Change size of object to track

[tshaible]

I am trying to increase the size of the tracked object, as I work with Monarchs rather than flies.

I have included the following line of code in an attempt to do so, but it doesn't work:

self._object_expected_size = 0.05 # proportion of the roi main axis
self._max_area = (2.5 * self._object_expected_size) ** 2

Any help would be greatly appreciated

[ggilestro]

Changing the expected size should suffice

ethoscope/src/ethoscope/trackers/adaptive_bg_tracker.py

Line 249 in 0521648

self._object_expected_size = 0.05 # proportion of the roi main axis

If you send me a short video (few minutes) I can try to have a look.

[tshaible]

day8.avi <https://drive.google.com/file/d/1jLeAMAFmzBl4fMZ9g4b-0EveUzQANfpH/view?usp=drive_web> moreblocks.png <https://drive.google.com/file/d/1cgBkUu8_QDfVmoLfNTSSzqyFU0M5nkRn/view?usp=drive_web> trimmed.mp4 <https://drive.google.com/file/d/1bD0EvVNj4-undaYpFVgOlcetNq32dRvl/view?usp=drive_web> Hello, I have finally been able to get the size change to work, although even when I've changed the expected size to be quite large, there are still many small objects tracked (see attached video named day8). I realize there is too much background noise in this video and in the future we are going to better account for this, but I'm wondering if something might be done to clean up the tracking? I attached the original video as well as the roi image I am using as well. Thanks for any help you can provide, Tierney Shaible

…

On Mon, Jan 18, 2021 at 6:29 AM Giorgio Gilestro ***@***.***> wrote: *External Email* Changing the expected size should suffice https://github.com/gilestrolab/ethoscope/blob/0521648f33103b8055342aea3ef37c06fde150a1/src/ethoscope/trackers/adaptive_bg_tracker.py#L249 If you send me a short video (few minutes) I can try to have a look. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#142 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ASO7LE7FQEMNOAZYMPT6VHDS2QLR7ANCNFSM4WC4XZOQ> .

[tshaible]

Just wanted to follow up on this matter to see if you might be able to provide any help or direct me to someone who can help. Thanks! Tierney On Wed, Jan 20, 2021 at 3:30 PM Tierney Marie Shaible < tshaible@email.arizona.edu> wrote:

…

day8.avi <https://drive.google.com/file/d/1jLeAMAFmzBl4fMZ9g4b-0EveUzQANfpH/view?usp=drive_web> moreblocks.png <https://drive.google.com/file/d/1cgBkUu8_QDfVmoLfNTSSzqyFU0M5nkRn/view?usp=drive_web> trimmed.mp4 <https://drive.google.com/file/d/1bD0EvVNj4-undaYpFVgOlcetNq32dRvl/view?usp=drive_web> Hello, I have finally been able to get the size change to work, although even when I've changed the expected size to be quite large, there are still many small objects tracked (see attached video named day8). I realize there is too much background noise in this video and in the future we are going to better account for this, but I'm wondering if something might be done to clean up the tracking? I attached the original video as well as the roi image I am using as well. Thanks for any help you can provide, Tierney Shaible On Mon, Jan 18, 2021 at 6:29 AM Giorgio Gilestro ***@***.***> wrote: > *External Email* > > Changing the expected size should suffice > > > https://github.com/gilestrolab/ethoscope/blob/0521648f33103b8055342aea3ef37c06fde150a1/src/ethoscope/trackers/adaptive_bg_tracker.py#L249 > > If you send me a short video (few minutes) I can try to have a look. > > — > You are receiving this because you authored the thread. > Reply to this email directly, view it on GitHub > <#142 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/ASO7LE7FQEMNOAZYMPT6VHDS2QLR7ANCNFSM4WC4XZOQ> > . >