Assisting Manual Dolphin Identification by Computer Extraction of Dorsal Ratio A. KREHO, 1 N. KEHTARNAVAZ, 1 B. ARAABI, 1 G. HILLMAN, 2 B. WU ¨ RSIG, 3 and D. WELLER 3 1 Electrical Engineering Department, Texas A & M University, College Station, TX, 2 University of Texas Medical Branch, Galveston, TX, and 3 Marine Mammal Research Center, Texas A & M University at Galveston, Galveston, TX (Received 15 January 1999; accepted 27 August 1999) Abstract—Marine biologists use a measurement called the ‘‘Dorsal Ratio’’ in the process of manual identification of bottlenose dolphins. The dorsal ratio denotes the relative dis- tances of the two largest notches from the tip on the dorsal fin. The manual computation of this ratio is time consuming, labor intensive, and user dependent. This paper presents a computer- assisted system to extract the dorsal ratio for use in identifica- tion of individual animals. The first component of the system consists of active contour modeling where the trailing edge of the dorsal fin is detected. This is followed by a curvature module to find the characteristic fin points: tip and two most prominent notches. Curvature smoothing is performed at vari- ous smoothing scales, and wavelet coefficients are utilized to select an appropriate smoothing scale. The dorsal ratio is then computed from the curvature function at the appropriate smoothing scale. The system was tested using 296 digitized images of dolphins, representing 94 individual dolphins. The results obtained indicate that the computer extracted dorsal ratio can be used in place of the manually extracted dorsal ratio as part of the manual identification process. © 1999 Biomedi- cal Engineering Society. S0090-69649900906-6 Keywords—Photoidentification, Image database, Feature ex- traction, Curve representation INTRODUCTION Early researchers of animal behavior and ecology rec- ognized that aspects of their studies were enhanced by the recognition of individuals. Photoidentification tech- niques are frequently used for identification of whales, dolphins, and porpoises, which usually have prominent and easily recognized markings on their dorsal fins. 24 Biologists commonly use distinctive, naturally occurring variations and notches in the dorsal fins of individual whales, dolphins, and porpoises to identify individuals within a study population. As the number of identified individuals within a population increases, however, the effort of reidentifying known dolphins, and determining new or unknown individuals becomes labor intensive and increasingly prone to error. Bottlenose dolphins Tursi- ops truncatus are particularly well suited to the process of photoidentification, in that the thin posterior edge of the dorsal fin becomes ‘‘notched’’ during interactions with conspecifics, predators, and humans, often resulting in recognizable patterns of notches. 25 Notch patterns once acquired by a dolphin are usually permanent with little change occurring over time. 21 Such patterns are relatively unique, analogous to human fingerprints. Data collection begins when dorsal fin photographs are taken from small research vessels with 35 mm cam- eras using color slide or black and white film. Laboratory photoanalysis begins with the initial sorting and identifi- cation of a collection of negatives from a photographic survey. Only high quality photographs of distinctive dor- sal fin notch patterns are included for analysis. These quality photographs or ‘‘type specimens’’ are then sorted into discrete individual fins. Type specimens are then rear projected and enlarged so that the fin silhouette may be traced in a 10–17 cm frame drawn on white paper. This tracing process results in uniform hand drawn rep- lications tracings of each individual identified in the initial sorting process. The dorsal ratio DR introduced by Defran et al. 9 is then calculated for all fins with two or more notches. 9,20 This ratio denotes the distance be- tween the two largest notches on the fin, divided by the distance from the lower notch to the top of the dorsal fin. As a relative measure the DR is unaffected by the size of the fin when photographed, enlarged or even under mod- erate cases of parallax. Once calculated, the DR is re- corded directly on the tracing, and used to facilitate matching to a collection of photographs that have been organized according to their DR. If a tracing cannot be matched in this way, then all tracings having all DR values are inspected twice. Although labor intensive, this systematic search process ensures that all previously sighted dorsal fin notch patterns, including those which may have changed in a new photograph, will be re- examined. If the tracing is not matched after thorough inspection of all catalogs, then the individual is consid- ered a new sighting. The use of the DR does not produce an exact match, but it greatly reduces the size of the manually searched database. Similar manual approaches for other species such as sperm whales and humpback Annals of Biomedical Engineering, Vol. 27, pp. 830–838, 1999 0090-6964/99/276/830/9/$15.00 Printed in the USA. All rights reserved. Copyright © 1999 Biomedical Engineering Society 830