Loading presentation...

Present Remotely

Send the link below via email or IM


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.


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

Android Data Capture Application

No description

charles waldron

on 16 April 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Android Data Capture Application

Android Data Capture Application
Meet The Team
Charles Waldron

Project Lead
Project Overview
Sponsored by Sargon Partners

OCR - Tesseract
Image Enhancement is Crucial
Image Capture and Crop
This helped to speed up the OCR processing and lowered the amount of errors
To be implemented as part of a larger inventory management system
Use Android device to capture data from manufacture's tags via Optical Character Recognition
Image Enhancement
Data Parsing
Serial Number
Asset Number
Part Number
Model Number
For Sargon Partners
Through research and testing we discovered that certain manipulations to the image improve the OCR results
We used the Leptonica open source library to manipulate images
Edge manipulation
Application Demonstration
Database Push
The data can then be moved to an external Database
Utilized Google's camera API to get highest resolution image possible from device's on-board camera.
OCR returns the best results from crisp, high res, in-focus images
This pushes the limits of the device memory
We wrote our own image cropping activity to remove unwanted text
Once the text is obtained, the app looks for the following:
Found text is then entered in respective text field for user to view
The user checks that all information is correct
The user can manually enter information
The user can select from the information obtained by the OCR
The user submits the information to the phone's internal database
The Research
Android Development
Optical Character Recognition
Google's Android Developer Knowledge Base
Android Programming: The Big Nerd Ranch Guide
Eclipse IDE
Google Code Tesseract Page
rmtheis' Tess-Two GitHub
Google Play App Store
The Requirements
Internet not required
Android Version 2.0+
Android Version
Android Version 4.0+
Image Capture
Image OCR
Text Placement
Push to SQLite DB
User Text Input
Target Device
Google Nexus 7 (Original)
Samsung Galaxy S3
The Prototypes
Built upon rmtheis' OCR Test application
Android Web Service
Android Camera App
Nick Busse

Bryan Farris

Mo'ath Nazzal

Document Manager
Research (OCR)
Testing (Image Enhancement)
Android Developer (OCR, Multi-Threading)
Research (Android, Image Enhancement, OCR)
Testing (Android, Image Enhancement, OCR)
Android Developer (Database, Parsing Algorithm, User Input)
Research (Android, OCR)
Testing (Android)
Android Developer (Image Capture, OCR, Multi-Threading, UI)
Research (Android, Image Enhancement, OCR)
Testing (Android, Image Enhancement, OCR)
Make the process as automated possible
The App
Originally developed at HP between 1984 and 1994
Most accurate open source OCR engine available
Our app achieved above 90 % for target text
Written in C++
Must be compiled with NDK to run in Android
Originally created to be used with flatbed scanners
This poses a problem when trying to implement on a mobile device
Road Blocks - What Went Wrong?
Problem Encountered:
Low Recognition Percentage
More Image Enhancement (Leptonica), Cropping The Image
Problem Encountered:
Communication w/ Client
Weekly Meetings on Google Chat
Problem Encountered:
Device Run time Memory Issues
Better Bitmap\Thread Management
Cost Analysis
Estimated Hours (September 2013 - March 2014)
Monetary Cost Estimation
Total Project Hours: 733
Full transcript