IEEE Trans Neural Netw Learn Syst. 2020 Nov;31(11):4637-4648. doi: 10.1109/TNNLS.2019.2956965. Epub 2020 Oct 29.
We propose a novel model, called stroke sequence-dependent deep convolutional neural network (SSDCNN), which uses the stroke sequence information and eight-directional features of Chinese characters for online handwritten Chinese character recognition (OLHCCR). SSDCNN learns the representation of OLHCCs by incorporating the natural sequence information of the strokes. Furthermore, it naturally incorporates the eight-directional features. First, SSDCNN inputs the stroke sequence and transforms it into stacks of feature maps following the writing order of the strokes. Second, the fixed-length, stroke sequence-dependent representations of OLHCC are derived through convolutional, residual, and max-pooling operations. Third, the stroke sequence-dependent representation is combined with the eight-directional features via a number of fully connected neural network layers. Finally, the Chinese characters are recognized using a softmax classifier. The SSDCNN is trained in two stages: 1) the whole architecture is pretrained using the training data until the performance converges to an acceptable degree. 2) The stroke sequence-dependent representation is combined with the eight-directional features by a fully connected neural network and a softmax layer for further training. The model was experimentally evaluated on the OLHCCR competition tasks of International Conference on Document Analysis and Recognition (ICDAR) 2013. The recognition error was a maximum 58.28% lower in SSDCNN than in a model using the eight-directional features alone (5.13% versus 2.14%). Owing to its high accuracy (97.86%), the proposed SSDCNN reduced the recognition error by approximately 18.0% as compared with that of the winning system in the ICDAR 2013 competition. SSDCNN integrated with an adaptation mechanism, called the SSDCNN+Adapt model, and reached a new state-of-the-art (SOTA) standard with an accuracy of 97.94%. The SSDCNN exploits the stroke sequence information to learn high-quality OLHCC representations. Moreover, the learned representation and the classical eight-directional features complement each other within the SSDCNN architecture.
我们提出了一种新的模型，称为笔画序列相关的深度卷积神经网络（SSDCNN），它使用汉字的笔画序列信息和八方向特征进行在线手写汉字识别（OLHCCR）。SSDCNN 通过结合笔画的自然序列信息来学习 OLHCC 的表示。此外，它自然地结合了八方向特征。首先，SSDCNN 输入笔画序列，并按照笔画的书写顺序将其转换为特征图的堆叠。其次，通过卷积、残差和最大池化操作，得到与笔画序列相关的固定长度的 OLHCC 表示。第三，通过多个全连接神经网络层将与笔画序列相关的表示与八方向特征相结合。最后，使用 softmax 分类器识别汉字。SSDCNN 分两个阶段进行训练：1）使用训练数据对整个架构进行预训练，直到性能收敛到可接受的程度。2）通过全连接神经网络和 softmax 层将与笔画序列相关的表示与八方向特征相结合进行进一步训练。该模型在 ICDAR 2013 年的 OLHCCR 竞赛任务中进行了实验评估。与仅使用八方向特征的模型相比（5.13%对 2.14%），SSDCNN 的识别错误降低了 58.28%。由于其高精度（97.86%），与 ICDAR 2013 竞赛中的获奖系统相比，所提出的 SSDCNN 将识别错误降低了约 18.0%。SSDCNN 与一种称为 SSDCNN+Adapt 的自适应机制集成，并达到了新的 SOTA 标准，准确率为 97.94%。SSDCNN 利用笔画序列信息来学习高质量的 OLHCC 表示。此外，在 SSDCNN 架构内，学习到的表示和经典的八方向特征相互补充。
