比度,提高可视化效果。Abstract: An underwater image restoration method was proposed to address the issues of background light estimation deviation and contrast imbalance that occurs during the restoration of water degradation images. First, the background light area and value are determined according to the superpixel image segmentation method, and then the red channel prior theory is used to obtain the estimated transmittance and the preliminary restored image. Finally, the color of the restored image is enhanced using the normalized power law correction-based contrast limited adaptive histogram equalization (CLAHE) algorithm. Three image quality evaluation standards are used to objectively analyze the experimental results, and it is found that the proposed method can effectively balance the contrast and improve the visualization effect.
-
表 1 各项评价指标值
Table 1. Evaluation index values
Image name Algorithm Information entropy Average gradient MCMA Sea floor Original image 11.889 4 0.671 4 - Scene depth prior 12.266 2 0.833 2 0.466 1 Fusion 14.846 2 1.357 6 0.572 6 Red channel prior 13.981 7 1.198 1 0.545 0 Proposed algorithm 15.031 6 2.450 8 0.579 0 Coral Original image 14.797 1 2.712 6 - Scene depth prior 15.438 5 3.162 8 0.635 9 Fusion 15.759 0 2.839 1 0.603 4 Red channel prior 16.060 6 3.765 4 0.618 3 Proposed algorithm 16.746 3 5.365 3 0.670 1 Fishes Original image 11.622 0 1.170 2 - Scene depth prior 14.890 8 2.687 7 0.589 4 Fusion 15.221 2 4.395 0 0.719 7 Red channel prior 15.385 4 3.436 3 0.711 6 Proposed algorithm 15.948 7 6.651 1 0.764 5 Diver Original image 13.940 8 1.369 7 - Scene depth prior 14.573 1 1.793 6 0.596 5 Fusion 15.522 6 2.835 6 0.716 9 Red channel prior 14.917 5 1.633 1 0.608 5 Proposed algorithm 15.527 2 2.846 9 0.661 3 -
[1] 冯雨, 易本顺, 吴晨玥, 等. 一种红通道反转的水下图像复原算法[J]. 小型微型计算机系统, 2019, 40(1): 194-198. https://www.cnki.com.cn/Article/CJFDTOTAL-XXWX201901038.htmFENG Y, YI B S, WU C Y, et al. Underwater Image Restoration Algorithm Based on Reversed Red-channel[J]. Journal of Chinese Computer Systems, 2019, 40(1): 194-198. https://www.cnki.com.cn/Article/CJFDTOTAL-XXWX201901038.htm [2] 张凯, 金伟其, 裘溯, 等. 水下彩色图像的亮度通道多尺度Retinex增强算法[J]. 红外技术, 2011, 33(11): 14-18. doi: 10.3969/j.issn.1001-8891.2011.11.003ZHANG K, JIN W Q, QIU S, et al. Multi-scale Retinex enhancement algorithm on luminance channel of color underwater image[J]. Infrared Technology, 2011, 33(11): 630-634. doi: 10.3969/j.issn.1001-8891.2011.11.003 [3] HE K M, SUN J, TANG X O. Single Image Haze Removal Using Dark Channel Prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-235. doi: 10.1109/TPAMI.2010.168 [4] LI C Y, GUO J C, PANG Y W, et al. Single underwater image restoration by blue-green channels dehazing and red channel correction[C]// IEEE International Conference on Acoustics, 2016: 1731-1735. [5] 曹美, 盛惠兴, 李庆武, 等. 基于暗原色先验模型的水下彩色图像增强算法[J]. 量子电子学报, 2016, 33(2): 140-147. https://www.cnki.com.cn/Article/CJFDTOTAL-LDXU201602003.htmCAO M, SHENG H X, LI Q W, et al. Underwater color image enhancement algorithm based on dark primary color prior model[J]. Chinese Journal of Quantum Electronics, 2016, 33(2): 140-147. https://www.cnki.com.cn/Article/CJFDTOTAL-LDXU201602003.htm [6] 徐岩, 曾祥波. 基于红色暗通道先验和逆滤波的水下图像复原[J]. 激光与光电子学进展, 2018(2): 221-228. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201802027.htmXU Y, ZENG X B. Underwater image restoration based on red-dark channel prior and inverse filtering[J]. Laser & Optoelectronics Progress, 2018(2): 221-228. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201802027.htm [7] 李炼, 李维嘉, 吴耀中. 基于红色暗通道先验理论与CLAHE算法的水下图像增强算法[J]. 中国舰船研究, 2019, 14(s1): 175-182. https://www.cnki.com.cn/Article/CJFDTOTAL-JCZG2019S1027.htmLI L, LI W J, WU Y Z. An underwater image enhancement algorithm based on RDCP and CLAHE[J]. Chinese Journal of Ship Research, 2019, 14(s1): 175-182. https://www.cnki.com.cn/Article/CJFDTOTAL-JCZG2019S1027.htm [8] Ancuti C, Ancuti C O, Haber T, et al. Enhancing Underwater Images and Videos by Fusion[C]// IEEE Conference on Computer Vision & Pattern Recognition, 2012: . 81-88 [9] Peng Y T, Cosman P C. Underwater Image Restoration Based on Image Blurriness and Light Absorption[J]. IEEE Transactions on Image Processing, 2017, 26(4): 1579-1594. doi: 10.1109/TIP.2017.2663846 [10] SHI Z H, FENG, Y N, ZHAO, M H, et al. Normalised gamma transformation-based contrast-limited adaptive histogram equalisation with colour correction for sand–dust image enhancement[J]. IET Image Processing, 2020, 14(4): 747-756. doi: 10.1049/iet-ipr.2019.0992 [11] 暴婉婷, 王俊平, 魏书蕾, 等. 天空区域分割修正的图像去雾新算法[J]. 西安电子科技大学学报, 2019, 46(2): 170-175. https://www.cnki.com.cn/Article/CJFDTOTAL-XDKD201902027.htmBAO W T, WANG J P, WEI Shulei, et al. Novel image defogging algorithm for sky region segmentation correction[J]. Journal of Xidian University: Natural science, 2019, 46(2): 170-175. https://www.cnki.com.cn/Article/CJFDTOTAL-XDKD201902027.htm [12] Galdran A, Pardo D, Picón A, et al. Automaticred-channel underwater image restoration[J]. Journal of Visual Communication and Image Representation, 2015, 26: 132-145. doi: 10.1016/j.jvcir.2014.11.006 [13] ZHAO X W, Jin T, Qu S. Deriving inherent optical properties from background color and underwater image enhancement[J]. Ocean Engineering, 2015, 94: 163-172. doi: 10.1016/j.oceaneng.2014.11.036 [14] Abdoli M, Nasiri F, Brault P, et al. Quality assessment tool for performance measurement of image contrast enhancement methods[J]. Image Processing, IET, 2019, 13(5): 833-842. doi: 10.1049/iet-ipr.2018.5520