Multiple Mouse Tracking

Project Discussion
1

Hi there,

I’m interested in using the OpenMV for chronic tracking of 4-5 mice in their home cage. I work at a university and will receive approval for any procedure I mention, so don’t worry that I’m some renegade experimenting on mice myself.

We don’t currently have the server capacity for storing and processing such a large amount of video data, so I’m hoping to use the OpenMV to extract necessary information regarding the position of each mice and dump most of the video data. Ideally, we would be able to capture the center position of each mouse’s body, as well their paws and head.

The most straightforward way people have tracked mice in the past was by coloring them. Unfortunately our mice are brown, and we don’t want to bleach them so we can dye their fur. I was hoping the multi-color detection may help here. Do you think if each mouse had say a different color collar, then they could be identified by looking for a combination of brown and their collar color?

For detecting the paws/head, I was thinking something reflecting IR could work. Then I could easily match the paws/head to each mouse based on proximity to the mice by multi-color blob detection.

These are just some ideas, if anyone has any other insight I’d love to hear it. We’ll need to wait for our OpenMV to arrive as well as regulatory approval to start any experiments.

Thanks,
Alec

2

You could use tags, barcodes etc… too (if the collars are big enough), but colors should work fine. It’s hard tell what would work best, you have to try a few things first and find out.

3

You just need good lighting, color tracking, and some high contrast over saturated color to track. The color you pick and the lighting are key.