The Ancient Pictogram Recognition Based on Convolution Neural Network

2017 16th International Symposium on Distributed Computing and Applications to Business, Engineering and Science (DCABES) Pub Date : 2017-10-01 DOI:10.1109/DCABES.2017.28

Qiao Cui, Yutong Zheng

引用次数: 0

Abstract

We design and implement a recognition system of ethnic pictogram based on convolutional neural networks, and obtain a high recognition rate in the test samples. In addition, we also test the characters of simulated natural scenes, and also achieve higher accuracy.

查看原文本刊更多论文

基于卷积神经网络的古代象形文字识别

设计并实现了一种基于卷积神经网络的民族象形文字识别系统，并在测试样本中获得了较高的识别率。此外，我们还对模拟自然场景的人物进行了测试，也达到了更高的准确率。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2017 16th International Symposium on Distributed Computing and Applications to Business, Engineering and Science (DCABES)

自引率

0.00%

发文量