You are on page 1of 14

B.M.

S COLLEGE OF ENGINEERING
Bull Temple Road, Basavanagudi, Bangalore - 560 019, Karnataka, India

DICTONARY APP DEVELOPMENT


Ashwin N – 1BM14EC047
Krishna Sai Tarun - 1BM14EC046
Kushal N - 1BM14EC047
Nikhilesh -1BM14EC062
Under the guidance of
Lalitha S
Assistant Professor

Major Project
2017 – 18
Department of Electronics and Communication Engineering
Introduction
• Digitalization has turned the world into a very accessible place to each and every
person on the planet wherever the person is. In this rapid broadening of
digitalization, a mobile phone has played a major role. Most of the world’s
population possess a mobile phone these days which acts an interface to every
other thing on the world. Digging a layer into the mobile phone, we find that there
are certain mobile applications that allow people to perform actions.
• In Karnataka, the official language for communication is Kannada. The urbanization
and technological advancements have made this state a gateway for talent from all
across the world. In today’s Bangalore’s population, roughly 47% of the people
either don’t know to read Kannada or speak the language.
Problem definition
• The primary goal of this project is to develop an android application
which helps the user in getting the meaning of a desired word. In
our project we deal with two languages namely English and
Kannada. The application should provide the meaning of a desired
Kannada word in English and vice versa.

• The meaning of only the desired words (Kannada/English) should be


the output of the application. The source of the input can be taken
from anywhere whichever is convenient the user.
PROPOSED SOLUTION
• The problem is mainly approached using the concepts of Optical Character
Recognition (OCR) concept. It is the mechanical or electronic conversion of
images of typed, handwritten or printed text into machine-encoded text,
whether from a scanned document, a photo of a document, a scene-photo
(for example the text on signs and billboards in a landscape photo) or from
subtitle text superimposed on an image.
Different software's used

1) Android Studio

Android Studio is the official integrated


development environment (IDE) for Google's
Android operating system, built on JetBrains
IntelliJ IDEA software and designed
specifically for Android development.
2) Tesseract

Tesseract is an open source text recognizer


(OCR) Engine, available under the Apache
2.0 license. It can be used directly, or (for
programmers) using an API to extract printed
text from images. It supports a wide variety of
languages.
3) SQLite Database
SQLite is a relational database management system
(RDBMS). If most RDBMSs such as MySQL, Oracle, etc. are
standalone server processes, then SQLite is embedded
because it is provided in the form of a library that is linked
in applications.

Like other RDBMSs, data is accessed in a SQLite database


by using Structured Query Language (SQL).
Flow of Android Application
Results

The OCR flow of the application is shown in the above figures. The application is user
friendly. In the Welcome screen, the user can either upload/take a picture or type in
the required word. If the user chooses to upload/take a picture, the OCR engine is
initialised. The crop function gives the user the freedom to select only the desired
word in the picture. The enhance button, makes sure that the threshold of
binarization is set according to the surroundings
The user can also type the word in the welcome screen, to get the desired
meaning and translation of the required word. On clicking the “Search Madi”
button on the application, the typed word is searched in the offline SQLite
database and the respective results are displayed on the screen.
The OCR system is tested for complex Kannada ottakshara’s as seen in Figure.The word
“ಸ್ವಾ ತಂತ್ರ್ ಯ ” which has double ottakshara is fed into the OCR system. As we can see in the
figure the OCR engine has no problem with detecting complex words of Kannada language
as well. Different words have been tried to test the OCR engine and there is remarkable
accuracy as the binarization threshold has been given to user. The OCR engine is tested
across many words and many combinations of words and results are obtained with better
accuracy compared to Google’s 49% for Kannada Language.
Conclusion
We learnt and implemented the following for the successful completion of the
mobile application:
• Android App Development using Android Studio (Java Programming)
• Android SDK (Compiling the code for different Android SDKs)
• Android NDK (To import C++ libraries that can be used in Android)
• Interfacing Teserract with Android (for the OCR Implementation)
• Interfacing SQLite database with Android (for the database)

As proposed, the application can perform and it possesses the following:


• Convenient user interface making it easy to use
• Dual Input (Text or Image)
• Pre-Processing the image: Cropping and Binarization
• Processing using Kannada or English OCR engine
• OCR output or Text Input which can used to obtain the desired meaning by
referring the database.
FUTURE WORK
• Implementing a voice input of the required word and get the meaning in both textual
as well as in voice output format.
• More words in the database can be added to get wide range of word’s meaning.
• The efficiency of the application can be improved along with the reduction of noises.
• The application can be made to take live pictures and get the meaning immediately.
Images from a live video can be taken to get the meaning of the desired word
provided the noises produced by it is reduced or taken care of to increase the
efficiency.
• Along with getting the meaning of the word it can be improvised by also including
the pronunciation of the word feature.
• This application is not only limited to the Kannada language, it can be implemented
with other languages provided if there is trained data and appropriate database.
• The application can be made to recognise handwritten characters as well in the
future. All types of fonts should be recognised by the application.
• After getting the meaning of the word, an example sentence where the particular
word is included in the sentence can be made to make it more helpful for the users
to understand the language.
Thank you

You might also like