激光与光电子学进展, 2019, 56 (7): 072001, 网络出版: 2019-07-30
基于多特征卷积神经网络的手写公式符号识别 下载: 1692次
Handwritten Formula Symbol Recognition Based on Multi-Feature Convolutional Neural Network
光计算 稠密卷积神经网络 手写公式符号 稠密残差块 深度特征 细粒度特征 optics in computing dense convolutional neural network handwritten formula symbols dense residual blocks deep features fine-grained features
摘要
提出了基于多特征稠密卷积神经网络的模型框架(DenseNet-SE)。与传统方法相比,DenseNet-SE采用数据驱动的方法,无需手工提取特征。该框架包含了稠密残差块的结构,能够获取深度特征。通过跳跃连接的方式,从浅层获取细粒度特征来辅助深度特征。同时,融合特征有助于网络结构获取更多全局信息,更好地表示公式符号的类别。利用在线手写数学表达式识别的竞赛组织(CROHME)提供的标准数学公式符号库来验证所提算法,结果表明,CROHME2014和CROHME2016的识别率分别达到93.38%和92.93%,高于目前已有算法的识别率。
Abstract
A model framework called DenseNet-SE is proposed based on a multi-featured dense convolutional neural network. Compared with the conventional methods, the DenseNet-SE adopts the data-driven approach and the manual extraction of features is not necessary. It contains the dense residual blocks so that the deep features can be acquired. In the jump-joining way, the fine-grained features are obtained from the shallow layers to assist the deep features. The fused features can help the network structure obtain more global information and better represent the categories of formula symbols. The standard mathematical formula symbol library provided by the competition organization on recognition of online handwritten mathematical expression (CROHME) is used to verify the proposed algorithm, results show that the recognition rates of CROHME2014 and CROHME2016 are 93.38% and 92.93%, respectively, higher than those of the existing algorithms.
方定邦, 冯桂, 曹海燕, 杨恒杰, 韩雪, 易银城. 基于多特征卷积神经网络的手写公式符号识别[J]. 激光与光电子学进展, 2019, 56(7): 072001. Dingbang Fang, Gui Feng, Haiyan Cao, Hengjie Yang, Xue Han, Yincheng Yi. Handwritten Formula Symbol Recognition Based on Multi-Feature Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(7): 072001.