Visual Saliency and Text Confidence Map Based Background Suppression for Scene Text

JIANG Wei; LU Zhao-yang; LI Jing; LIU Xiao-pei; YAO Chao

doi:10.3969/j.issn.0372-2112.2015.01.010

您当前的位置：

首页 >

文章列表页 >

Visual Saliency and Text Confidence Map Based Background Suppression for Scene Text

更新时间：2025-07-16

- Visual Saliency and Text Confidence Map Based Background Suppression for Scene Text
- Acta Electronica Sinica Vol. 43, Issue 1, Pages: 62-68(2015)
- 作者机构：
  
  西安电子科技大学综合业务网国家重点实验室,陕西,西安,710071
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.60872141);Fundamental Research Funds for the Central Universities (No.K50510010007);Huawei Innovation Research Project of Higher Education Institutes (No.IRP-2012-03-06)
- DOI：10.3969/j.issn.0372-2112.2015.01.010
  CLC： TP391
- Published：2015
- 稿件说明：
移动端阅览
JIANG Wei, LU Zhao-yang, LI Jing, et al. Visual Saliency and Text Confidence Map Based Background Suppression for Scene Text[J]. Acta Electronica Sinica, 2015, 43(1): 62-68.
DOI：

JIANG Wei, LU Zhao-yang, LI Jing, et al. Visual Saliency and Text Confidence Map Based Background Suppression for Scene Text[J]. Acta Electronica Sinica, 2015, 43(1): 62-68. DOI： 10.3969/j.issn.0372-2112.2015.01.010.

摘要

复杂背景是影响场景文字自动定位算法性能的重要因素.为消除背景干扰

本文首先根据谱残差理论获取显著性区域

然后利用基于笔画特征的点对梯度方向差比例图与基于傅里叶谱统计特征的后验概率图取得显著性区域的文字置信图

最后借助图模型融合视觉显著性图、文字置信图与HSI颜色特征最终达到抑制自然图像背景并突出前景文字的效果.在ICDAR2011场景文字定位竞赛数据库与实验室场景中文数据库中的实验结果表明

本算法可较好的抑制自然图像中复杂背景

并有效提升算法的性能.

Abstract

The complex background is an important inference factor of scene text automatic localization algorithm.To eliminate the impact of background

this scheme firstly obtains the saliency region in saliency map according to the spectral residual theory.Then it computes text confidence map in saliency region

which is weighted sum of the fraction map of corresponding pairs based on stroke feature and the posterior probability map based on Fourier spectrum statistical feature.Finally

it combines saliency map

text confidence map and HSI color feature with the graph model to suppress the background and highlight the foreground text.The experimental results on both the ICDAR2011 scene text localization competition test dataset and laboratory Chinese dataset show that the scheme can effectively suppress the complex background and improve the scene text localization algorithm.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Error Concealment for Frame Losses in MVD Video Using Saliency

Review of the Methods for Salient Object Detection Based on Deep Learning

Extracting Attention Information Algorithm Based onContrast Sensitivity and Markov Chain

Singular Value Decomposition Band-Pass-Filter for Image Background Suppression and Denoising

Infrared Background Suppression for Segmenting and Detecting Small Target

Related Author

CUI Jin-peng

ZHOU Yang

YIN Hai-bing

HUANG Xiao-feng

LU Yu

YUAN Pu

TONG Kang

LUO Hui-lan

Related Institution

School of Telecommunication Engineering, Hangzhou Dianzi University

College of Electronic and Information Engineering， Nanjing University of Aeronautics and Astronautics

College of Information Engineering， Jiangxi University of Science and Technology

'长春师范学院计算机科学与技术学院

'吉林大学符号计算与知识工程教育部重点实验室

⁰