| 基于小波和形态学的图像文本定位方法． |
| |
| 引用本文： | 宋文,陈国龙.基于小波和形态学的图像文本定位方法．[J].宿州学院学报,2013,28(2):76-78. |
| |
| 作者姓名： | 宋文 陈国龙 |
| |
| 作者单位： | 1. 淮北师范大学计算机科学与技术学院,安徽淮北,235000
2. 宿州学院信息工程学院,安徽宿州,234000 |
| |
| 基金项目： | 安徽高校省级自然科学研究项目"多级过滤器组合的图像垃圾邮件检测关键技术研究" |
| |
| 摘 要： | 为解决图像中文本定位问题，提出一种基于小波和形态学的图像文本定位方法。该方法首先采用二尺度的小波变换对可能是文本的区域进行粗检测；然后再通过形态学的相关技术去除图像中一些小的虚假的非文本区域；最后对提取的文本区域中的连通区域进行外接矩形标定，从而实现对文本区域的细定位。实验结果表明，该算法具有较好的实时性、精确性和鲁棒性。
|
| 关 键 词： | 小波分解 小波重构 形态学 文本定位 |
Image Text Location Method Based on Wavelet and Morphology |
| |
| SONG Wen
,
CHEN Guo-long.Image Text Location Method Based on Wavelet and Morphology[J].Journal of Shuzhou College,2013,28(2):76-78. |
| |
| Authors: | SONG Wen CHEN Guo-long |
| |
| Institution: | 1. School of Computer Science and Technology,Huaibei Normal University,Huaibei Anhui, 235000 ; 2. School of Information Engineering,Suzhou University,Suzhou Anhui,234000,China) |
| |
| Abstract: | In order to solve the problem of text location in images, the article presents an image text location method based on wavelet and morphology. The method first uses two-scale wavelet transformation to do a coarse detection to potential text regions ;and then it removes some small false text regions in images using technologies related to morphology ; finally it carries out the external rectangular calibration to the connected component in the extracted text regions,so as to realize the fine location of the text regions. The results of experiments show that there are good real-time performance, accuracy and robustness in the algorithm. |
| |
| Keywords: | Wavelet decomposition Wavelet reconstruction morphology text location |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
