| 视频图像中文本的检测、定位与提取 |
| |
| 引用本文： | 孙李娜,樊来耀.视频图像中文本的检测、定位与提取[J].电子科技,2009,22(1). |
| |
| 作者姓名： | 孙李娜 樊来耀 |
| |
| 作者单位： | 西安电子科技大学,电子工程学院,陕西,西安,710071 |
| |
| 摘 要： | 视频数据中的文本是视频语义理解和检索的重要信息来源.文中对视频中文本的检测、定位、提取、增强和识别进行了研究.提出了应用小波模极大值算法检测视频帧文本所在的位置,用由粗到精的多层定位方法以及金字塔模型,对于多尺度的静止和滚动中英文文字进行提取,最后对文本区域进行二值化.实验表明文中方法取得了良好的效果.
|
| 关 键 词： | 视频检索 高层语义 文本提取 小波模极大值 文本定位 |
Text Detection,Localization and Extraction from Video Images |
| |
| Sun Lina,Fan Laiyao.Text Detection,Localization and Extraction from Video Images[J].Electronic Science and Technology,2009,22(1). |
| |
| Authors: | Sun Lina Fan Laiyao |
| |
| Affiliation: | School of Electronic Engineering;Xidian University;Xi'an 710071;China |
| |
| Abstract: | Texts in video bring important clues for video indexing and retrieval.This paper studies the detection,localization,extraction,enhancement and recognition of video texts.It proposes that the wavelet modulus maximum algorithm be used to detect video texts.The coarse-to-fine localization scheme and the pyramid model are preformed to identify the multi-scale Chinese and English texts both static and scrolling.Finally the text region is binarized.The experiment results show the effectiveness of the method. |
| |
| Keywords: | video retrieval high-level semantic text extraction wavelet modulus maximum text localization |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
