Text Analysis Based On Mobile Devices In Natural Scene

With the development and popularity of multi-media technology and mobile devices, it has a great demand for the text analysis based on mobile devices, such as the applications of card identification for mobile phones. However, text extraction from natural scenes is more desired, such as the intelligent tour guide systems for mobile phones. The key and difficult problem is how to locate text information fast and accurately. The complicated and changefully of nature scenes affects the accuracy of text location, and hardware resources for mainstream mobile devices also limit the transplant of some outstanding location algorithms(such as methods based on connected component, texture and edge, or their advantages'integration). This work is an exploratory study of the text location and extraction on the mobile devices which supported by the Chinese National Natural Science Foundation—Research on Key Technology of Text Analysis and Processing under Natural environment.This paper aims at the study of the natural scene analysis of the text based on mobile devices. Two major problems will be solved: one is how to get the image data captured by mobile phone cameras; the other is how to locate the text information on mobile phone devices in real-time. In order to deal with these two issues, firstly, this paper focuses on the studying of a Windows Mobile cell phone camera. And then by the study of kinds of operate systems of cell phone, DirectShow is selected as the tool to get the image data captured by cameras; secondly, in view of the hardware conditions of mobile devices, it is important to choose simple edge detection and stroke detection algorithms to detect text and then extract them from the complex background. A comprehensive system-wide is designed and implemented in this paper. Furthermore, the performance of the location algorithm is tested by 300 natural scene images captured by the mobile phone. The design ideas and methods in this paper can give some useful suggestion to the field of text information analysis based on the mobile devices.