Scribd
Upload
358 views23 pages

Mobile Document Retriver

The document presents a mobile document retriever project by four students under the guidance of two professors and project coordinators. The project allows users to interact with physical documents by taking photos of text with their phone cameras and retrieving corresponding digital information. It discusses challenges of the project like image quality and network latency. The literature review covers existing technologies like OCR, feature extraction, and mobile visual search systems that retrieve documents through image matching.

Uploaded by

Pradeep Kumar
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
358 views23 pages

Mobile Document Retriver

The document presents a mobile document retriever project by four students under the guidance of two professors and project coordinators. The project allows users to interact with physical documents by taking photos of text with their phone cameras and retrieving corresponding digital information. It discusses challenges of the project like image quality and network latency. The literature review covers existing technologies like OCR, feature extraction, and mobile visual search systems that retrieve documents through image matching.

Uploaded by

Pradeep Kumar
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 23

FIRST PROJECT SEMINAR

On

MOBILE DOCUMENT
RETRIVER
Presented by
GIRISH.A 4JN07CS025
JAIMINI.M.TAPAS 4JN07CS028
PRADEEP.P 4JN07CS056
SUSHRUTH.D.BELAGUR 4JN07CS097

Under the Guidance of Project Coordinators


Mrs.Poornima K.M B.E,M.Tech
Mr.R.SANJEEV KUNTE
B.E,M.Tech,Phd
Asst.Professor, Dept of CS & E,
Professor, Dept of CS & E, JNNCE, Shimoga.
JNNCE, Shimoga. Mr. Chetan K.R B.E,M.Tech
Sr.Lecturer, Dept of CS & E,
JNNCE, Shimoga.
Jawaharlal Nehru National College of
Engineering, Shimoga – 577204
Department of Computer Science & Engineering
CONTENTS

1.Abstract.
2.Introduction to the area.
3.Application
4.Problem Specification.
5.Literature Survey and Summary.
6.System Architecture.
7.References.
ABSTRACT:-
• The popularity of camera phones enables many exiting
multimedia applications. A camera based document
image retrieval system which is targeted towards camera
phones. Mobile document retriever application allows
user to interact with paper documents, books,
magazines, etc... Interaction is between physical and
digital world is achieved by pointing camera at a patch of
text on a document and sending it to the server where it
is matched and corresponding information is retrieved.
INTRODUCTION
Mobile applications
 Mobile Apps are apps or services that can be pushed to a
mobile device or downloaded and installed locally.
 Classification
• Browser-based: apps/services developed in a markup
language
• Native: compiled applications (device has a runtime
environment). Interactive apps such as downloadable
games.
• Hybrid: the best of both worlds (a browser is needed
for discovery)
Mobile Platforms
• A wide variety of device supporting different platforms
 Windows Mobile
 BlackBerry
 Palm OS
 Symbian
• RunTime environment and apps
 Browser-based apps (WAP)
 Flash-lite
 Java ME
 Qualcomm’s BREW
 Google’s Android
OVERALL PROCESS
APPLICATIONS
• Real Estate Guide
• Advertisement posters
• IT administrators to quickly learn server
configuration details
• Conference room application
• Retrieving imaged documents in digital
libraries and offices.
Real estate guide

• Synchronization of mobile with pc and database


Advertisement poster
• URL’s used on the posters can be used to obtain more
information regarding poster
IT administrator to quickly learn server
details
• Configuration details are immediately available on mobile
screen and can get required information.
Conference room application
• pending
Problem specification
• pending
Technical Challenges
• Most of camera enabled mobiles lack
macro and auto-focus facility to get better
quality of image.
• Achieving good Network Latency.
• Providing language independency.
• Searching time should be minimal even
though the database is larger.
• Handling of distorted image to certain
extent.
Significance
• As mobile has become part of our life it is
better that we have more and more
options in mobile to perform various tasks
in it.
• Linking physical and digital world would
ease many work
Literature survey
OCR Technology

• It examines scanned bitmap images of machine-


printed text and translates the characters into
ASCII text files that can be edited.
• It is widely used to convert books and
documents into electronic files, to computerize a
record-keeping system in an office etc..
• OCR recognition technology requires high-
quality images with excellent contrast, character
and clarity. 
Feature extraction

• Transforming the input data into the set of


features is called feature extraction.
• It involves simplifying the amount of resources
required to describe a large set of data
accurately.
• It can be used in the area of image
processing which involves using algorithms to
detect and isolate various desired portions or
shapes (features) of a digital image. 
Mobile Retriever
Xu Liu ,David Doermann

• OCR is applied to the documents while


retrieving from the database.
• It is proposed that token pairs and triplets
which respectively retrieve and verify the
candidates of retrieval at high speed and
accuracy.
Mobile visual search, linking printed
documents to digital media
xu liu, Jonathan J,Hull

• Feature extraction method is used in


retriving documents from the database.
• Quality of the input image is estimated
before sending it to recognition.
• Recognition takes place based on image
features extraction and efficient database
lookup
Document Image Matching and
retrieval with Multiple Distortion-
Invariant Descriptors
Jonathan J. Hull
• setup time (compile the database)
for each image in the database
1. locate features in the image
2. calculate descriptors from groups of features
3. enter passage identifier and location in a hash
table

• run time (match a query image to the database)


1. locate features in the input image
2. construct descriptors from groups of features
3. for each descriptor
increment accumulator cell corresponding to
eachpassage identifier
Document Finder

• Here kd-tree is used to retrieve documents


from database.
• Kd-tree is a popular data structure for fast
data point search.
• It is used with SIFT for retrieving.
Mobile Document Retrieval
Architecture
Mobile Document Retrieval
Architecture

You might also like