Overview
We've got that covered.
Home | Project Overview | Process | Planning | People | Weekly Reports | Prototypes | Documentation | Task Tracker
Business Need

The Digital Data Embedding Technologies team of the Air Force Research Laboratory, Multi-Sensor Exploitation Branch (DDET) needs the ability to easily gather groups of images from a large database. Images should be retrievable by various criteria and automatically packaged into a portable archive format (a zip file, for example). The images themselves must be converted to one of many formats (jpeg, gif, etc.) and have processing algorithms (blurring, encrypting, etc.) applied to them before packaging. Adding new image formats and image processing algorithms after the product is delivered must be easy.

There must be a strong focus on the process used to develop this product. It is important to illustrate the benefits of a process by producing clear, appropriate, consistent, and useful documents. The process will be based on the Spiral Model, an iterative software development methodology used by the Department of Defense.

Project Vision and Scope

Overview

The Extensible Steganographic Database Interface (ESDI) will be a web application that allows users to quickly specify images from the STARS database, apply transformation algorithms to them, and package them into steganographic test sets. New image algorithms and package formats will be extremely easy to add.

The process developed for ESDI will be an implementation of the Spiral Model for software development. The model is very broad, and has been described as more of a process generator than a process itself, but there are a few key ideas. First, the model is risk-driven. Throughout projects, risks are identified and work is planned and completed in response to those risks. Second, there are no discrete phases in which particular artifacts are begun and completed. Instead, artifacts are developed concurrently throughout projects with the focus shifting to whatever artifacts will resolve the current risks. For example, in most spiral projects a greater amount of time is spent on the requirements specification early on to combat the risk of building a product that the users won’t like.

The goal of the ESDI process is to be useful, clear, and repeatable. If the process becomes too heavy to be followed correctly, it will not be useful. It also should be documented clearly so it can be repeated by this development team and others in future projects.

For more information on the ESDI spiral model process that will be implemented, see An Overview of the ESDI Process and ESDI Risks.

Assumptions

Listed below are DDET’s assumptions about the project.

Product

  • ESDI will be built as a Java web application.
  • ESDI must be completed by five students working part-time within a 6-month time period: December 2003 through mid-May 2004.
  • The ESDI features must be accessible to users with Windows XP and Solaris 8 systems.
  • DDET will provide all necessary image manipulation, formatting, and packaging algorithms to the development team.

Process

  • The process will be based on the spiral model for software development.
  • The process must be easily repeatable so that it can be used to promote good software practices across the Air Force Research Laboratory.

Limitations

  • ESDI will only interface with STARS at the Oracle 9i database level.