[1]桂丹萍 陈佳祥 何红生.视频检索在汉字识别中的应用研究[J].计算机技术与发展,2010,(10):207-210.
　GUI Dan-ping,CHEN Jia-xiang,HE Hong-sheng.Application Research of Video Retrieval Model on Chinese Character Recognition[J].,2010,(10):207-210.
点击复制

视频检索在汉字识别中的应用研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2010年10期
页码:
207-210
栏目:
应用开发研究
出版日期:
1900-01-01

文章信息/Info

Title:
Application Research of Video Retrieval Model on Chinese Character Recognition
文章编号:
1673-629X（2010）10-0207-04
作者:
桂丹萍 陈佳祥 何红生
集美大学
Author(s):
GUI Dan-ping; CHEN Jia-xiang; HE Hong-sheng
Jimei University
关键词:
汉字识别; 视频检索模型; SIFT特征; KMEANS; TFIDF
Keywords:
chinese character recognition;  video retrieval model;  SIFT feature;  KMEANS;  TFIDF
分类号:
TP391.43
文献标志码:
A
摘要:
传统的OCR技术在汉字识别领域趋于成熟,对背景清晰的正体汉字有很高的识别正确率,然而当汉字图片在复杂背景中或经旋转、加噪处理后,OCR软件的识别正确率大大下降。当今有关视频检索的研究正在快速发展中,其中一种行之有效的方法是通过提取模板视频的关键帧及其特征向量,应用聚类算法形成关键字,并通过快速的检索算法来实现匹配。创新性地将该模型应用到汉字识别研究中,通过大量实验数据的研究发现,该模型在上述情况中相对于传统的OCR技术优势明显,在未来实际应用中具有广阔的前景
Abstract:
Traditional OCR has achieved a degree of maturity in the field of Chinese character recognition,which obtains a high recognition accuracy on Chinese character with a clean background and no rotation.However,when images are preprocessed in a complex background with low quality like affine transform and addition of noise,its recognition accuracy declined significantly.The current research on video retrieval is growing rapidly,where an effective method is to extract key frames from the video template and their feature vectors,apply clustering algorithm to form and retrieve the target video through a fast search algorithm.Innovatively apply the model to the study of Chinese character recognition.Through a large number of experimental data; this model outperforms traditional OCR under such variances.Therefore; this model enjoys a good prospect of application in the future

相似文献/References:

[1]宓浩 张燕平.基于特征投影和交叉覆盖神经网络的车牌识别[J].计算机技术与发展,2007,(10):76.
　MI Hao,ZHANG Yan-ping.License Plate Location Based on Projection Character and Alternate Covering Neural Network[J].,2007,(10):76.
[2]王建平 陈军 徐晓冰 王熹徽.基于SVM的脱机手写汉字机器学习识别方法研究[J].计算机技术与发展,2006,(10):104.
　WANG Jian-ping,CHEN Jun,XU Xiao-bing,et al.Research on Method of Off- Line Handwritten Chinese Characters Recognition Based on SVM[J].,2006,(10):104.
[3]武子毅,刘亮亮,张再跃.基于集成注意力层卷积神经网络的汉字识别[J].计算机技术与发展,2018,28(08):100.[doi:10.3969/ j. issn.1673-629X.2018.08.021]
　WU Zi-yi,LIU Liang-liang,ZHANG Zai-yue.Chinese Character Recognition of Convolutional Neural Network of Integration Attention Layer[J].,2018,28(10):100.[doi:10.3969/ j. issn.1673-629X.2018.08.021]

备注/Memo

备注/Memo:
福建省自然科学基金（2007J0202）桂丹萍（1983-）,女,硕士研究生,研究方向为非线性方程、图像处理
更新日期/Last Update: 1900-01-01