Image processing software identifies animal gender and age from footprint

Aug. 20, 2013
Image processing software from WildTrack detects physical footprint characteristics to determine an animal’s gender and age and can also be used for the monitoring and censusing of species.

Image processing software from WildTrack detects physical footprint characteristics to determine an animal’s gender and age.

The imaging process, called the footprint identification technique, or FIT, begins when images of six to eight different animal tracks, which are taken along with a scale such as a ruler, are collected by professional trackers for analysis. The images, which are at a size of 1200 x 1600 or greater, are uploaded onto a laptop and optimized using standard commercial image manipulation software.

From there, GPS coordinates are added and the images of the footprints are loaded into the software and matched against a library of footprints from known wild animals or from captive animals for the purpose of developing the species algorithm. The algorithm then compares the elements of the photographed footprint against this database to determine an effective classification of species, age, and gender. FIT goes beyond just classification, however, as it is also being used for censusing and monitoring of various species, according to WildTrack.

For censusing, FIT uses a canonical pairwise comparison technique (CPCT). For each pair of tracks, a group from each track is formed, and a third group, the reference centroid value which consists of all the footprints in the library other than those belonging to the two test tracks, is added. CPCT computes the two canonical variants for the three groups and inspects the corresponding plot in canonical space of group means and 95% confidence regions. If the confidence regions of the two tracks being tested overlap the tracks, they are from the same animal. If the tracks do not overlap, they are from different animals. Using CPCT, WildTrack has found the accuracy for censusing white rhinos to be between 91-99%.

For monitoring, WildTrack uses a canonical ellipse reduction technique (CERT), where each known individual is represented by a set of footprints and an unknown track is tested against these to see which it matches. JMP Software computes the first two canonical variants for these data and produces the corresponding 2D plot of group means and 95% confidence ellipses. Groups whose confidence ellipses do not overlap with the unknown track are set aside and the procedure is repeated. If the process results in the overlapping of the unknown track with exactly one group, the group’s identity is assigned to the track. Otherwise, the track remains unclassified. WildTrack has found that the accuracy of CERT for monitoring white rhino to be between 95-98% accuracy.

FIT is being utilized to track a variety of animals in different habitats, including tigers in Russia, tapir in South America, and polar bears in the Canadian province of Nunavut, according to MIT Technology Review.

View more information on WildTrack.

Also check out:
Thermal imaging system protects whales from underwater noise pollution
Vision-enabled robot tracks Great Whites for Shark Week special
Crabster robot will explore shipwrecks below the sea

Share your vision-related news by contacting James Carroll, Senior Web Editor, Vision Systems Design

To receive news like this in your inbox,
click here.

Join our LinkedIn group | Like us on Facebook | Follow us on Twitter | Check us out on Google +

About the Author

James Carroll

Former VSD Editor James Carroll joined the team 2013.  Carroll covered machine vision and imaging from numerous angles, including application stories, industry news, market updates, and new products. In addition to writing and editing articles, Carroll managed the Innovators Awards program and webcasts.

Voice Your Opinion

To join the conversation, and become an exclusive member of Vision Systems Design, create an account today!