Assessing the taxonomy of
fish is important in order to manage fish
populations, regulate fisheries, and remove the
exotic invasive species. Automating this process
saves valuable resources of time, money, and
manpower. Current methods for automatic fish
monitoring rely on a human expert to design
features necessary for classifying fish into a
taxonomy. This work uses a method using
Evolution-COnstructed (ECO) features to
automatically find features that can be used to
classify fish into a taxonomy. Rather than relying
on human experts to build features sets to tune
their parameters, our method uses simulated
evolution to construct series of transforms that
convert the input signal of raw pixels of fish
images into high quality features. The
effectiveness of ECO features is shown on a
dataset of four fish species where using five-fold
cross validation an average classification rate of
99.4% is achieved. Although we used four fish
species to prove feasibility, this method can be
easily adapted to new fauna and circumstances.
Fourth place in the Michigan State's Great Lakes Invasive Carp Challenge.
Michigan Department of Natural Resources
||Dr. Dong Zhang, Sun Yat-sen University, Guangzhou,
Lillywhite and Beau Tippetts
- D. Zhang, D.J. Lee, M. Zhang, B.J.
Tippetts, and K.D. Lilywhite, "Object Recognition Algorithm for the Automatic Identification and Removal of Invasive Fish, “ Biosystems and Engineering, vol. 145, p. 65-75, May 2016.
- D. Zhang, K.D. Lillywhite, D.J. Lee, and B.J.
Tippetts, “Automated Fish Taxonomy using
Evolution-COnstructed Features for
invasive species removal,” Pattern Analysis and
Applications, vol. 18/2. p. 451-459, May 2015.
Lillywhite and D.J. Lee, “Automated Fish
Taxonomy using Evolution-COnstructed
Features,” Lecture Notes in Computer
Science (LNCS), Part I, LNCS 6938, p. 541-550,
International Symposium on Visual Computing
(ISVC), Las Vegas, NV, U.S.A., September
image to view.)