plan.md

Plan

• Deep neural networks for classification of multispectral images
• Niklas Lindorfer
• 2265846l
• Professor Roderick Murray-Smith

Week-by-week plan for the whole project.

Winter semester

• Week 1
  • initial environment setup
  • first research into technologies and applications
• Week 2
  • continue research
• Week 3
  • continue research
• Week 4
  • acquire some dataset for testing
  • explore some traditional classification methods (Logistic Regression, SVM)
• Week 5
• Week 6
  • reproduce previous project for testing and training purposes
• Week 7
  • find off-the-shelf dataset for initial development
• Week 8
• Week 9
  • Image registration problem
• Week 10
• Week 11 [PROJECT WEEK]
  • Develop mobile application
• Week 12 [PROJECT WEEK]
  • Status report submitted.

Winter break

Spring Semester

• Week 13
  • Select application
• Week 14
  • Capture dataset
• Week 15
• Week 16
• Week 17
  • Evaluate different models
• Week 19
• Week 20
• Week 21
  • Dissertation writing time
• Week 22
• Week 23 [TERM ENDS]
• Week 24
  • Dissertation submission deadline and presentations.