Advanced search

Message boards : Image and Video Discussion : To the scientists/imagers

Author Message
count jim moriarty
Send message
Joined: 4 Sep 15
Posts: 11
Combined Credit: 6,634,955
DNA@Home: 106,647
SubsetSum@Home: 270,506
Wildlife@Home: 6,257,802
Wildlife@Home Watched: 964,833s
Wildlife@Home Events: 588
Climate Tweets: 3,934
Images Observed: 1,016

Message 6490 - Posted: 2 Dec 2016, 11:46:29 UTC

I would like to discuss the following trail cam image:

Okay, have looked at 3 of these images having looked first at the training views.

I have a few points needing to addressed:
1) Where is the scale? How high up is the UAV? This is important as the white spots could be rocks, boulders other animals or other birds.

2) Training picture lets you know what to look for regards target species...however no overhead pictures of all the other species when trying to work out if the white spot is a goose or another species. Please fix this, it makes it awkward not knowing what these look like, so we can discount them during our searches.

3) Image pixellates as zoom in commences increasing difficulty, which I gather is not the intended idea.It makes ID'ing even harder not easier.

Lastly, comments regarding using UAV, fantastic idea, but you need a camera with better resolution. Having low resolution photos with no scale and no overhead pictures of the other species of birds and animals makes for a large amount of false positives.Which would it seems to me to defeat the whole point of the exercise.
Looking forward to your reply

Marshall Mattingly
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Send message
Joined: 3 Feb 16
Posts: 27
Combined Credit: 74,756,298
DNA@Home: 0
SubsetSum@Home: 0
Wildlife@Home: 74,756,298
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 1,335

Message 6491 - Posted: 6 Dec 2016, 0:48:51 UTC - in response to Message 6490.

Thanks for you suggestions! I work on the computer science / interface side, so I'll do the best I can do address these:

1. This is a great suggestion. We have this information available in the database and I'll look to expose it to the interface to provide a better sense of "height" as suggested.

2. I'll forward this on to the biology team to try and get some more comprehensive examples.

3. I've been working on tweaking the interface so zooming is less of an issue, and hope to have that out soon.

Post to thread

Message boards : Image and Video Discussion : To the scientists/imagers