一种基于图像统计学的复杂表格交点聚类提取方法

本发明公开了一种基于图像统计学的复杂表格交点聚类提取方法。其采用的方法步骤为1：待分割表格图像预处理阶段：将多分辨率的图像数据进行压缩灰度化，使用OTSU进行二值化图像，为后续直线拟合阶段鉴定基础；2：横向直线方程拟合阶段：通过横向投影直方图统计，结合统计学计量得到横向长直线方程；3：纵向直线方程拟合阶段：通过纵向投影直方图统计，结合统计学计量得到纵向长直线方程；4：单元格分割处理阶段：通过直线方程的横纵交叉确定横纵坐标，实现单元格的分割并进行非文字区域的剔除。该方法经过实际测试具备较强的鲁棒性和自适应性，提高了表格交点检测准确率，在输入纵向有效聚类数目的前提下，交点检测率可达100％，具有一...

Full description

Saved in:
Bibliographic Details
Format Patent
LanguageChinese
Published 02.02.2024
Subjects
CALCULATING
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:本发明公开了一种基于图像统计学的复杂表格交点聚类提取方法。其采用的方法步骤为1：待分割表格图像预处理阶段：将多分辨率的图像数据进行压缩灰度化，使用OTSU进行二值化图像，为后续直线拟合阶段鉴定基础；2：横向直线方程拟合阶段：通过横向投影直方图统计，结合统计学计量得到横向长直线方程；3：纵向直线方程拟合阶段：通过纵向投影直方图统计，结合统计学计量得到纵向长直线方程；4：单元格分割处理阶段：通过直线方程的横纵交叉确定横纵坐标，实现单元格的分割并进行非文字区域的剔除。该方法经过实际测试具备较强的鲁棒性和自适应性，提高了表格交点检测准确率，在输入纵向有效聚类数目的前提下，交点检测率可达100％，具有一定的创新可拓展性和实际可行性。 The invention discloses a complex table intersection clustering extraction method based on image statistics. The method comprises the following steps: 1, performing a to-be-segmented table image preprocessing stage: carrying out compression graying on multi-resolution image data, and carrying out binarization on an image by using OTSU to serve as an identification basis of a subsequent straight line fitting stage; 2, performing a transverse linear equation fitting stage: obtaining a transverse long linear equation through transverse projection histogram statistics in combination with statistical measurement; 3, performing a longitudinal linear equation fitting stage: obtaining a longitudinal long li
Bibliography:Application Number: CN202010564503