摘要
提出了一种基于子图像的变异灰度直方图 (VCH)的检测与定位文本的方法 ,对子图像VCH的映射特征及其与图像文本的对应关系作了分析 ,并给出文本检测与定位的有效算法。该算法通过对行分割子图像VCH C中的凸台和列分割子图像VCH R中的凹谷的识别与定位 ,实现在图像中快速检测与定位文本。文章给出了计算试验的结果 。
It has been attracting wide attention to retrieval, analysis and understanding of huge video data. A number of content based methods for indexing and retrieving have been presented. Text detection and extraction in video images is an effective way for achieving image-content-based retrieval. This paper presents a method for detecting and locating text areas in images based on sub-image VCH. The mapping features of sub-image VCH and the relation between texts and mapping features is analyzed. And the algorithm for detecting and locating text areas in images is described. Some computing experiment results are given and analysis and evaluation are made for it in the end of the paper.
出处
《武汉大学学报(信息科学版)》
EI
CSCD
北大核心
2003年第3期354-358,共5页
Geomatics and Information Science of Wuhan University
基金
安徽省自然科学基金资助项目 ( [2 0 0 2 ] 0 4 0 30 42 2 0 7)
关键词
变异灰度直方图
子图像
基于内容的检索
文本检测与定位
variant color histogram (VCH)
sub-image
content-based retrieval
text detection and location