Journal of Computer Applications ›› 2020, Vol. 40 ›› Issue (3): 638-644.DOI: 10.11772/j.issn.1001-9081.2019081461

• Artificial intelligence • Previous Articles     Next Articles

Image style transfer network based on texture feature analysis

YU Yingdong, YANG Yi, LIN Lan   

  1. College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China
  • Received:2019-08-22 Revised:2019-10-15 Online:2020-03-10 Published:2019-11-06
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61373106).

结合纹理特征分析的图像风格转换网络

余英东, 杨怡, 林澜   

  1. 同济大学 电子与信息工程学院, 上海 201804
  • 通讯作者: 余英东
  • 作者简介:余英东(1995-),男,江西九江人,硕士研究生,主要研究方向:深度学习、机器学习、图像处理;杨怡(1996-),男,河南洛阳人,硕士研究生,主要研究方向:深度学习、人工智能;林澜(1969-),女,广东台山人,副教授,博士,主要研究方向:图像处理与分析、机器学习、深度学习。
  • 基金资助:
    国家自然科学基金资助项目(61373106)。

Abstract: Focusing on the low efficiency and poor effect of image style transfer, a feedforward residual image style transfer algorithm based on pre-trained network and combined with image texture feature analysis was proposed. In the algorithm, the pre-trained deep network was applied to extract the deep features of the style image, and the residual network was used to perform deep training and realize image transfer. Meanwhile, by analyzing the influence of input style image and content image texture on transfer effect, the corresponding measures were adopted for different input images to improve the transfer effect. Experimental results show that the algorithm can achieve better output visual effect, lower normalized style loss and less time consumption. Besides, according to the information entropy and moment invariant calculation of the input image to guide the setting and adjustment of the network parameters, the network was optimized pertinently, and good effect was obtained.

Key words: image style transfer, deep residual network, image texture feature, pre-trained network, Gram matrix

摘要: 针对图像风格转换效率不高、效果不佳的问题,提出一种结合图像纹理特征分析,并基于预训练网络的前馈残差图像风格转换算法。该算法利用预训练深层网络来提取风格图的深度特征,采用残差网络来进行深层训练以及进行图像变换;同时通过分析研究输入风格图与内容图的纹理特征对转换效果的影响,针对不同输入图像采取相应的处理方法来提升转换效果。实验结果表明,与现有深度图像风格转换算法相比,该算法的输出视觉效果更佳,归一化风格损失更小,耗时更短,并且根据输入图像的信息熵与不变矩的计算来指导网络参数的设定与调整,能够针对性地优化网络,取得了良好的效果。

关键词: 图像风格转换, 深度残差网络, 图像纹理特征, 预训练网络, 格拉姆矩阵

CLC Number: