摘要
随着电子影象文档应用的日益广泛,相关文档图象的自动处理日益成为人们研究的热点。本文针对电子表格的特点,对从表格中自动提取填充内容这一问题进行了研究。系统首先在模版库中对输入表格图象进行自动匹配;然后对表格图象与相应的模版图象进行自动配准,从表格图象中将模版信息去除,并对误删除信息进行修补;最终,提取出表格的填充内容。通过对系统的实际测试,实验结果表明该方案是很有效的。
With the widely usage of the electronic files, the automatic processing of the corresponding documents has become a hot topic. In this paper, we consider the features of the files and have a research on the question. First, we get the template automatically of the input electronic file-image from the template base, and have a registration between the two images. Secondly, we delete the template data from the input image, and retain the mis-deleted information. Finally, we get the filled information of the original form. The experimented result shows that the method is effective and practicable.
出处
《系统仿真学报》
EI
CAS
CSCD
2004年第11期2611-2613,共3页
Journal of System Simulation
基金
国家自然科学基金项目(60173035
60373070)
国家"863"(项目2003AA411310)资助。
关键词
表格处理
倾斜校正
版面识别
格式去除
form processing
skew emendation
template recognition
format removal