基于SWT法网络社交平台图像文本检测  

Social networking platform image text detection based on SWT algorithm

作　　者：乔平安 刘佩龙 QIAO Pingan;LIU Peilong(School of Computing,Xi’an University of Posts and Telecommunications,Xi’an 710100,China;Shaanxi Provincial Key Laboratory of Network Data Analysis and Intelligent Processing,Xi’an 710100,China)

机构地区：[1]西安邮电大学计算机学院,陕西西安710100 [2]陕西省网络数据分析与智能处理重点实验室,陕西西安710100

出　　处：《现代电子技术》2019年第20期48-52,共5页Modern Electronics Technique

基　　金：陕西省教育厅专项科研计划项目(16JK1689);国家自然科学基金(61105064)~~

摘　　要：网络社交平台图像包含丰富的文本信息,而文本检测是提取这些文本信息的基础。针对网络社交平台图像的特点,提出一种基于笔画宽度变换(Stroke Width Transform)的检测图像文本的方法。该方法首先预处理输入的图像,接着按照分布规则检测图像文本区域,然后根据形态规则和笔画特征规则筛除非文本区域,最后输出检测的文本区域结果。仿真实验结果表明,该方法能够准确检测网络社交平台图像文本区域,效率较好。Social networking platform image contains abundant text information,and text detection is the basis of extract?ing text information.In allusion to the features of the social networking platform image,a method for text image detection based on SWT(stroke width transform)is proposed.In this method,the input image is preprocessed,the image text area is detected according to the distribution rules,and those non?text areas are eliminated according to the morphology rules and stroke feature rules.Finally,the detected results of text area are output.The simulation experimental results show that the method can accu?rately detect the image text area of the social networking platform,and has better detection efficiency.

关 键 词：图像文本检测 网络社交平台 笔画宽度变换 文本区域检测 算法流程 仿真验证 

分 类 号：TN919.8?34[电子电信—通信与信息系统] TP391.1[电子电信—信息与通信工程]