A GUI for OpenPIV-python

A Graphical User Interface (GUI), providing fast and efficient workflow for evaluating and post-processing particle image velocimetry (PIV) images or results. This repository is based off of openpiv_tk_gui.

Warning

This GUI was developed hastily for a PIV project and thus has many bugs. Additionally, the GUI may not be tailored to your needs as it has features that was found useful for an in-house PIV system.

Features

• Works with a wide range of image types (RGB converted to grayscale using 0.299R 0.587B 0.144G)
• Create movies (mp4, avi, non-optimized gif)
• Process movies (converts to images)
• Custom image frequencing
• Extensive image pre-processing with batch image processing (saves as 8 bit images)
• Interactive calibration
• High precision processing using iterative image deformation
• 3 subpixel estimators (gaussian, parabolic, centroid)
• Ensemble correlation with iterative image deformation
• Rectangular interrogation windows supported (25% slower)
• Average all results (excluding ensemble correlation)
• Parallel batch processing
• Extensive post-processing
• Import vectors for post-processing
• Synthetic image generation
• No proprietary environments

Example output

Documentation

See wiki

To-do list

• [ ] second peak search
• [ ] nxn 2D subpixel estimator
• [ ] direct correlation for limited search area
• [ ] precomputed dense predictor for first pass
• [ ] temporal background removal
• [ ] Uncertainty quantification
• [ ] proper orthogonal decomposition
• [ ] image rectification (calibration)
• [ ] pressure estimation
• [ ] Add stereo or other 3D processing options (maybe a new GUI for this too)
• [ ] do all PIV processing in c (including parallel)

References

• openpiv_tk_gui as a starting platform
• openpiv-python as core processor
• PIVlab for synthetic image generation and auto scaling of vectors
• Particle Image Velocimetry: A Practical Guide got me started in PIV