[1]李志杰,袁鹏泰,李博涵.基于透视变换的手机身份证字符分割算法[J].计算机技术与发展,2018,28(07):58-62.[doi:10.3969/ j. issn.1673-629X.2018.07.013]
 LI Zhi-jie,YUAN Peng-tai,LI Bo-han.A Segmentation Algorithm of Identification Card Characters of Mobile Phone Based on Perspective Transform[J].,2018,28(07):58-62.[doi:10.3969/ j. issn.1673-629X.2018.07.013]
点击复制

基于透视变换的手机身份证字符分割算法()
分享到:

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年07期
页码:
58-62
栏目:
智能、算法、系统工程
出版日期:
2018-07-10

文章信息/Info

Title:
A Segmentation Algorithm of Identification Card Characters of Mobile Phone Based on Perspective Transform
文章编号:
1673-629X(2018)07-0058-05
作者:
李志杰1 袁鹏泰1 李博涵2
1. 南京航空航天大学 计算机科学与技术学院,江苏 南京 211106;
2. 江苏易图地理信息科技股份有限公司,江苏 扬州 225009
Author(s):
LI Zhi-jie1 YUAN Peng-tai1 LI Bo-han2
1. School of Computer Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China;
2. Jiangsu Province Easy Map Geographic Information Technology Company Limited by Shares,Yangzhou 225009,China
关键词:
Android 手机复杂背景身份证字符分割Canny 边缘检测Hough 变换透视变换投影法
Keywords:
Android mobile phonecomplex backgroundsegmentation of ID card charactersCanny edge detectionHough transformperspective transformprojection method
分类号:
TP301.6
DOI:
10.3969/ j. issn.1673-629X.2018.07.013
文献标志码:
A
摘要:
在 Android 手机下,提出一种改进的从复杂背景中分割出无畸变身份证字符图像的技术。 对于手机扫描得到的满足要求的图像进行预处理,包括:灰度化、缩放、中值滤波、自适应阈值 Canny 边缘检测;在此基础上使用概率 Hough 变换得到边缘检测图中的直线。 为了避免同一条直线出现断裂的情况,需要对 Hough 变换后的图像进行同一条直线的连接处理。 在连接后的直线中,找出身份证外围的四条轮廓线并求出四条线的交点,利用这四个交点和透视变换分割出无畸变、无背景的目标身份证图像。 根据所需信息区域在身份证上位置固定,利用比例坐标,在之前分割出的无背景身份证图像上,分割出含有所需信息的矩形区域图像,最终用投影法做行分割和单字符分割。 实验结果表明,该方法具有较高的分割速度与正确率。
Abstract:
We put forward an improved technique to segment the image of the undistorted ID card characters from the complex background in Android mobile phone. The image obtained by scanning and satisfying special conditions is preprocessed firstly including graying,scaling,median filtering and self-adaptive threshold Canny edge detection. Based on previous steps,the probabilistic Hough transform is used for line detection in edge detection image. To avoid a fracture in the same straight line,it is necessary to connect the same line of the image after the Hough transformation. In the straight line after the connection,we find out the outer four contours of the ID card and find the intersection point of the four lines. Using the four intersection points and the perspective transformation,we can segment the target ID card image without distortion and background. Because the required information is fixed on the ID card,the rectangular image of the area containing the required information can be segmented on the ID card image without background which is previously segmented by using the proportional coordinate. Finally,the projection method is used for row segmentation and single character segmentation. Experiment shows that this method has higher segmentation speed and correct rate.
更新日期/Last Update: 2018-08-27