PIVlab tutorial, part 2/3: Pre-processing, analysis and data validation

Introduction

In this video, the presenter demonstrates how to process data in PIVlab after importing images, loading a video or doing image acquisition directly in PIVlab.

Importing and Selecting Images

• The presenter selects all the images except for the calibration image and imports them.

• The pairwise image sequencing style is used.

• The presenter toggles between the images to check their quality.

• A mask is drawn around the boundary of a nozzle using draw mask focus frame.

• The polygon is closed by double-clicking and then applied to all frames.

Image Pre-processing Settings

• Image pre-processing settings are set by enhancing contrast, enabling contrast-limited adaptive histogram equalization filter, suppressing low-frequency information with high pass filters, and applying denoising filters.

• To get rid of background signal, mean intensity of all images is subtracted.

• Different correlation robustness settings can be selected based on signal-to-noise ratio in your image.

PIV Settings

• PIV settings are selected based on FFT window deformation algorithm which is the most advanced algorithm in PIVlab.

• Large interrogation areas should be used to capture high displacements.

• Different correlation robustness settings can be selected based on signal-to-noise ratio in your image.

Conclusion

The presenter provides an overview of how to process data in PIVlab after importing images, loading a video or doing image acquisition directly in PIVlab. They demonstrate how to select images, draw masks around regions of interest and apply different pre-processing settings such as enhancing contrast and suppressing low-frequency information with high pass filters. Finally, they show how to select different correlation robustness settings based on signal-to-noise ratio in your image when selecting PIV settings.

Parallel Computing Toolbox

The speaker discusses how the parallel computing toolbox speeds up calculations quite dramatically.

Processing Frames in Parallel

• The speaker has the parallel computing toolbox, which allows for processing all frames in parallel.

• This results in a significant speedup of calculations.

Converting to Real World Units

The speaker explains how to convert pixel units to real-world units using calibration.

Calibration Panel

• To convert pixel units to real-world units, the speaker loads a calibration image and selects a reference distance.

• They set the x-axis to increase towards the right and y-axis towards the top.

• They set the origin of the coordinate system at the bottom of the ruler.

• After applying this conversion, clicking on an image shows coordinates in meters and velocities in meters per second.

Post Processing

The speaker discusses two validation methods: velocity-based validation and image-based validation.

Image-Based Validation

• The speaker disables all velocity-based filters and enables image-based validation.

• They apply a low contrast filter suggested by PIVlab and set a threshold of 0.5 for bright objects.

• These filters remove erroneous vectors from areas with low contrast or bright objects.

Velocity-Based Validation

• The most powerful velocity-based filter is velocity limit, where acceptable velocities are selected by drawing a rectangle around them.

• A standard deviation filter is applied globally across all velocities, while local median filter looks at every single vector and surrounding vectors.

click so this is now the average flow velocity of my experiment Plotting Velocity Magnitude

In this section, the speaker demonstrates how to plot the velocity magnitude in an experiment and export data to different formats.

Plotting Velocity Magnitude

• The speaker shows how to plot the velocity magnitude in their experiment.

• The plotted graph is displayed on screen.

• Other types of data can also be plotted using the same method.

• Data can be exported to different formats.

is something that i will show you in the next video Conclusion

In this section, the speaker concludes their presentation and provides information about future videos.

Conclusion

• The speaker concludes their presentation by thanking viewers for watching.

• Links to future videos are provided in the video description.