SahanaOCR

SahanaOCR is the automated form processing system of the Sahana disaster management system. The main objective is to eliminate the bottleneck at data feeding stage. This standalone application provides users to one-click data feeding option.

Architecture

Architecture

Overall Description

Functional Requirements

  • Scan application forms
  • Recognize the Data in forms
  • Upload data to the server
  • Alert user for incorrectly recognize data

Non-Functional Requirements

  • Validate form
  • Higher accuracy
  • Bulk upload
  • Detect errors
  • Present current progress
  • Platform independent

Current Status

  • Development (GSoC 2009 Project)

Development Roadmap

  • Pre-Development:
    • Domain Ontology
    • Requirement gathering
    • High-level component Design
    • XML schema design
  • Initial Development:
    • Loading and validating XForm
    • Image processing and segmentation
    • OCR
    • UI design
    • Uploading result
  • Improvements
    • Improve OCR performance
    • Better image processing algorithms
    • Make platforms independent
    • Parallel processing

Drafts/Discussion

Discussions

References


Navigation
QR Code
QR Code sahanaocr (generated for current page)