Photo-Identification by Internet

A very big part of the research we do involves individually identifying the animals we study. Conveniently, the species we work with all have pigmentation patterns which are permanent and unique to each individual. This removes any need to tag or mark individuals and worry about tags getting lost, or the ethical issues with injuring an animal to mark it.

So what we do is take photographs of the animals, process them into a uniform digital format, then compare them to catalogues of animals identified previously. We have been using this process quite successfully to identify grey whales, humpback whales, bottlenose dolphins and red-legged frogs. Our catalogues are growing steadily, with most in the low hundreds.

The process of comparing photographs to the catalogues is not difficult in principle, but, as you can imagine, it does take time, and lots of it! Until recently, the matching has been done by our staff & students, helped out very significantly by a large number of Earthwatch volunteers.

Now, however, we are looking to expand our capabilities, so that we can identify and catalogue more individual animals. We are particularly interested in collecting so-called "incidental" photographs from whale-watchers in the areas where we work. These shots will not always be exactly what we are looking for in a high-quality identification photograph, but many will be very useable, and so if we can sort through them, we will be tapping into a resource that we would simply not otherwise have access to.

The problem will be that this exercise will completely swamp our ability to process and catalogue the photographs. Hence this page. We are asking for your help to sort through the mountain of photographs. We hope that we will be able to harness the power of the internet and recruit a large number of individuals willing to give us a few hours of their time to hunt through the catalogues for matches.

We've been doing this for a while now, and recognise that not every potential matcher was created equal... :-) So what we're trying to set up is a system whereby each individual contributes the matches they believe they have to a common database. Then as more and more people go through the pictures and make the same matches, any mistakes will slowly filter themselves out.

This is all a bit of an experiment right now, so please bear with us while we get these pages up and running.

If you're interested in helping us out, then please follow these steps in order:

1. Introduction

For a brief introduction to photo-id, please see this page.

2. Training

For some more training before you get started on the real thing, please see this page for a longer, self-marking "exam".

3. Familiarisation

To see our on-line photo-id catalogue and familiarise yourself with the controls, please follow this link.

4. Matching

If you're ready to get started, then please enter your email address in the box below, and click "go". That will take you to the catalogue proper, where you will be working live with our database. The controls should be obvious.

Thank you very much in advance for all your help!

All the best,

William