Welcome to Snapshot Serengeti

Hi! And welcome to Snapshot Serengeti. We are all incredibly excited to be working with you to turn photographs into scientific discoveries. You might be wondering what this is all about, so let me start with some introductions. This is Ali:

ali_Fabio

Ali is a researcher at the University of Minnesota. She studies the big carnivores (lions, hyena, cheetahs, and leopards) in the Serengeti. Every year she flies to Tanzania, loads up on supplies in Arusha, and then drives for a day – mostly on dirt roads – out into Serengeti National Park.

TanzaniaSerengetiNP

This is Craig:
Craig-Packer-Tanzania-3

Craig is a professor at the University of Minnesota and Ali’s advisor. He runs the Lion Research Center has been studying lions out in the Serengeti for decades. He has radio collars on lions in many prides, which allows him to keep track of lots individual lions over many years.

This is Daniel:

DanielRosengren

And this is Stan:

Stan

They are field assistants who work for Craig out in the Serengeti. Daniel is responsible for driving around and finding lions, while taking pictures of them and recording lots of information about what he sees. Stan is responsible for going out to the camera traps, making sure they’re still working fine, and changing the cameras’ memory cards when they fill up. Daniel and Stan live in Serengeti year-round at Lion House, where facilities are basic, but the scenery is amazing.

LionHouse

When Ali goes out the Serengeti, she stays at Lion House, too. Once she’s there, she makes observations that help her understand the big carnivores. A couple years ago, she installed a bunch of camera traps so she could see where the carnivores roamed when she wasn’t present. The cameras worked really well and the images were so useful that she installed some more. Now there are 225 of these cameras automatically taking pictures out the Serengeti!

camera

My name is Margaret. This is me:

Margaret2
Like Ali, I’m a researcher at the University of Minnesota, and Craig is my advisor, too. Ali became inundated with the images the camera traps produced – a million per year! I have a reputation around here as a computer fundi – a Swahili word that translates as ‘master’ or ‘expert’ – and Ali asked me if there was a way to automate the process of turning images into data. See, the images by themselves aren’t that useful for research; Ali needs to know what species are in the pictures so she can do her analyses. For example, if she knows which images contain wildebeest and zebra, she can use that data put together a map that shows their density across the landscape. (The size of the circles show how many wildebeest and zebra there are in various places — bigger circles mean more wildebeest and zebra.)

animalMapUnfortunately, I had to tell her that computers aren’t that good yet. They can’t yet reliably pick out objects from a picture, except under very controlled situations. But human eyes are remarkable in their ability to find objects in images. As I started looking through Ali’s images, I was blown away by how beautiful many of them are. And I wondered if we could ask for help from people. Lots of people. Hundreds. Thousands. So we started to think about how to do that.

The end result is Snapshot Serengeti, a collaboration with Zooniverse. We’d like to ask you to help us turn all these pictures from the Serengeti into scientific data by identifying what animals are in the images and what they’re doing. And in this blog, we will keep you updated on how the project is progressing, share cool information about the Serengeti and African wildlife, as well as hopefully answer a lot of questions you may have about animal behavior, ecology, and science in general.

So, check out the camera trap images. Tell us what you see in them. And let us know if you have questions. Thanks! You can get started by clicking here.

About Margaret Kosmala

I am an ecologist exploring the complex dynamics of plant and animal systems. I am especially interested in understanding how species communities change over time and how humans impact them.

26 responses to “Welcome to Snapshot Serengeti”

  1. Paul Miles says :

    I’ve tried to follow the tutorial – but get no further than the first page of this. I have been prompted to choose ‘Antelope/deer’ from the list of animals on the right of this prompt. However, the choice of ‘antelope/deer’ is not visible to choose. Frustrating!!!

  2. Lalitha says :

    Love this website. What a fantastic idea, and practical way to teach high school kids some aspects of environmental science. Any ideas how I can make this into an educational tool for k-12 and plan activities tied up to this website? The most obvious being tag as many pics as you can..

    • jon brees thogmartin architect says :

      why doesn’t the classify work w/ mac os 10.7.5?
      this is another zoo universe that does not work. what is the problem w/ this website? it does not get off the intro to classifying.

