Table Recognition Based on CNN and Grid Feature
基于细胞神经网络和网格特征的碑刻文字识别

As a carrier of the text, the table has very important historical and artistic value. However, because of a long time, a lot of noises on the rubbings have impact on recognizing text. This paper proposes a new method for this case. Firstly, remove the noises of a single table using CNN technology. And then recognize the table using the grid feature extraction method. The simulation example is based on MATLAB software, and we achieve good results.