基于色彩空间的最大稳定极值区域的自然场景文本检测

doi:10.11772/j.issn.1001-9081.2017061389

计算机应用 ›› 2018, Vol. 38 ›› Issue (1): 264-269.DOI: 10.11772/j.issn.1001-9081.2017061389

• 虚拟现实与多媒体计算 • 上一篇下一篇

基于色彩空间的最大稳定极值区域的自然场景文本检测

范一华, 邓德祥, 颜佳

武汉大学电子信息学院, 武汉 430072

收稿日期:2017-06-07 修回日期:2017-08-05 出版日期:2018-01-10 发布日期:2018-01-22
通讯作者: 邓德祥
作者简介:范一华(1993-),女,河南新乡人,硕士研究生,主要研究方向:图像自然语言处理、字符识别;邓德祥(1961-),男,湖北荆州人,教授,硕士,主要研究方向:计算机视觉、目标跟踪;颜佳(1983-),男,湖北天门人,讲师,博士,主要研究方向:目标跟踪、图像质量评价。

Natural scene text detection based on maximally stable extremal region in color space

FAN Yihua, DENG Dexiang, YAN Jia

College of Electronic Information, Wuhan University, Wuhan Hubei 430072, China

Received:2017-06-07 Revised:2017-08-05 Online:2018-01-10 Published:2018-01-22

摘要/Abstract

摘要： 针对传统的最大稳定极值区域（MSER）方法无法很好地提取低对比度图像文本区域的问题，提出一种新的基于边缘增强的场景文本检测方法。首先，通过方向梯度值（HOG）有效地改进MSER方法，增强MSER方法对低对比度图像的鲁棒性，并在色彩空间分别求取最大稳定极值区域；其次，利用贝叶斯模型进行分类，主要采用笔画宽度、边缘梯度方向、拐角点三个平移旋转不变性特征剔除非字符区域；最后，利用字符的几何特性将字符整合成文本行，在公共数据集国际分析与文档识别（ICDAR）2003和ICDAR 2013评估了算法性能。实验结果表明，基于色彩空间的边缘增强的MSER方法能够解决背景复杂和不能从对比度低的场景图像中正确提取文本区域的问题。基于贝叶斯模型的分类方法在小样本的情况下能够更好地筛选字符，实现较高的召回率。相比传统的MSER进行文本检测的方法，所提方法提高了系统的检测率和实时性。

关键词: 文本检测, 边缘增强, 最大稳定极值区域, 颜色空间, 贝叶斯模型

Abstract: To solve the problem that the text regions can not be extracted well in low contrast images by traditional Maximally Stable Extremal Regions (MSER) method, a novel scene text detection method based on edge enhancement was proposed. Firstly, the MSER method was effectively improved by Histogram of Oriented Gradients (HOG), the robustness of MSER method was enhanced to low contrast images and MSER was applied in color space. Secondly, the Bayesian model was used for the classification of characters, three features with translation and rotation invariance including stroke width, edge gradient direction and inflexion point were used to delete non-character regions. Finally, the characters were grouped into text lines by geometric characteristics of characters. The proposed algorithm's performance on standard benchmarks, such as International Conference on Document Analysis and Recognition (ICDAR) 2003 and ICDAR 2013, was evaluated. The experimental results demonstrate that MSER based on edge enhancement in color space can correctly extract text regions from complex and low contrast images. The Bayesian model based classification method can detect characters from small sample set with high recall. Compared with traditional MSER based method of text detection, the proposed algorithm can improve the detection rate and real-time performance of the system.

Key words: text detection, edge enhancement, Maximally Stable Extremal Region (MSER), color space, Bayesian model

中图分类号:

TP391.1

范一华, 邓德祥, 颜佳. 基于色彩空间的最大稳定极值区域的自然场景文本检测[J]. 计算机应用, 2018, 38(1): 264-269.

FAN Yihua, DENG Dexiang, YAN Jia. Natural scene text detection based on maximally stable extremal region in color space[J]. Journal of Computer Applications, 2018, 38(1): 264-269.

