结合WTLBP特征和SVM的复杂场景文本定位方法

doi:10.3969/j.issn.1001-2400.2012.04.019

J4 ›› 2012, Vol. 39 ›› Issue (4): 103-108.doi: 10.3969/j.issn.1001-2400.2012.04.019

结合WTLBP特征和SVM的复杂场景文本定位方法

刘晓佩;卢朝阳;李静

(西安电子科技大学综合业务网理论及关键技术国家重点实验室，陕西西安 710071)

收稿日期:2011-06-26 出版日期:2012-08-20 发布日期:2012-10-08
通讯作者: 刘晓佩
作者简介:刘晓佩(1976-)，女，西安电子科技大学博士研究生，E-mail: liuxiaopei2007@163.com．
基金资助:
国家自然科学基金资助项目(60872141);中央高校基本科研业务费专项基金资助项目(K50510010007);陕西省自然科学基础研究计划资助项目(2009JQ8019)

Complex scene text location method based on WTLBP and SVM

LIU Xiaopei;LU Zhaoyang;LI Jing

(State Key Lab. of Integrated Service Networks, Xidian Univ., Xi'an 710071， China)

Received:2011-06-26 Online:2012-08-20 Published:2012-10-08
Contact: LIU Xiaopei

摘要/Abstract

摘要：

针对自然环境中文字结构差异性大，文字和非文字难以有效区分而造成定位虚警率较高的问题，提出利用小波变换(WT)和多尺度LBP算子相结合的方法(WTLBP)提取文字特征，并将其用于对候选文字区域的分类确认，以降低文本定位虚警率．算法首先利用文字笔画边缘实现快速的文本区域检测，获得候选的文字区域;然后，提取候选文字区域的WTLBP纹理特征，结合支持向量机(SVM)分类器对候选文字区域进行分类确认．实验结果表明， WTLBP文字特征具有较高的区分度，能够有效区分文字和非文字区域，将其用于对候选区域的分类确认可大大降低复杂场景中文本定位虚警率．

关键词: 文本定位, 复杂场景, WTLBP, 支持向量机

Abstract:

To reduce the false alarm rate in a complex scene text location, a text mode description method based on the combination of wavelet transform and LBP is proposed, which is used to classify candidate text regions as text or non-text regions. In the procedure of text location, firstly, the stroke edge is used to detect text regions, and the candidate text regions are obtained; secondly, WTLBP features of candidate text regions are extracted, and the support vector machine (SVM) classifier is employed to determine whether candidate text regions are text regions or not. Experimental results show that the proposed text feature is of high discrimination, and that using it in the procedure of verification can greatly reduce the false alarm rate of the scene text location.

Key words: text location, complex scene, WTLBP, support vector machine

刘晓佩;卢朝阳;李静. 结合WTLBP特征和SVM的复杂场景文本定位方法[J]. J4, 2012, 39(4): 103-108.

LIU Xiaopei;LU Zhaoyang;LI Jing. Complex scene text location method based on WTLBP and SVM[J]. J4, 2012, 39(4): 103-108.

