基于改进概率霍夫变换的表格字符定位算法  

Table Character Location Algorithm Based on Improved Probabilistic Hough Transform

作　　者：张倩[1] 陈其浩 杨硕[1] ZHANG Qian;CHEN Qi-Hao;YANG Shuo(School of Surveying,Mapping and Spatial Information,Shandong University of Science and Technology,Qingdao 266590,China)

机构地区：[1]山东科技大学测绘与空间信息学院,山东青岛266590

出　　处：《软件导刊》2021年第11期204-209,共6页Software Guide

摘　　要：现有表格字符定位算法在扫描高质量的图像时可以实现精确定位效果,但对低质量的表格图像字符定位精度较差。为实现低质量表格图像中字符的精确定位,提出一种基于改进概率霍夫变换的表格字符定位算法。该算法首先对图像进行预处理以提取目标信息;其次,利用改进概率霍夫变换算法检测表格线,并对其进行倾斜校正;最后,根据表格线位置对字符进行粗定位,再利用投影法对字符进行精确定位。该算法实现了字符的精确定位,相比于其他字符定位算法,该算法定位准确率有所提升,达99.3%。The existing table character positioning algorithms can achieve the effect of accurate positioning when scanning high-quali⁃ty images,but the character positioning accuracy of low-quality table images is poor.In order to realize the accurate location of charac⁃ters in low-quality table images,a table character location algorithm based on improved probability Hough transform is proposed in this paper.Firstly,the image is preprocessed to extract the target information;Secondly,the improved probabilistic Hough transform algorithm is used to detect the table line and correct its tilt;Finally,the characters are roughly located according to the table line posi⁃tion,and then the characters are accurately located by projection method.The algorithm realized the accurate location of the character.Compared with other character location algorithms,the location accuracy of this algorithm is improved to 99.3%.

关 键 词：概率霍夫变换 投影法 直线检测 字符定位 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]