|
Testing / Example Zooplankton Image Database
| |
|
 |
 |
|
| |
Hello all!
I'm currently comparing a couple of image recognition software programs and before I scan zooplankton we've collected in the field, I was hoping to do a test run with some readily available image data, preferable sorted into taxonomic groups such that they could be used as a training set.
I know some of these data are available with ZooImage's example data, however these training data images all have little scale bars in them and I was hoping to find a training set without these scale bars.
Thanks for the help!
-Heckler511
|
|
|
|
|
|
|
| |
|
|
|
|
| |
Would anybody be willing to share zooplankton image data they have collected and / or worked with in the past?
Thanks
|
|
|
|
|
|
|
| |
|
|
|
|
| |
Hi Heckler
I think the images with scale bars that you are referring to is the vignettes that are automatically generated from the main image of a sample? The scale bar is thise are automatically placed and shouldn't pose a problem. However, using these images for actualy analysis might not be of much use.
Could you explain what type of sampels you are working on e.g. mesh sizes/volumes?
|
|
|
|
|
|
|
| |
|
|
|
|
| |
Hi jensr,
Thanks for the reply. We have been doing vertical tows with 150 micron mesh. Samples are collected in 500ml bottles. We will probably need to split many of the samples up for analysis.
In the meantime I was hoping to play around with some plankton data that people have already used for analysis. I could be a little confused, as perhaps this won't work, but it seems to me that if someone had a training set and a set of unknowns I should be able to use them for analysis.
|
|
|
|
|
|
|
| |
|
|
|
|
| |
There has been discussion about creating a "Golden Standard" training set that owuld contain a library that samples could be ran against. However, to my knowledge, such a thing has not yet been established.
Under all circumstances you will probably still hit some problems when running it against your own scanned sampled in terms of making sure the right species/groups are included and that all the "junk" also gets categorized appropriately.
I junk management is a big challenge in its own in dealing with multiple types of items like broken off appendages, marine snow etc. I suspect that it might be diifucult to actually establish something that could be used generically for that - there would still be some local adaptation of a training set needed.
|
|
|
|
|
|
|
 |
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum
|
|
|
|
|
|
|
|
|
| |