  3. William Wilson Hutchinson says :

    great idea

  4. Marina Isaac (@rockhyrax) says :

    Yes, great idea, Margaret. The only thing I’ve noticed missing is the “hyrax” classification. Or are you unsure whether people would be able to distinguish between the species you’d expect to see?

    I think there are quite a few of us who will be taking part in the exercise specifically in the hope of seeing one of these beautiful animals…

    • Margaret Kosmala says :

      I’m afraid hyraxes and other small mammals don’t tend to show up in our camera trap images very often. The cameras are a few feet off the ground, and so the smaller animals don’t tend to trigger them.

  5. Georgia Lake says :

    Absolutely gorgeous. I’ve spent over an hour just classifying the photos, and it’s so rewarding to get the occasional elephant, giraffe, warthog, or relaxed pride of lions!

  6. Alejandro Torres Frías says :

    I am happy to be of help in this project. I’d like to confirm that/if an ipad can be used to tag along, since it’s the device I have handy on the times I can collaborate on this. I took the tutorial just fine with it (ipad) but keep on getting stuck when pressing “finish” to start clasifying. Thanks!

    • jon brees thogmartin architect says :

      using a macbook pro and it keeps on getting stuck when pressing “finish” to start clasifying. Thanks!

  7. Kyla says :

    I love hearing that this work is being done at the University of Minnesota! Great work :)

    However, I’m using a Macbook Pro with iOS Mountain lion on firefox and the finish button is unlinked! I cannot submit my classifications because the finish button does not work. Please fix, this looks like fun!

  8. Charlie Francis says :

    Clicking Finish after I identify something, does nothing. I continue to click and it does not go to the next capture. If I click Refresh a new capture comes up, but the previous capture is not recorded. I’ve tried Chrome and IE, both do the same.

  9. drake lupus says :

    I am really happy to help, on whatever free time I have

  10. D. Patrick Hynes says :

    Wonderful Project. Excellent Website. Sorting images while listening to good Irish Music….. Now thats good for my soul.

  11. Megan says :

    I’m loving this! But one option I’m missing is an “I have no clue” button — I have found several photos where there was definitely SOMETHING showing, like a random paw or ear that did not allow identification at all, and I didn’t want to tick “Nothing here” because well, there was definitely an animal there. Just not identifiable to me.

  12. Corrie says :

    Would like to help with classification but cannot get past the tutorial page. Any suggestions?

  13. Szura says :

    This is such a wonderful idea! :) I really enjoy looking at this great photos and if while doing it I can help someone – I’m just very happy.

  14. bluepip says :

    Loving this project! :D I’ve already developed ‘just one more classification’ syndrome. lol

    I do agree with Megan about the need for an option to indicate where an animal is present within an image, but can’t be classified. Would be most useful :)

    • Gilly says :

      I agree too. I find it difficult if the animals are too far away. I’d like to record that there’s something there, and maybe someone with better software or more experience can identify them.

  15. Lenka says :

    Hello, I like help with your project, I think it is a very good idea. I tried it. But I need other option too, for example – on the image is the animal, but this animal not in options, – on the image is only a part of animal, i have problem with identificantion (but it is not option nothing here).

  16. carolfrueh says :

    Great fun! I hope lots of folks get involved. Needs some sort of timer like Snood so you can set yourself to play/classify only one more! I’ll be back…again and again.

  17. Daniel Cornwall says :

    Reblogged this on Alaskan Librarian and commented:
    Here’s an actual introductory post to the Snapshot Serengeti project I’m playing a teensy part in as a layman classifier. It’s fun work. So much fun that the researchers have about half their photos analyzed and the project has only been live a few days.

  18. cesarakg says :

    Loved the images.

    Is there a way to get a bigger copy of them? The ones I marked as “favorite”, or the ones I marked some animal?

    Best regards

  19. Carol Russo says :

    I just sent an email out about this site to the entire Yale Neuroscience community (I’m the administrator for the grad program) and I am totally in support of grad student research!

Trackbacks / Pingbacks

  1. Hunt for African Wildlife From Your Computer | Smart News - December 12, 2012

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 4,734 other followers

%d bloggers like this: