边缘缺失的畸变图像校正算法研究  

Research on the Correction Algorithm of Edge Missing Distortion Image

作　　者：魏剑飞 张荣福[1] 张仁杰[1] WEI Jian-fei;ZHANG Rong-fu;ZHANG Ren-jie(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《软件导刊》2020年第9期215-219,共5页Software Guide

摘　　要：目前,尚未有对边缘缺失畸变图像进行校正的有效方法。四边缘确定的畸变图像通过透视变换矫正,先确定边缘缺失图像的4个顶点,再通过透视变换实施校正。对边缘缺失的多边缘线图像,使用Hough变换的边缘检测及桥接膨胀,使主要边缘线延伸相交成4个点的封闭图形以确定四顶点。此外,拍摄图片有时背景复杂,对有效信息所在的四边形区域辨认带来一定困难。因此,在整幅图像中先获取所有连通区域,建立所有连通区域最小外接矩形,计算各外接矩形整幅占比,模糊比较后确定图像所在目标区域。在得出目标区域4个顶点后,采用透视变换进行畸变校正,再进行有效信息提取识别。在Matlab实验中,识别矫正成功率达93%,说明该方法准确率较高,适用于多边缘缺失图像矫正。There is no cost-effective method to implement the correction of edge missing distorted image. Distorted image and four-edge determination are corrected by perspective transformation. The four vertices of edge missing image are determined first,and then the correction is implemented by perspective transformation. For the image of multiple edge lines with missing edge,Hough transform’s edge detection and bridge expansion make the main edge lines extend and intersect into a closed graph of four points to determine the four vertices. In addition,sometimes the background of the image is complex,which brings some difficulties to the recognition of effective information in the quadrilateral region. Therefore,in this study,all connected regions are first obtained in the whole image,and the minimum enclosing rectangles of all connected regions are established. Then,the proportion of each enclosing rectangle is calculated,and the target region of the image is determined after fuzzy comparison. After the four vertices of the target region are obtained,the perspective transformation is used to correct the distortion and then the effective information is extracted and recognized.In Matlab,our experiment can achieve the recognition and correction success rate of 93%. Therefore,this method has high accuracy and is suitable for correction of multi-edge missing images.

关 键 词：边缘缺失畸变图像 HOUGH变换 顶点确定 目标鉴别 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]