基于小波统计特征的行块级朝汉文种辨识  被引量:2

Script identification for document images between Chinese and Korean language based on wavelet statistic features at the level of text row

在线阅读下载全文

作  者:金璟璇[1] 崔荣一[1] 崔旭[1] 

机构地区:[1]延边大学工学院计算机科学与技术系智能信息处理实验室,吉林延吉133002

出  处:《延边大学学报(自然科学版)》2013年第4期277-280,共4页Journal of Yanbian University(Natural Science Edition)

摘  要:提出了一种基于小波统计特征的朝汉文种识别方法.首先计算行文档图像的垂直及水平方向的投影;其次对垂直方向的投影及水平方向的投影进行一维小波分解并分别计算小波统计特征,然后将两个方向上的小波统计特征进行合并作为该文档图像的特征向量(本文方法构造的特征向量仅为11维);最后通过神经网络进行训练和测试,结果显示平均测试准确率超过94%.A script identification method between Chinese and Korean language based on wavelet statistic feature is presented. To reduce the dimension and improve calculation efficiency, each 2D row-document image partitioned from original document image is converted into 1D projection signal in both vertical and horizontal direction. 1D wavelet decomposition is implemented on the both projections. Then, wavelet statistic features are calculated for both projections and merged tistic feature vector is evaluated by BP neural as feature vector of row-document. Effectiveness of wavelet sta network. The experimental results show that the identification accuracy average around 94o/00.

关 键 词:文种识别 行块级 小波分析 神经网络 

分 类 号:TP391.1[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象