On Demand Webinar
Producing synthetic thermal imagery at the volume machine learning training demands is slow work, and most of that time goes to scene development and compositing. In this on-demand webinar, ThermoAnalytics engineers preview two new custom tools that clear both bottlenecks.
The Scene Creation Tool allows users to transform satellite imagery and topographical data into simulation-ready environments. Whether using real-world digital elevation maps or Perlin-noise-based topology, you can build realistic, full-fidelity thermal backdrops for Overhead Persistent Infrared (OPIR) training in under an hour.
The Image Compositor Tool automates the integration of simulated target signatures into measured or synthetic backgrounds. The tool enables the generation of tens of thousands of unique composite images without manual intervention, handling pre- and post-processing to replicate specific sensor effects.
Key Takeaways
- Workflow Integration: Implement the MuSES Scene Creation Tool and Image Compositor Tool in typical scene and data generation pipelines via CoTherm.
- Enhanced Accuracy: Why high-fidelity backgrounds are critical for realistic thermal simulations and signature renderings
- ML Optimization: How automated composite images improve machine learning training models.
- Speed vs. Fidelity: Leveraging compositing to produce high-volume data faster than full scene rendering.
Who Should Watch
Engineers, data scientists, researchers, and defense technologists working with thermal imaging, machine learning, and UAV/ground vehicle detection
Presenters
Audrey Levanen
Thermal/CFD Engineer
Eli Datema
Thermal/CFD Engineer
