Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Targemly Graduation Project

Construction planning: Methods & equipments
by

Mahmoud Gamal

on 30 June 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Targemly Graduation Project

In good lighting and text font conditions our app accuracy reaches over 90% accuracy.
Targemly
Under supervision of:
Prof. Reda Abd El-Wahab
Dr. Mohammed El-Ramly

what?
Server Side
Technologies used
Web service was written in:
( PHP and java )
In general the work flow is:
Correcting errors
Extracting patterns
Extracting information
Why?
Android development environment ( java )
Tesser-act OCR library
PHP
Java-bridge
MySQL
Bing-Translate API
Receiving
request
Sending response
how?
thanks for listening!!
I hope you liked my presentation

Who WE are:
Amr Fouad CS-IT 20090219
Eslam Gamal CS-IT 20090041
Mahmoud Gamal CS-IT 20090302
Karim Mostafa IT-DS 20090238
Mohamed Ahmed IT-DS 20090249
Problem Statement
Receive request
Correct errors
Translate text ( Bing API )
Extract text patterns ( emails, websites , dates )
Extract info about keywords
Send response
Web service is written in PHP
Upon receiving request using Java-Bridge allows PHP to execute java-written code
OCR depends mainly on lighting conditions some errors may appear like:
Using regular expressions many word patters may be extracted:
Emails : user@site.edu.eg
Websites : www.test.com ,fci.cu.edu.eg
Dates in many patterns:
19-5-2010
Jan 15 2013
Data base filled with the famous keywords (countries, cities, famous dates, famous people).
MySQL management system used.
Full-Text search used to extract word.
All previous data are packed then it is sent to the client.
Sent in the format of JSON (JavaScript Object Notation).
JSON is used to serialize data in a way allows developers to act in an OOP way on both client and server side.
Correcting errors
Dynamic programming approach was used to fix errors.
The algorithm given a dictionary goes as follow:
Starting with min Error=3.
For each word:
Calculate edit distance.
Minimize with min Error.
If error < 3 return the found word else return the given word.
Edit distance problem is what is the minimum number of steps to convert one string to another.
Allowed steps is to insert or delete or change a letter.
We put threshold 2 characters per word.
Algorithm Steps:
Evaluation
Future work
Make some image enhancement on the taken photo
Add map features for mentioned placed in text
Improve overall system performance
wait a minute!! Something is missing!
your site is
not sustainable
!
help!
What do I do?!!
Don't worry!
I have the answer!
Android app (written in java)
Send HTTP request.
Display translated text & information.
Run OCR on the taken picture.
Get a picture.
Parse response.
Client Side
Display output
OCR
How to get the picture?
Client Side in details
Display output (cont.)
Access the camera as a hardware not as a service.
Process a frame every time-stamp.
Convert it to bitmap.
Send it to OCR.
Tesser-act library was used.
Originally developed by HP labs and in 2006 Google added some word then was released as open source.
Modified version suitable for Android environment was used called “tess-two”
Detect text regions.
Ignore detected regions out of our region of focus.
Draw translated text in white boxes on top on the camera using canvas.
Parse the information came from the server into their corresponding categories.
Links are shown as hyper links.
Emails can be used by other applications to send mails.
Other translated data will be received with a link to Wikipedia for more information.
imagine someone with a sign he can't read, then he needs fast and easy translation way.
We live in a world that became small and everyone is used to get information in a fast and easy way through technology.
mobile devices became most popular and internet connection became almost everywhere.
Idea
develop an mobile application that uses camera to extract and translate text.
OUR Rivals
take pictures then highlight text area to extract text and translate it.
Word lens
Google translate
open camera view then replace text with translated text.
targmely
same as word lens with extract information feature.
Release more APIs for developers
Full transcript