fbpx

WCA Partner publishes two important research papers

Home » News » WCA Partner publishes two important research papers

HARRY ECKMAN

Chief Executive Officer

ELIZABETH CUEVAS ZIMBRÓN

Whale Heritage Site Project Manager

MIKI TILLETT

Communications Manager

PATICE TALAUE

Certification Manager

STEFF EATON

Operations Manager

ANDREW SCOON

Sussex Dolphin Project, Project Support Officer

THEA TAYLOR

Sussex Dolphin Project, Lead

DYLAN WALKER

Senior Adviser - Whale Heritage Sites

JEAN-MICHEL COUSTEAU

HONORARY PRESIDENT

IAN LEWIS

Trustee, Life College, UK

ROGER MANN

Trustee, Individual Partners

SUZANNE ROGERS

Trustee, Change for Animals Foundation, UK

WCA Partner Jonian Dolphin Conservation has recently published two important scientific papers related to Risso’s dolphins and Artificial Intelligence.

The first, published in November 2018 in Scientific Reports, is entitled DolFin: an innovative digital platform for studying Risso’s dolphin in the Northern Ionian Sea (North-eastern Central Mediterranean) by Maglietta et al (2018).

More recently the team published A SIFT-based software system for the photo-identification of the Risso’s dolphin (Reno et al. 2019) in Ecological Informatics.

Both papers present the results of using fully automated tools for identifying individual Risso’s dolphins from images of their dorsal fins. Photo-identification has long been used as a non-invasive technique in studying whales and dolphins. It is based on the assumption that a single individual can be recognised by unique markings (such as scratches, scars or notches) in multiple photos captured at different times. Researchers use this information to understand site fidelity, habitat use, movement in both space and over time, as well as looking at population dynamics and to calculate abundance.

However most current systems for processing and analysing photo-identification images are still manual or semi-automated, requiring huge effort and time to analyse. The innovative system described in these papers presents a computer vision (enabling computers to see, identify and process images in the same way that human vision does) approach for the identification of Risso’s dolphins. The results indicate the approach can efficiently assist researchers in collecting and analysing large amounts of data.

The DolFin platform contains the digital catalogue of Risso’s dolphin sightings data in the Gulf of Taranto and the automated photo-ID algorithm. It is accessible through a web interface and facilitates exchange among scientists. The study has provided unprecedented insights into the spatial distribution of Risso’s dolphin in the Northern Ionian Sea.

This work has been realised between Jonian Dolphin Conservation, University of Bari Department of Biology and the National Council of Researches.

Author:

More Posts