期刊文献+

基于轻量化神经网络的快速汉字笔画分类模型

Chinese Stroke Classification Model Based on Lightweight Neural Network
下载PDF
导出
摘要 在汉字书写场景中,我们可以通过实时捕捉手写者的手写轨迹数据,并可以通过笔画类型、空间关系等构字要素逻辑地位等评价策略,实现汉字书写质量智能测评,以达到指导学习书写的目的。其中笔画分类是手写字体评估中的重要任务。然而,现有笔画级标注的汉字数据集很少,并缺少轻量化并可以应用于嵌入式设备的笔画分类方案,难以工程化落地。在不同设备上采集用户的手写笔画数据(如会议平板、电子白板、学习机等),并使用特征工程,采用一维深度可分离卷积构建轻量化的快速汉字笔画分类模型,并把模型进行量化部署到嵌入式设备中。模型量化后的tflite格式文件大小仅为11kb,而在数据集上达到了97%的精确度。把量化后的模型部署到嵌入式设备中并进行了商业化上线。
出处 《工业控制计算机》 2022年第4期112-114,共3页 Industrial Control Computer
基金 S202111845107基于深度神经网络的情景互动康复训练系统项目 202111845015基于视觉识别的独居老人异常行为检测系统项目资助。
  • 相关文献

参考文献1

二级参考文献9

  • 1Shimodaira H, Sudo T, Nakai M, et al. On-line Overlaid-Hand- writing Recognition Based on Substroke HMMs[C]//Seventh International Conference on Document Analysis and Recogni- tion. Washington DC: IEEE, 2003 : 1043-1047.
  • 2Wan Xiang, Liu Chang-song, Zou Yan-ming. On-line Chinese Character Recognition System for Overlapping Samples[C] // 2011 International Conference on Document Analysis and Re- cognition. Washington DC: IEEE, 2011 : 799-803.
  • 3Zou Yan-ming, Liu Ying-fei, Liu Ying, et al. Overlapped hand- writing input on mobile phones[C]//2011 International Confe- rence on Document Analysis and Recognitiom Washington DC: IEEE, 2011 : 369-373.
  • 4Wang Qiu-feng, Yin Fei, Liu Cheng-lin. Handwritten Chinese Text Recognition by Integrating Multiple Contexts[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012,34(8) : 1469-1481.
  • 5Yin Fei,Wang Da-han,Wang Qiu-feng. CASIA Online and (3ff- line Chinese Handwriting Databases [C]//1 2011 International Conference on Document Analysis and Recognition. Washington DC: IEEE, 2011 : 37-41.
  • 6Wang Da-han, Liu Cheng-lin, Zhou Xiang-dong. An approach for real-time recognition of online Chinese handwritten sentences [J]. Pattern Recognition, 2012,45 (10) : 3661-3675.
  • 7Liu Cheng-lin. Classifier Combination Based on Con{idence Transformation[J]. Pattern Recognition, 2005,38 (1) : 11-28.
  • 8Teng Long,Jin Lian-wen. Building compact MQDF classifier for large character set recognition by subspace distribution sharing [J]. Pattern Recognition, 2008,41 (9) : 2916-2925.
  • 9Wang Yong-qiang, Huo Qiang. Building compact recognizers of handwritten Chinese characters using precision constrained Gaussian model, minimum classification error training and pa- rameter compression [J]. International Journal on Document Analysis and Recognition, 2011,14 (3) : 255-262.

共引文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部