The project aims to enhance target recognition and classification for 5th Generation Fighter Jets by using images taken by Electro-optic and Infrared Radar Systems’ sensors and detect/classify the targets from those images.

A labeled dataset of military aircraft in the IR spectrum is crucial !!! Detecting and classifying military aircrafts using IR imagery.

• PAGE

👁️‍🗨️ To Do

• [ ] Dataset Investigate (Kaggle, Yolo, Tensorflow Creat Dataset)
• [ ] Project Proposal Report
• [ ] Image Labeling Can It Done? (just for training) Search

🚀Progress

• [ ] Paper Research INFRAGAN paper
• [ ]

📣 Done

• [x] Select Paper
• [ ]

1. SSL-RGB2IR - thesis project related
2. Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks - CycleGAN base
   • https://junyanz.github.io/CycleGAN/
3. InfraGAN - ML PROJECT
   • InfraGAN: A GAN architecture to transfer visible images to infrared domain

Önce CycleGAN - > uçak IR data olmadığı için onu bi üretelim normal RGB den daha sonra, RBG + IR data üzerinden uçakların detection / classification yapılabilir.

InfraGAN network iyileştirme çalışmaları

Simulation Environment