[1]陈德运,尹芳,吴锐.基于消失点的场景文本透视变形校正方法[J].南京理工大学学报(自然科学版),2011,(04):436-441.
 CHEN De-yun,YIN Fang,WU Rui.Perspective Distortion Correction Method of Scene Text Based on Vanishing Point[J].Journal of Nanjing University of Science and Technology,2011,(04):436-441.
点击复制

基于消失点的场景文本透视变形校正方法
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
期数:
2011年04期
页码:
436-441
栏目:
出版日期:
2011-08-31

文章信息/Info

Title:
Perspective Distortion Correction Method of Scene Text Based on Vanishing Point
作者:
陈德运1尹芳1吴锐2
1. 哈尔滨理工大学计算机科学与技术学院,哈尔滨150080; 2. 哈尔滨工业大学计算机科学与技术学院,哈尔滨150001
Author(s):
CHEN De-yun1YIN Fang1WU Rui2
1. School of Computer Science and Technology,Harbin University of Science and Technology,Harbin 150080, China; 2. School of Computer Science and Technology,Harbin Institute of Technology,Harbin 150001,China
关键词:
场景文本 透视变形 变形校正 消失点
Keywords:
scene texts perspective distortion distortion correction vanishing points
分类号:
TP391. 4
摘要:
针对自然场景文本识别过程中文本图像的透视变形问题,提出一种有效的基于消失点 的场景文本透视变形校正方法。该方法首先利用由文本字符边缘端点拟合得到的文本基线检 测消失点,然后利用消失点信息估计单应矩阵部分参数,使用单应矩阵实现图像校正,得到正视 的文本图像。与其他方法相比,该方法受文本版式的限制更少。实验结果表明,该方法能有效 校正文字行数较少的场景文本,明显提高字符识别率。
Abstract:
An effective perspective distortion correction method is presented to resolve the perspective distortion correction in text recognition of natural scene. A vanishing point is detected by text baseline obtained from the tip point of text character. A vanishing point position is used to get some deformation parameters of homography matrix,which is utilized to correct the image. The front view image is achieved using a homography matrix. Compared with other methods, this method is less restricted by text layout. Experimental results show this method can effectively correct the text image with a few lines in natural scene and increase recognition rate significantly.

参考文献/References:

[1] Lienhart R,Wernicke A. Localizing and segmenting text in images and videos[J]. IEEE Transactions on Circuits and Systems for Video Technology,2002,12 ( 4) : 256-268.
[2] Clark P,Mirmehdi M. Rectifying perspective views of text in 3D scenes using vanishing points[J]. Pattern Recognition, 2003, 36: 2673-2686.
[3] Dance C R. Perspective estimation for document images [A]. Proceedings of SPIE [C]. San Jose,United States: SPIE, 2002: 244-254.
[4] Lu Shijian,Chen B M,Ko C C. Perspective rectification of document images using fuzzy set and morphological operations[J]. Image and Vision Computing, 2005, 23: 541-553.
[5] Hartley R,Zisserman A. Multiple view geometry in computer vision[M]. 2nd. Cambridge: Cambridge University Press, 2003: 87-129.
[6] ZhangWangbo,Li Xiaoyuan,Ma Xingjie. Perspective correction method for Chinese document image[A]. IITA 2008 Workshop [C]. Shanghai,China: IEEE Computer Society, 2008: 467-470.
[7] Tong Lijing,Zhang Yan. Correction of perspective text image based on gradient method[A]. ICINA[C]. Kunming, China: IEEE Computer Society, 2010: 312-316.
[8] Liang Jian. Processing camera-captured document images: Geometrics rectification,mosaicing,and layout structure recognition[D]. College Park: Maryland University, 2006.
[9] Fischler M A,Bolles R C. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography[J]. CACM, 1981, 24( 6) : 381-395.
[10] 陈付幸,王润生. 基于预检验的快速随机抽样一致 性算法[J]. 软件学报. 2005, 16( 8) : 1431-1437.
[11] 吴锐,刘家峰,唐降龙,等. 基于Gabor 小波变换的 汉字识别方法[J]. 高技术通讯, 2005, 15( 3) : 7-10.

备注/Memo

备注/Memo:
基金项目: 国家自然科学基金( 60672090) ; 黑龙江省自然科学基金( QC2009C35) ; 黑龙江省教育厅科学技术研究 项目( 12511098) 作者简介: 陈德运( 1962-) ,男,博士,教授,博士生导师,主要研究方向: 探测与成像技术、图像处理,E-mail: chendeyun@ hrbust. edu. cn。
更新日期/Last Update: 2012-10-23