【作者】 姜映映； 王绪刚； 敖翔； 戴国忠；

【Author】 Jiang Yingying+, Wang Xugang, Ao Xiang, Dai Guozhong (Intelligence Engineering Lab, Institute of Software, Chinese Academy of Science, Beijing P.O.Box 8718, 100080, Beijing)

【机构】 中国科学院软件研究所人机交互与智能信息处理实验室；

【摘要】 手写输入设备日趋成熟,但它并未在化学公式输入领域得到很好的应用。本文提出了一个在线的手写化学公式切分识别算法。算法采用基于识别的切分,充分利用了化学领域知识,采用局部优化的方式,满足了实时性的要求。该算法被应用于一个基于语音和笔输入的多通道化学公式编辑系统中。初步的用户调查表明,该化学公式编辑系统在使用上自然高效,识别率也能满足用户要求更多还原

【Abstract】 The handwriting input devices are becoming mature, but it hasn’t been well used in chemical formula input domain. This paper proposes an online segmentation and recognition algorithm for handwritten chemical formula. The algorithm adopts recognition-based segmentation, utilizes knowledge in chemistry domain and uses local optimization instead of global optimization, which satisfies real-time demand. The recognition result implies a layered structure of chemical formula. A multimodal chemical formula editor is built on the algorithm, in which the user can use speech and pen to input chemical formulas. The speech inputs are translated into digital ink and presented in the form of digital ink. Elementary user investigation shows that the system is natural and efficient to use and the recognition performance basically satisfies users’ demand.更多还原