计算机科学 ›› 2016, Vol. 43 ›› Issue (4): 313-317.doi: 10.11896/j.issn.1002-137X.2016.04.064

• 图形图像与模式识别 • 上一篇    下一篇

结合显著性检测和中心分割算法的文本检测方法

许肖,顾磊   

  1. 南京邮电大学计算机学院 南京210003,南京邮电大学计算机学院 南京210003
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受国家自然科学基金项目(61302157)资助

Saliency Text Detection Combining Graph-based Manifold Ranking with C entral Segmentation

XU Xiao and GU Lei   

  • Online:2018-12-01 Published:2018-12-01

摘要: 针对复杂背景下的文本检测问题,提出了显著性检测与中心分割算法相结合的文本检测技术。对于输入的图像,首先分别使用前景与背景作为标准的显著性检测方法,背景检测时将图像的四边分别作为基准,前景检测时将背景检测中得到的非背景区域作为基准,最终可得到较准确的备选文本区。然后使用中心分割算法,得到精确的边缘图。由于显著性图备选区域准确边缘细节缺失,而边缘图边缘精确但无法得出备选文本区,因此将两者进行融合处理,得到最终文本区域。实验表明,所提出的方法有较好的检测效果。

关键词: 文本检测,显著性检测,中心分割

Abstract: To detect text from images with different backgrounds,a new algorithm based on saliency detection via graph-based manifold ranking and central segmentation was proposed. The image elements (pixels or regions) of input image are ranked by similarity with foreground cues or background cues at first.The boundary prior is exploited by using the nodes on each side of image as labelled background queries,and then binary segmentation is applied on the achieved saliency map and the labelled foreground nodes are taken as salient queries.The accurate alternative text area can be obtained at last.Then the central segmentation algorithm is used to obtain the precise edge graph.As the saliency map cannot locate the boundary and the edge graph cannot locate the text areas,we integrated both of them and got the final result.Experiment shows that the proposed method can effectively detect the text in the image.

Key words: Text detection,Saliency detection,Central segmentation

[1] Liu Chun-mei,Wang Chun-heng,Dai Ru-wei.Text detection in images based on unsupervised classification of edge-based features [C]∥Document Analysis and Recognition.2005:610-614
[2] Mariano V Y,Kasturi R.Locating uniform-colored text in video frames [C]∥15th International Conference on Pattern Recognition.2000:539-542
[3] Cai M,Song J,Lyu M R.A new approach for video text detection [C]∥2002 International Conference on Image Processing.2002:117-120
[4] Yang Chuan,Zhang Li-he.Saliency Detection via Graph-BasedManifold Ranking[C]∥IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2013).2013:3166-3173
[5] Grady L,Jolly M,Seitz A.Segmentation from a box[C]∥IEEE International Conference on Computer Vision.2011:367-374
[6] Lempitsky V,Kohli P,Rother C,et al.Image segmentation with a bounding box prior[C]∥IEEE International Conference on Computer Vision.2009:277-284
[7] Zhou D,Weston J,Gretton A,et al.Ranking on data manifolds[M]∥Advances in Neural Information Processing Systems,2003:169-176
[8] Ng A,Jordan M,Weiss Y,et al.On spectral clustering:Analysisand an algorithm[C]∥Neural Information Processing Systems Foundation.2002:849-856
[9] Scholkopf B,Platt J,Shawe-Taylor J,et al.Estimating the support of a high-dimensional distribution[J].Neural Computation,2001,13(7):1443-1471
[10] Achanta R,Smith K,Lucchi A,et al.SLIC superpixels:149300[R].EPFL,2010
[11] Borji A,Sihite D,Itti L.Salient object detection:A benchmark[C]∥European Conference on Computer Vision.2012:414-429
[12] Wang H,Oliensis J.Rigid Shape Matchingby Segmentation Averaging[J].IEEE Transactions on Pattern Analsysi and Machine Intelligence,2009,32(4):619-635
[13] Boykov Y,Veksler O,Zabih R.Fast Approximate EnergyMinimization via Graph Cuts[J].IEEE Trans.Pattern Analysis andMachine Intelligence,2001,23(11):1222-1239
[14] Lucas S M.ICDAR 2005 text locating competition results [C]∥Document Analysis and Recognition,2005:1-5
[15] Shivakumara P,Phan T Q.A laplacian approach to multi-oriented text detection in video [J].IEEE Transactions on Software Engineering,2010,33(2):412-419
[16] Chen D,Odobez J M,Thiran J P.A localization/verificationscheme for finding text in images and video frames based on contrast independent features and machine learning [J].Signal Processing Image Communication,2004,19(3):205-217
[17] Wong E K,Chen M.A New Robust Algorithm for Video Text Extraction[J].Pattern Recognition,2003,36(6):1397-1406

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!