改进的CycleGAN网络用于水下显微图像颜色校正  

Improved CycleGAN network for underwater microscopic image color correction

作　　者：王昊天 刘庆省 陈亮[1] 叶旺全 卢渊[1] 郭金家[1] 郑荣儿[1] WANG Haotian;LIU Qingsheng;CHEN Liang;YE Wangquan;LU Yuan;GUO Jinjia;ZHENG Ronger(College of Physics and Optoelectronic Engineering,Faculty of Information Science and Engineering,Ocean University of China,Qingdao 266100,China)

机构地区：[1]中国海洋大学信息科学与工程学部物理与光电工程学院,山东青岛266100

出　　处：《光学精密工程》2022年第12期1499-1508,共10页Optics and Precision Engineering

基　　金：山东省重点研发计划(国际科技合作)项目(No.2019GHZ010);山东省自然科学基金项目(No.ZR2020MF123)。

摘　　要：针对海洋水体及悬浮颗粒物吸收和散射所导致的水下显微图像的颜色信息失真问题,本文提出了一种改进的循环一致性对抗网络(Cycle-consistent Adversarial Network,CycleGAN)算法,实现对水下目标物图像的颜色校正。通过在原始水下降质图像和重构水下图像之间加入R、G、B三个通道的结构相似性(Structure Similarity Index Measure,SSIM)损失函数,度量二者图像之间的信息损失,进而实现R、G、B三个通道颜色的精准调控,不仅增强了CycleGAN网络的整体性能,也提高了生成器生成图像的质量。然后,利用水下多色自制标靶及天然矿石的显微图像组成的训练数据集对本文所提的改进网络进行训练,所得的模型可用于实际矿石样品表面的显微图像颜色校正。结果表明,本文所提的改进的CycleGAN算法较其它方法在颜色校正方面有着明显的优势。与传统的Retinex算法相比,处理后的图像峰值信噪比(Peak Signal-to-Noise Ratio,PSNR)和结构相似性指标分别提高41.85%、35.62%,而且,从主观视觉角度可发现经过校正的水下显微图像与空气中图像颜色相似度最高。综上,本文方法可以有效地对水下目标物图像进行颜色校正,并提升水下显微图像的质量,有望在海洋地质和海洋生物学方面得到应用。The absorption and scattering of light by marine water and suspended particles lead to the distortion of color in underwater microscopic images. This paper presents an improved cycle generative adversarial network(CycleGAN)algorithm for effectively correcting the color of microscopic images of underwater targets. The structural similarity index(SSIM)loss function,which measures the loss of color information among images,of the R,G,and B channels was added between the original underwater images and the reconstructed images. Therefore,the color of the R,G and B channels was accurately regulated.This enhanced not only the overall performance of the CycleGAN,but also the quality of images produced by the generator. Subsequently,the improved network was trained by using a training data set,which consisted of underwater multicolor self-made target images and microscopic images of natural stones. The trained network model was used to correct the color of the microscopic images of underwater stones. The results showed that the improved CycleGAN algorithm had distinct advantages in color correction over other methods. The peak signal-to-noise ratio and SSIM of the images processed by using this algorithm were41. 85% and 35. 62% higher than those processed by using the traditional Retinex algorithm,respectively. Moreover,in terms of subjective vision,the corrected underwater microscopic images had the highest color similarity with the images taken in air. In conclusion,this method can effectively correct the color of underwater target images and improve the quality of underwater microscopic images. It can be applied in marine geology and marine biology.

关 键 词：海底深部微生物观测 水下显微成像 SSIM损失函数 CycleGAN 颜色校正 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] TH691.9[自动化与计算机技术—计算机科学与技术]