吉林大学学报（理学版）2025，Vol.63Issue(3)：804-814,11.DOI:10.13413/j.cnki.jdxblxb.2023490
基于数据增强循环生成对抗网络的图像水墨画风格迁移方法
Style Transfer Method of Image Ink Painting Based on Data Enhanced CycleGAN
摘要
Abstract
Aiming at the problem of poor effect of the style transfer of existing image ink painting,we proposed a new data enhanced cycle generative adversarial network(GAN)for the style transfer of ink painting of unpaired natural landscape photos.Firstly,the binary synthesizer and discriminator structure was designed to effectively improve the mapping constraints of one-way GAN models.Secondly,we used multiple loss functions to optimize the model,introduced total variational loss and identity mapping loss,and designed a new cyclic consistency loss function combined with multi-scale structural similarity to better capture the characteristics of traditional ink painting.Finally,data enhancement techniques were used to increase the amount and variety of real and generated data to improve generator performance.The comparative experimental results show that this method can effectively transfer natural landscape images to traditional ink painting style images.关键词
循环生成对抗网络/图像水墨画风格迁移/损失函数/数据增强Key words
CycleGAN/style transfer of image ink painting/loss function/data enhancement分类
信息技术与安全科学引用本文复制引用
李伟伟,傅博,王贺霏,孙文燕,薛玉利..基于数据增强循环生成对抗网络的图像水墨画风格迁移方法[J].吉林大学学报（理学版）,2025,63(3):804-814,11.基金项目
国家自然科学基金(批准号:61702246)、山东省本科教学改革研究重点项目(批准号:Z2024429)和山东青年政治学院博士科研启动基金(批准号:XXPY21023). （批准号:61702246）