参考文献

［1］ Zhang J, Kasturi R. Extraction of Text Objects in Video Documents: Recent Progress［C］//Proc of the 8th International Conference on Pattern Recognition. New York: IEEE Computer Society, 2008: 5-17.
［2］ Liang J, Doermann D, Li H P. Camera-based Analysis of Text and Documents: a Survey［J］. International Journal on Document Analysis and Recognition, 2005, 7(2-3): 84-104.
［3］ Zhang J, Goldgof D, Kasturi R. A New Edge Based Text Verification Approach for Video ［C］//Proc of the Eighth International Conference on Pattern Recognition. New York: IEEE Computer Society, 2008: 1-4.
［4］ Anthimopoulos M, Gatos B, Pratikakis I. A Two-stage Scheme for Text Detection in Video Images ［J］. Image and Vision Computing, 2010, 28(12): 1413-1426.
［5］ Pan Yifeng, Hou Xinwen, Liu Chenglin. Text Localization in Natural Scene Images based on Conditional Random Field［C］//Proc of the 10th International Conference on Document Analysis and Recognition. New York: IEEE Computer Society, 2009: 6-10.
［6］ Pan Yifeng, Hou Xinwen, Liu Chenglin. A Hybrid Approach to Detect and Localize Texts in Natural Scene Images ［J］. IEEE Trans on Image Processing, 2011, 20(3): 800-813.
［7］ Ye Qixiang, Jiao Jianbin, Huang Jun, et al. Text Detection and Restoration in Natural Scene Images［J］. Journal of Visual Communication ＆ Image Representation, 2007, 18(6): 504-513.
［8］ Ojala T, Pietikainen M. Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns［J］. IEEE Trans on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971-987.
［9］温浩, 卢朝阳, 高全学. 融合小波变换和张量PCA的人脸识别算法［J］. 西安电子科技大学报, 2009, 36(4): 602-607.
Wen Hao, Lu Zhaoyang, Gao Quanxue. Face Recognition Algorithm Based on Wavelet Preprocessing and Tensor PCA ［J］. Journal of Xidian University, 2009, 36(4): 602-607.
［10］ Lu Zhaoyang, Ando S, Kataoka K, et al. Text Locating and Verification Algorithm for Scene Images Based on Gray-Scale CC Analysis［J］. Portrait Electronics Society of Japan, 2007, 36(4): 509-519.
［11］ Chang C C, Lin C J. LIBSVM: a Library for Support Vector Machines［R/OL］. ［2010-5-20］. http://www.csie. ntu.edu.tw/～cjlin/libsvm.
［12］ Gllavata J, Ewerth R, Freisleben B. Text Detection in Images Based on Unsupervised Classification of High-Frequency Wavelet Coefficients［C］//Proc of the 17th International Conference on Pattern Recognition. New York: IEEE Computer Society, 2004: 425-428.
［13］ Yan Hao, Zhang Yi, Hou Zenggang, et al. Automatic Text Detection In Video Frames Based on Bootstrap Artificial Neural Network And CED［C］//Proc of the 11th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision. Plzen: Science Press, 2003: 1-6.
［14］ Lucas S M, Panaretos A, Sosa L, et al. ICDAR 2003 Robust Reading Competitions: Entries, Results, and Future Directions［J］. International Journal on Document Analysis and Recognition, 2005, 7(2): 105-122.

[1]	王俊祥,黄霖,张影,倪江群,林朗. 低复杂度的增强图像来源检测算法[J]. 西安电子科技大学学报, 2021, 48(1): 96-106.
[2]	李海,尚金雷,孙婷逸,冯青,庄子波. MC-DTSVMs的双偏振气象雷达降水粒子分类方法[J]. 西安电子科技大学学报, 2020, 47(4): 132-140.
[3]	安亚利,周水生,陈丽,王保军. 鲁棒支持向量机及其稀疏算法[J]. 西安电子科技大学学报, 2019, 46(1): 64-72.
[4]	李美玲;胡耀垓;周晨;赵正予;张援农;刘静;邓忠新. 支持向量机用于电离层foF2的短期区域预报[J]. J4, 2015, 42(5): 147-153+206.
[5]	蔡艳艳;宋晓东. 针对非平衡数据分类的新型模糊SVM模型[J]. J4, 2015, 42(5): 120-124+160.
[6]	王瑞;王康晏;冯玉田;张海燕;金彦亮;张有正. 复杂场景下声频传感器网络核稀疏表示车辆识别[J]. J4, 2015, 42(4): 114-120.
[7]	姜建国;赵媛;孟宏伟;李博. 采用类电磁机制算法的SVM决策树多分类策略[J]. J4, 2014, 41(6): 83-88.
[8]	杨曦;李洁;韩冰;高新波. 一种分层小波模型下的极光图像分类算法[J]. J4, 2013, 40(2): 18-24.
[9]	杨曦高新波李洁韩冰. 一种分层小波模型下的极光图像分类算法[J]. J4, 2013, 40(2): 18-24.
[10]	王世强;张登福;毕笃彦;薛德友;雍霄驹. 双谱二次特征在雷达信号识别中的应用[J]. J4, 2012, 39(2): 127-132+191.
[11]	秦传东;刘三阳;张市芳. 一种肿瘤基因的支持向量机提取方法[J]. J4, 2012, 39(1): 191-196.
[12]	厍向阳;李崇贵;姚顽强. 遥感图像几何校正的支持向量机算法研究[J]. J4, 2011, 38(5): 121-128+146.
[13]	杜亚勤；洪波；郭雷；杨宁. 一种曲面拟合图像边缘特征提取算法[J]. J4, 2011, 38(3): 164-168+188.
[14]	常甜甜;刘红卫;冯筠. 多源性数据SVM集成算法研究[J]. J4, 2010, 37(1): 136-141.
[15]	杜喆;刘三阳. 最小二乘支持向量机变型算法研究 [J]. J4, 2009, 36(2): 331-337.

结合WTLBP特征和SVM的复杂场景文本定位方法

Complex scene text location method based on WTLBP and SVM

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0