一种汉字字体家族生成算法  

A Chinese Font Family Generation Algorithm

作　　者：李莉 徐张昕子 王存睿 战国栋 LI Li;XU Zhangxinzi;WANG Cunrui;ZHAN Guodong(School of Computer Science and Engineering,Dalian Minzu University,Dalian Liaoning 116605,China;School of Design,Dalian Minzu University,Dalian Liaoning 116605,China;Dalian Chinese Font Design Technology Innovation Center,Dalian Minzu University,Dalian Liaoning 116605,China)

机构地区：[1]大连民族大学计算机科学与工程学院,辽宁大连116605 [2]大连民族大学设计学院,辽宁大连116605 [3]大连民族大学大连市汉字计算机字库设计技术创新中心,辽宁大连116605

出　　处：《大连民族大学学报》2024年第5期439-443,480,共6页Journal of Dalian Minzu University

基　　金：大连市创新基金(2023JJGX026);民族教育信息化教育部重点实验室联合基金(EIN2024B002)。

摘　　要：为提高字体家族生成效果,针对字体家族特性,调整网络关注字体的局部特征和字体笔画相交的部分,提出融合SCConv和注意力机制的字体家族生成方法。该方法使用空间通道重建卷积代替普通卷积来提高网络生成效率,同时模型加入置换注意力(ShuffleAttention,SA),将字体特征进行混合后分组用以增加字体特征交互,使用多头注意力机制对字体特征进行融合。实验结果表明:本模型生成字体家族字符在指标和效果优于DG-Font算法。To enhance the effectiveness of font family generation,this paper proposes a method that integrates SCConv and attention mechanisms,specifically targeting the unique characteristics of font families by adjusting the network to focus on local font features and stroke intersections.This method replaces standard convolution with spatial-channel reconstruction convolution(SCConv)to improve the efficiency of network generation.Additionally,the model incorporates shuffle attention(SA),which combines and groups font features to increase feature interaction,and uses a multi-head attention mechanism to fuse the font features.Experimental results show that the proposed model outperforms the DG-Font algorithm in both metrics and visual outcomes for generating font family characters.

关 键 词：字体家族 字体生成 注意力机制 空间通道重建卷积 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]