🥳 New version is out, 0.8.0, with new matching algorithm and an integrated background subtraction in the segmentation module

[ronshnapp]

MyPTV is continuing to grow and evolve, and now, we have two new additions that bring MyPTV into version 0.8.0.

The first addition is a new stereo matching algorithm that can decrease the computational time and the eventual quality of the results in experiments with numerous particles. The new algorithm takes advantage of a new blob search strategy, and a time marching of particles that were already found in previous frames. More details are in the user manual.

The second addition is a new utility that was integrated into the segmentation module. In experiments with non-idea lighting conditions, inhomogeneities in the particle images can make the segmentation process challenging. Issues such as these can often be eased with a background subtraction. Therefore, the new version of MyPTV includes functions that can automatically calculate the background image and subtract it from the other images during the segmentation. Again, more details are in the manual.

I believe these are two significant improvements of MyPTV, as that can help obtain better results in difficult experiments that include numerous particles and hard lighting conditions. I personally used these new methods in a new work of mine and I saw significant improvements both in terms of the quality of the results (length and number of trajectories) and in terms of computational time of the matching process. I hope you will find them useful for your own work as well.

Ron