Limelight 2022.1 (Critical Fix, First cargo tracking improvement, Python Updates)

2022.1 Introduces a critical fix and an essential feature for teams experimenting with ball tracking.


  • We acquired information from one of our suppliers about an issue (and a fix!) that affects roughly 1/75 of the CPUs specifically used in Limelight 2 (it may be related to a specific batch). It makes sense, and it was one of the only remaining boot differences between the 2022 image and the 2020 image.

  • Fix the upload buttons for GRIP inputs and SolvePNP Models


Hue rainbow and inversion

  • The new hue rainbow makes it easier to configure the hue threshold. Here’s an example of filtering for blue pixels:

  • The new hue inversion feature is a critical feature if you want to track red objects, as red is at both the beginning and the end of the hue range:

  • If you’re trying to track balls, the aspect ratio filter (set a tight range around “1”) and the fullness filter (you want above 70%) should work quite well. Explicit “circle” filtering is relatively slow and wont work as well as people assume. Gifs of ball tracking will be posted later today.

New Python libraries

  • We’ve added scipy, scikit-image, pywavelets, pillow, and pyserial to our python sandbox. I wonder what everyone’s planning given the recent feature requests :thinking:

“Hue rainbow and inversion” is an awesome feature! Thanks.


Fixed my problem Limelight 2! And the new slider looks great.

1 Like

Our team tried to update the firmware of the limelight and afterwards the status lights would not turn on, the fans did turn on though. When I accidentally tried to mount the limelight image on windows, it said that the image was corrupted. I should probably also mention that when we tried the limelight finder it wasn’t able to find the limelight.

I am dumb, I unzipped the image file.

1 Like

You probably won’t be able to mount the image at all on Windows as its root partition is ext4 formatted.

1 Like

Ok that makes sense, but I remember I accidentally mounted the 2022.0.3 version and it worked. I might be remembering wrong though.

1 Like

What settings are people using to detect the upper hub vision target?

Will you add the files needed for pose estimation with the new vision targets?

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.