计算机科学 ›› 2017, Vol. 44 ›› Issue (9): 300-303.doi: 10.11896/j.issn.1002-137X.2017.09.056

• 图形图像与模式识别 • 上一篇    下一篇

基于边缘检测和特征融合的自然场景文本定位

王梦迪,张友梅,常发亮   

  1. 山东大学控制科学与工程学院 济南250061,山东大学控制科学与工程学院 济南250061,山东大学控制科学与工程学院 济南250061
  • 出版日期:2018-11-13 发布日期:2018-11-13
  • 基金资助:
    本文受国家自然科学基金项目(61673244),高等学校博士学科点专项科研基金资助

Text Localization Based on Edge Detection and Features Fusion in Natural Scene

WANG Meng-di, ZHANG You-mei and CHANG Fa-liang   

  • Online:2018-11-13 Published:2018-11-13

摘要: 文本定位作为文本识别的基础和前提,对图像深层信息的理解至关重要。针对自然场景下的文本定位受光照、复杂背景等因素影响较大的问题,提出了一种基于多方向边缘检测和自适应特征融合的自然场景文本定位方法。该方法首先将自然场景图像进行三通道八方向的边缘检测;然后 通过启发式规则 对得到的边缘图像进行过滤从而提取出备选文本域,进而对备选文本域进行自适应权值的HOG-LBP特征提取与融合;最后采用支持向量机进行特征分类学习,实现文本定位。实验结果表明,该方法能准确定位自然场景图片的文本区域,对光照和复杂背景具有较强的鲁棒性。

关键词: 自然场景,文本定位,边缘检测,特征融合

Abstract: As the basis and premise of text recognition,text localization has an important influence on the analysis of images.Since the text localization in natural scene can be effected by illumination and the complex backgrounds significantly,we proposed a text localization method based on edge detection and features fusion.The method began with edge detection from three channels and eight directions,and then we filtered the detected edge images with heuristic rules to extract candidate text regions.On top of that,the HOG-LBP features were extracted and fused by adaptive weights.Finally,we applied support vector machine (SVM) to classify the candidate regions and realized text localization.Experimental results indicate that the proposed method can locate the text region accurately in natural scene images while reducing the influence of illumination and complex backgrounds effectively.

Key words: Natural scene,Text localization,Edge detection,Feature fusion

[1] YE Q X,DOERMANN D.Text Detection and Recognition in Imagery:A Survey[J].IEEE Transactions on Pattern Analysis &Machine Intelligence,2015,7(7):1480-1500.
[2] YANG H J,QUEHL B,SACK H.Text detection in video images using adaptive edge detection and Stroke Width verification[C]∥19th International Conference on Systems,Signals andImage Processing (IWSSIP).IEEE,2012:9-12.
[3] RAJESHABABA M,ANITHA T.Detect and separate localization text in various complicated-colour image[C]∥International Conference on Circuits,Power and Computing Technologies.IEEE,2013:866-872.
[4] MORADI M,MOZAFFARI S.Hybrid approach for Farsi/Arabic text detection and localisation in video frames[J].Iet Image Processing,2013,7(2):154-164.
[5] YI C C,TIAN Y L.Text string detection from natural scenes bystructure-based partition and grouping[J].IEEE Transactions on Image Processing,2011,0(9):2594-2605.
[6] FENG Y Y,SONG Y H,ZHANG Y L.Scene text localization using extremal regions and Corner-HOG feature[C]∥IEEE International Conference on Robotics and Biomimetics.IEEE,2015:881-886.
[7] BHARDWAJ D,PANKAJAKSHAN V.Image Overlay Text Detec-tion Based on JPEG Truncation Error Analysis[J].IEEE Signal Processing Letters,2016,3(8):1027-1031.
[8] LYU M R,SONG J C,CAI M.A comprehensive method for multilingual video text detection,localization,and extraction[J].IEEE Transactions on Circuits and Systems for Video Technology,2005,5(2):243-255.
[9] SRIVASTAV A,KUMAR J.Text detection in scene imagesusing stroke width and nearest-neighbor constraints[C]∥2008 IEEE Region 10 Conference.IEEE,2008:1-5.
[10] YE J,HUANG L L,HAO X L.Neural network based text detection in videos using local binary patterns[C]∥Chinese Conference on Pattern Recognition(CCPR).IEEE,2009:1-5.
[11] MAO W G,CHUNG F L,LAM K K M,et al.Hybrid Chinese/English text detection in images and video frames[C]∥16th International Conference on Pattern Recognition.IEEE,2002,16(3):1015-1018.
[12] ZINI L,DESTRERO A,ODONE F.A classification architecture based on connected components for text detection in unconstrained environments[C]∥6th IEEE International Conference on Advanced Video and Signal Based Surveillance.IEEE,2009:176-181.
[13] SU F,XU H L.Robust seed-based stroke width transform for text detection in natural images[C]∥13th International Conference on Document Analysis and Recognition(ICDAR).IEEE,2015:916-920.
[14] KUTTY S B,SAAIDIN S,YUNUS P N A M,et al.Evaluation of canny and sobel operator for logo edge detection[C]∥International Symposium on Technology Management and Emerging Technologies.IEEE,2014:153-156.
[15] KAUR B,GARG A.Mathematical morphological edge detection for remote sensing images[C]∥3rd International Conference on Electronics Computer Technology(ICECT).IEEE,2011:324-327.
[16] DALAL N,TRIGGS B.Histograms of oriented gradients forhuman detection[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2005:886-893.
[17] HUANG F F.Research on face recognition based on LBP operator[D].Chongqing:Chongqing University,2009.(in Chinese) 黄非非.基于LBP的人脸识别研究[D].重庆:重庆大学,2009.
[18] OJALA T,PIETIKAINEN M,MAENPAA T.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,4(7):971-987.
[19] LIU Y Y,YU F Q,CHEN Y.Text location in image based on connected-component and statistical features[J].Computer Engineering and Applications,2016,2(5):165-68.(in Chinese) 刘亚亚,于凤芹,陈莹.基于连通区域和统计特征的图像文本定位[J].计算机工程与应用,2016,2(5):165-168 .
[20] LUCAS S M.ICDAR 2005 text locating competition results[C]∥8th International Conference on Document Analysis and Recognition(ICDAR’05).IEEE,2005:80-84.
[21] YI C C,TIAN Y L.Text detection in natural scene images by stroke gabor words[C]∥International Conference on Document Analysis and Recognition(ICDAR).IEEE,2011:177-181.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!