Educating new generations of marine mammal scientists is important for dolphin conservation efforts -- not just in the U.S. but worldwide -- and we're glad to help students such as Goldie develop a strong base of knowledge that they can apply to help species in their own local regions. We wish Dr. Phillips well!

And finally, I'd like to make a personal request that you mark your calendars for the 2016 Giving Challenge, scheduled for Sept. 20-21, and consider making a donation to the Sarasota Dolphin Research Program during that time. The annual Giving Challenge is a 24-hour event sponsored by the Community Foundation of Sarasota County, with support strengthened by The Patterson Foundation, the Manatee Community Foundation, John S. and James L. Knight Foundation, William G. and Marie Selby Foundation and the Herald-Tribune Media Group.

During the Challenge, donations made to participating nonprofit organizations -- such as the Sarasota Dolphin Research Program -- may be eligible to receive matching funds from the Patterson Foundation. Here's how it works:

• All donations of $25 to $100 will be matched 1 to 1 --  a $50 donation becomes $100.
• Donations of $25 to $100 from new donors will be matched 2 to 1 -- a $50 donation becomes $150. You're considered a new donor if you did not give to SDRP through the Giving Challenge last year.

 

This year, funds we raise through the Giving Challenge will support a new project that I've asked SDRP Postdoctoral Researcher Dr. Reny Tyson to lead. Please page down for details and know that donations made to SDRP truly do support dolphin conservation worldwide -- a subject that has been my personal calling for more than four decades.

As always, I wish you good luck, good health and plenty of dolphin sightings in your future!

 

Randy

You've heard of automated facial recognition software, right? Well, we hope to raise enough funds through our annual Giving Challenge project to develop automated fin recognition software in partnership with another nonprofit organization called Wild Me.

The Giving Challenge offers an opportunity for nonprofit organizations like ours to receive matching grants for donations made between noon on Sept. 20 and noon Sept. 21.

This year, we are dedicating those dollars to a project that will help us work with computer vision experts and software developers to create a new tool that will revolutionize how we track dolphins in Sarasota Bay and beyond.

Right now, we have more than 685,000 images of dolphin fins in our database, including more than 5,500 individuals that have been verified and placed in our photographic catalog that covers the central west coast of Florida. We add thousands of new images to our database each year. That means every time we take a picture of a dolphin, a staff member has to compare it to our catalog to determine whether the animal is known to us or not. For unfamiliar or potentially new dolphins, it's a process that takes about nine hours -- time for one person to review the catalog and verify the animal in the picture and a second person to confirm.

But how many more animals could we identify -- and how many other tasks could we complete -- if we could automate this process? The need for automation is becoming even more pressing as our Gulf-wide collaborative bottlenose dolphin identification catalog called GoMDIS grows. To date, we have received and incorporated 18 catalogs into GoMDIS, including 13,630 dolphins and 22,586 images, with at least one catalog from each U.S. Gulf state as well as Cuba. Based on firm expressions of interest from others, we expect to receive catalogs from at least 15 more sites.

Our goal is to raise at least $7,500 so we can work with Wild Me to support development of a new algorithm in Wildbook that will help us automatically ID dolphins here in Sarasota Bay. The funds will help to support a Ph.D. student in the Department of Computer Science at Rensselaer Polytechnic Institute dedicated to developing this tool. While Wild Me has already supported the development of software that helps research teams track individual animals of other species -- from whale sharks, to manta rays to zebras -- they've been a little "stuck" by dolphins.

"Dolphins are harder because they don't have complex coloration or patterns like zebras or whale sharks or even humpback whale flukes. Instead, we rely on the nicks and notches on dolphin fins and -- because the images we take are often at different angles -- that can be problematic for developing the algorithms to automate the process of reliably identifying individual animals," says SDRP's Dr. Reny Tyson. "But because we have such an extensive photographic catalog of known individual dolphins in Sarasota Bay, we think we can work together to develop, test and refine the algorithms we need for successful dolphin identification."

Once developed, this software will not only help us more quickly ID Sarasota Bay dolphins, it will help our Gulf of Mexico dolphin research partners identify dolphins in their own regions and support similar dolphin research programs around the world -- Wild Me bases its tools on open-source software, which means the technology will be open to our dolphin research colleagues worldwide.

• Save the Date: Giving Challenge Sept. 20-21

Age: 45

Sex: Female