INTERPRETING TAMIL TEXT by USING MOBILE PHONE CAMERA Jayasakthi velmurugan K. 1 , Dorairangaswamy M.A. 2 1 Research Scholar, Sathyabama University, Chennai, India, jaisakthi21@gmail.com 2 Professor, St.Peter’s University, Chennai, India, drdorairs@yahoo.co.in Abstract -- Interpreting Tamil Text from Signboard is a well-known image processing technology. This Process consists of four steps: capture the image from mobile Phone camera, pre-processing, Text Area detection, Tamil Text extraction, character segmentation and character recognition. Signboard plates are available in various styles and colors in various Places. Every State has its own National Language Text. Each LPR system use different combination of algorithms. The information which is present in the signboard image can be detected by using the mobile phone camera. Once the Tamil Text is detected it will be converted into whatever the form which is understandable by the non native user. Keywords - pre-processing, Text detection, Text extraction, segmentation, recognition. I. INTRODUCTION Signboard information plays an important role in our society. Their format is often concise and direct, and the information they give is usually very useful. However, the foreign visitors may not understand the language that the signboard is written in, with the consequently loss of all that important information. The gaining momentum of portable mobile devices, the growing of their computational power and the inclusion of cheap digital cameras on them makes it possible to change from the classical hand dictionary translation to a new faster, comfortable and affordable way. In this sense, it is expected that a high percentage of the world population will own a mobile phone with an embedded camera, which is all that our system needs. We propose, then, a method that detects the text Information within an image taken with a mobile phone Camera, extracts it, recognizes it and translates and phonetically transcribes it from Tamil to English language. This method may have to deal with different such as: low resolution and low quality images, perspective distortion due to the viewing angle used when taking the picture, blurring effects due to object movement or incorrect camera focusing, uneven lighting from shadowing, reflections, or inter-reflections, and multiple text formats within the same image. Nevertheless, since the computational resources are limited, our system is based on non-expensive algorithms in order to achieve the highest accuracy in the lowest processing time possible. II. SYSTEM ARCHITECTURE The proposed system deals with Mobile Phone camera which is used to recognize the Tamil Text and the interpreted English Message is given back to the User. From that message the user may able to know the information which is existing in the signboard. In the below Fig.1, the camera becomes acting as a front end of the mobile phone. Once the signboard image is taken by the camera, the image is given to the back end of the system. It will do the work of finding the text message and converting it into an English text which is understandable by the user. ISSN (Print) : 2319-8613 ISSN (Online) : 0975-4024 Jayasakthi velmurugan K et al. / International Journal of Engineering and Technology (IJET) DOI: 10.21817/ijet/2017/v9i4/170904159 Vol 9 No 4 Aug-Sep 2017 3259