一种基于强化学习的中文字库自动生成方法及系统

本发明公布了一种基于强化学习的中文字库自动生成方法及系统，通过创建强化学习模块输出薄板样条TPS插值函数参数，并应用TPS插值函数将标准汉字骨架的每个笔画转化为目标风格；再对笔画实际位置与尺寸进行预测，将笔画骨架组合为汉字骨架；对汉字骨架进行渲染生成目标汉字图片，从而得到完整的中文字库文件。系统包括：强化学习模块、笔画位置预测模块、渲染模块。本发明实现了对标准笔画骨架的修改和笔画边界框的预测，确保生成更好的目标风格汉字骨架。模型结构简单，无需进行大规模的预训练处理，即可合成高质量的汉字骨架，自动生成高质量的中文字库。 The invention discloses a Chinese character library automatic generation method and system based on reinforcement learning, and the method comprises the steps of building a reinforcement learning module to output a thin plate spline TPS interpolation function parameter, and converting each stroke of a standard Chinese character skeleton into a target style through employing a TPS interpolation function; predicting actual positions and sizes of strokes, and combining stroke skeletons into a Chinese character skeleton; and rendering the Chinese character skeleton to generate a target Chinese character picture so as to obtain a complete Chinese character library file. The system comprises a reinforcement learning module, a stroke position prediction modul