A fast HMM algorithm for on-line handwritten character recognition
A fast HMM algorithm for on-line handwritten character recognition
复制标题
一种在线手写字符识别的快速HMM算法
DOI：
10.1109/icdar.1997.619873
复制
发表时间：
1997
期刊：
影响因子：
--
通讯作者：
T. Matsumoto
中科院分区：
文献类型：
--
作者：
K. Takahashi;H. Yasuda;T. Matsumoto
提出了一种用于联机手写字符识别的快速隐马尔可夫模型（HMM）算法。在预处理之后，输入笔划被离散化，使得可以使用离散HMM。这种特殊的离散化自然会导致分配初始状态和状态转移概率的简单过程。在训练阶段，不执行关于状态的完全边缘化（约束维特比）。一个简单的平滑/地板程序产生快速和强大的学习。基于归一化的最大似然比的标准，给出了用于决定何时创建一个新的模型为同一个字符在学习阶段，以科普笔划顺序的变化和大的形状变化。在东京农业技术大学的新Kuchibue数据库上进行了初步实验。结果似乎令人鼓舞。
A fast HMM algorithm is proposed for on-line hand written character recognition. After preprocessing input strokes are discretized so that a discrete HMM can be used. This particular discretization naturally leads to a simple procedure for assigning initial state and state transition probabilities. In the training phase, complete marginalization with respect to state is not performed (constrained Viterbi). A simple smoothing/flooring procedure yields fast and robust learning. A criterion based on the normalized maximum likelihood ratio is given for deciding when to create a new model for the same character in the learning phase, in order to cope with stroke order variations and large shape variations. Preliminary experiments are done on the new Kuchibue database from the Tokyo University of Agriculture and Technology. The results seem to be encouraging.