Unwarp is an open-source tool that enhances image resolution by eliminating scintillations caused by atmospheric turbulence and similar distortion phenomena. The software processes a series of images of the same subject, aligning and stacking them using advanced feature selection algorithms and phase correlation approaches.
The core technique matches features between images, applies triangulation across the entire frame, and warps each pixel to its optimal position. The resulting aligned images are then averaged into a single output, with automatic outlier elimination for cleaner results.
Unwarp is equally effective for telephoto image series of terrestrial objects affected by heat shimmer or atmospheric distortion.
Unwarp supports common image formats and provides an intuitive interface for both beginners and advanced users. The project is currently in early development, with active work on improving algorithm performance and expanding feature sets.
Features
- Eliminates atmospheric scintillation and turbulence effects from image series
- Advanced feature matching algorithms for precise alignment
- Phase correlation approach for accurate pixel registration
- Automatic triangulation-based image warping
- Intelligent outlier rejection during image stacking
- Support for both astronomical and terrestrial photography
- Customizable processing parameters for different imaging conditions
- Batch processing capability for large image sets
- Cross-platform compatibility (Linux, Windows, macOS)
- Open-source development with active community support
- GPU acceleration for faster processing (planned feature)
- Integrated image preprocessing tools