基于LHBP多尺度向性滤波的文字检测算法

许鹏飞; 姚鸿勋; 纪荣嵘; 王积成; 孙晓帅

doi:10.11834/jig.20100806

图像分析和识别 | 浏览量 : 0 下载量: 253 CSCD: 0

PDF
导出
分享
收藏
专辑

基于LHBP多尺度向性滤波的文字检测算法
A Text Detection Algorithm Based on Local Haar Binary Pattern with Multi-scale Directional Filtering
2010年15卷第8期页码：1207
网络出版：2010-08-25，

纸质出版：2010
DOI： 10.11834/jig.20100806
稿件说明：

移动端阅览

许鹏飞, 姚鸿勋, 纪荣嵘, 王积成, 孙晓帅. 基于LHBP多尺度向性滤波的文字检测算法[J]. 中国图象图形学报, 2010,15(8):1207. DOI： 10.11834/jig.20100806.

XU Pengfei, YAO Hongxun, JI Rongrong, WANG Jicheng, SUN Xiaoshuai. A Text Detection Algorithm Based on Local Haar Binary Pattern with Multi-scale Directional Filtering[J]. Journal of Image and Graphics, 2010, 15(8): 1207. DOI： 10.11834/jig.20100806.

摘要

复杂光照条件和文字—背景的交融是自然场景图像中文字检测的主要难点。为解决该问题，提出了基于LHBP(local Haar binary pattern)多尺度向性滤波的文字检测算法。该算法首先采用对光强变化不敏感并具文字特征显式描述特点的LHBP模式的纹理描述算子；并在LHBP模式上采用多尺度向性滤波器MDF(multi-scale directional filtering)来确定候选文字区域；最后使用基于LHBP直方图的支持向量机法精确定位文字区域。实验结果表明，与其他主流算法相比，该算法能够去除复杂光照条件和文字—背景交融的影响，具有更好的性能。

Abstract

The main difficulties for text detection are due to complicated illumination condition and text-background weak contrast in natural scene images. In order to resolve them

this paper presents a local Haar binary pattern(LHBP) based algorithm for text detection. Firstly

LHBP texture description operator is presented. It is insensitive to illumination variance and can effectively describe the text feature. Secondly

the Multi-scale Directional Filtering based on LHBP is proposed for fast filtering to obtain candidate text regions. Finally

the LHBP-histogram-based SVM is presented to refine the text location. Comparing with state-of-the-art algorithms

the experiment results demonstrate the robustness of the proposed method with better accuracy.

关键词

Keywords

references

文章被引用时，请邮件提醒。

提交