A Novel Approach to Text Detection and Extraction from Videos by Discriminative Features and Density  

作　　者：WEI Baogang ZHANG Yin YUAN Jie LIU Yonghuai WANG Lidong 

机构地区：[1]College of Computer Science and Technology, Zhejiang University [2]Department of Computer Science, Aberystwyth University [3]Qianjiang College, Hangzhou Normal University

出　　处：《Chinese Journal of Electronics》2014年第2期322-328,共7页电子学报（英文版）

基　　金：supported by the National Natural Science Foundation of China(No.60673088);He-gao-ji National Major Technology Special Project(No.2010ZX01042-002-003);Chinese Knowledge Center of Engineering Science and Technology(CKCEST);Specialized Research Fund for the Doctoral Program of Higher Education(SRFDP)(No.20130101110136);China Academic Digital Associative Library(CADAL)

摘　　要：Text is very important to video retrieval,index, and understanding. However, its detection and extraction is challenging due to varying background, low contrast between text and non-text regions, and perspective distortion. In this paper, we propose a novel two phase approach to tackling this problem by discriminative features and edge density. The first phase firstly defines and extracts a novel feature called edge distribution entropy and then uses this feature to remove most non-text regions. The second phase employs a Support vector machine(SVM) to further distinguish real text regions from nontext ones. To generate inputs for SVM, additional three novel features are defined and extracted from each region:a foreground pixel distribution entropy, skeleton/size ratio, and edge density. After text regions have been detected, texts are extracted from such regions that are surrounded by sufficient edge pixels. A comparative study using two publicly accessible datasets shows that the proposed method significantly outperforms the selected four state of the art ones for accurate text detection and extraction.Text is very important to video retrieval, index, and understanding. However, its detection and ex- traction is challenging due to varying background~ low con- trast between text and non-text regions, and perspective distortion. In this paper, we propose a novel two phase approach to tackling this problem by discriminative fea- tures and edge density. The first phase firstly defines and extracts a novel feature called edge distribution entropy and then uses this feature to remove most non-text re- gions. The second phase employs a Support vector machine （SVM） to further distinguish real text regions from non- text ones. To generate inputs for SVM, additional three novel features are defined and extracted from each region： a foreground pixel distribution entropy, skeleton/size ra- tio, and edge density. After text regions have been de- tected, texts are extracted from such regions that are sur- rounded by sufficient edge pixels. A comparative study using two publicly accessible datasets shows that the pro- posed method significantly outperforms the selected four state of the art ones for accurate text detection and ex- traction.

关 键 词：Text detection Text extraction~ Edge distribution entropy  Foreground pixel distribution en- tropy  Edge density. 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]