参考文献

[1] CHEN X, YUILLE A L. Detecting and reading text in natural scenes[C]//CVPR 2004:Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2004:366-373.
[2] NEUMANN L, MATAS J. Scene text localization and recognition with oriented stroke detection[C]//Proceedings of the 2013 IEEE International Conference on Computer Vision. Piscataway, NJ:IEEE, 2013:97-104.
[3] CHEN H Z, TSAI S S, SCHROTH G, et al. Robust text detection in natural images with edge-enhanced maximally stable extremal regions[C]//ICIP 2011:Proceedings of the 2011 IEEE International Conference on Image Processing. Piscataway, NJ:IEEE, 2011:2609-2612.
[4] TURKI R, HALIMA M B, ALIMI A M. Scene text detection images with pyramid image and MSER enhanced[C]//ISDA 2015:Proceedings of the 2015 International Conference on Intelligent Systems Design and Applications. Piscataway, NJ:IEEE, 2015:301-306.
[5] AO C, BAI X, LIU W, et al. Detecting texts of arbitrary orientations in natural images[C]//CVPR 2012:Proceedings of the 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2012:1083-1090.
[6] MATAS J, CHUM O, URBAN M, et al. Robust wide baseline stereo from maximally stable extremal regions[J]. Image and Vision Computing, 2004, 22(10):761-767.
[7] EPSHTEIN B, OFEK E, WEXLER Y. Detecting text in natural scenes with stroke width transform[C]//CVPR 2010:Proceedings of the 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2010:2963-2970.
[8] YIN X C, YIN X, HUANG K, et al. Robust text detection in natural scene images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(5):970-983.
[9] YIN X, PEI W, ZHANG J, et al. Multi-orientation scene text detection with adaptive clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9):1930-1937.
[10] FORSSEN P E, LOWE D G. Shape descriptors for maximally stable extremal regions[C]//ICCV 2007:Proceedings of the 2007 IEEE International Conference on Computer Vision. Piscataway, NJ:IEEE, 2007:1-8.
[11] NEUMANN L, MATAS J. Real-time scene text localization and recognition[C]//CVPR 2012:Proceedings of the 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2012:3538-3545.
[12] 唐有宝,卜巍,邬向前.多层次MSER自然场景文本检测[J].浙江大学学报(工学版),2016,50(6):1134-1140.(TANG Y B, BU W, WU X Q. Natural scene text detection based on multi-level MSER[J]. Journal of Zhejiang University (Engineering Science), 2016, 50(6):1134-1140.)
[13] 李鉴鸿.基于MSER的图像文本定位的应用研究[D].广州:华南理工大学,2015:30-41.(LI J H. Application research on text location in image based on maximally stable extremal regions[D]. Guangzhou:South China University of Technology, 2015:30-41.)
[14] LUCAS S M, PANARETOS A, SOSA L, et al. ICDAR 2003 robust reading competitions[C]//ICDAR 2003:Proceedings of the 2003 International Conference on Document Analysis and Recognition. Berlin:Springer, 2003:682-687.
[15] KARATZAS D, SHAFAIT F, UCHIDA S, et al. ICDAR 2013 robust reading competition[C]//ICDAR 2013:Proceedings of the 2013 International Conference on Document Analysis and Recognition. Piscataway, NJ:IEEE, 2013:1484-1493.
[16] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//CVPR 2005:Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2005:886-893.
[17] LI Y, JIA W, SHEN C, et al. Characterness:an indicator of text in the wild[J]. IEEE Transactions on Image Processing, 2014, 23(4):1666-1677.
[18] REN X H, ZHOU Y, HE J H, et al. A convolutional neural network-based Chinese text detection algorithm via text structure modeling[J]. IEEE Transactions on Multimedia, 2017, 19(3):506-519.
[19] 张国和,黄凯,张斌,等.最大稳定极值区域与笔画宽度变换的自然场景文本提取方法[J].西安交通大学学报,2017,51(1):135-140.(ZHANG G H, HUANG K, ZHANG B, et al. A natural scene text extraction method based on the maximum stable extremal region and stroke width transform[J]. Journal of Xi'an Jiaotong University, 2017, 51(1):135-140.)
[20] 杨磊.复杂背景图像中文本检测与定位研究[D].广州:华南理工大学,2013:57-61.(YANG L. Research on text detection and location in complex background images[D]. Guangzhou:South China University of Technology, 2013:57-61.)
[21] NEUMANN L, MATAS J. A method for text localization and recognition in real-world images[C]//ACCV 2010:Proceedings of the 2010 IEEE Computer Asian Conference on Computer Vision. Piscataway, NJ:IEEE, 2010:770-783.

基于色彩空间的最大稳定极值区域的自然场景文本检测

Natural scene text detection based on maximally stable extremal region in color space

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	王一丁, 郝晨宇, 李耀利, 蔡少青, 袁媛. 基于深度学习的小样本中药材粉末显微图像识别[J]. 计算机应用, 2020, 40(5): 1301-1308.
[2]	李红, 王瑞尧, 耿则勋, 胡海峰. 基于多尺度梯度域引导滤波的低照度图像增强算法[J]. 计算机应用, 2019, 39(10): 3046-3052.
[3]	孙婷, 马磊. 巡检机器人中指针式仪表示数的自动识别方法[J]. 计算机应用, 2019, 39(1): 287-291.
[4]	任帅, 索丽, 张弢, 杨涛, 慕德俊. 基于3D高效率视频编码背景的信息隐藏算法[J]. 计算机应用, 2017, 37(6): 1616-1619.
[5]	张海艳, 高尚兵. 图像分割中改进空间约束贝叶斯网络模型的应用[J]. 计算机应用, 2017, 37(3): 823-826.
[6]	高智勇, 唐文峰, 贺良杰. 基于运动显著性的移动镜头下的运动目标检测[J]. 计算机应用, 2016, 36(6): 1692-1698.
[7]	陈伟, 陈继明. 基于贝叶斯模型的云服务服务质量预测[J]. 计算机应用, 2016, 36(4): 914-917.
[8]	张洛阳, 毛嘉莉, 刘斌, 吴涛. 基于贝叶斯模型的多标签分类算法[J]. 计算机应用, 2016, 36(1): 52-56.
[9]	邵良杉, 郭雅婵. 基于Codebook的视频火焰识别算法[J]. 计算机应用, 2015, 35(5): 1483-1487.
[10]	张永库, 李云峰, 孙劲光. 基于多特征融合的图像检索[J]. 计算机应用, 2015, 35(2): 495-498.
[11]	李鹏伟, 葛文英, 刘国英. 基于阴影概率模型的遥感影像阴影检测方法[J]. 计算机应用, 2015, 35(2): 510-514.
[12]	余艳. 融合KL散度和移地距离的高斯混合模型相似性度量方法[J]. 计算机应用, 2014, 34(3): 828-832.
[13]	盛帅曹丽萍黄增喜吴鹏飞. 基于改进稀疏编码的图像超分辨率算法[J]. 计算机应用, 2014, 34(2): 562-566.
[14]	伍杰华. 基于树状朴素贝叶斯模型的社会网络关系预测[J]. 计算机应用, 2013, 33(11): 3134-3137.
[15]	黄国伟许昱玮. 基于用户反馈的混合型垃圾邮件过滤方法[J]. 计算机应用, 2013, 33(07): 1861-1865.