融合域迁移和注意力机制的水下图像增强

打开文本图片集

YAO Tingting*，FENG Zihao，ZHAO Hengxin

（Information Science and Technology College，DalianMaritime University，Dalian116O26，China） *Corresponding author，E-mail：yttlO3o@dlmu.edu.cn

Abstract： Due to the attenuation and scattering of light in an underwater environment，the images directly captured by imaging equipment sufer from significant quality degradation. Although learning-based under water image enhancement methods improve the original image imaging quality to a certain extent，most of the existing methods use artificially synthesized or model-generated paired datasets for training.Meanwhile，there is alarge domain diference between artificial or model-generated images and real underwater images in distribution，which leads to problems of excessve enhancement and no obvious removal of color shift in the enhancement results.Focusing on these problems，an underwater image enhancement model that integrates domain transfer and attention mechanisms was proposed in this paper. First，an image generation network with domain transfer was designed and combined with the physical imaging model and the water type clasifier. In this way，the feature description mapping between images in diferent domains and scenarios could be learned，thereby reducing the difference between the generated imagesand the real images.Furthermore，a multi-scale hybrid atention encoder-decoder network was designed. With the help of efficient feature connections and diferent attention-fused structures，the model's ability to recover local image details was improved. Finally，a global domain association consistency loss function was proposed to better train the network model parameters and improve the quality of image enhancement by constructing content and structure consistent associations of the generated images at each stage of the domain transfer.The proposed model achieved accuracies of 3.140 1，0.602 1 and 3.076 8，O.612 4 for the UIQM and UCIQE metrics on the underwater real datasets UIEB and EUVP，respectively. The experiments show that the proposed model could efectively improve the color recovery ability of underwater images， and more detailscouldbe recovered.

Key words：underwater image enhancement；domain transfer；generative adversarial networks；attntion enhancement

1引言

海洋中蕴含大量的生物和矿产资源，使其成为热门的勘探区域，依靠视觉成像系统拍摄到的水下图像进行海洋探索和资源开发获得了广泛的应用[1]。（剩余16715字）