Abstract The positive phototactic and rheotactic innate responses of guppies ( Poecilia reticulata ) were used to induce them to swim through narrow channels so as to render them ready for inspection and sorting by a computer vision system. Over 300 fish of each gender of the red-blond strain were tested. Guppies in a Y channel consistently preferred the uncovered illuminated arm to the darkened arm, irrespective of their relative position. Three types of sorting apparatuses were tested; each consisted of a straight narrow channel, branching into two, three or four arms. All arms were darkened and equipped with lamps to allow their illumination. Male guppies were tested with the two-choice apparatus under two illumination regimes and two temperature ranges, using a 2×2 factorial experimental design. In 100% of the trials males chose to swim through the illuminated arm (correct choice) irrespective of the illumination regime and the temperature, but temperature and the illumination regime affected the time they spent in the Y channel before deciding which arm to enter. When a guppy was tested with the three- or four-arm apparatus, the branching arms were always darkened, and one of them was illuminated (according to a randomly predetermined schedule) when a fish reached the junction point. In these cases, at least 83% of the males chose to swim through the illuminated arm (correct choice). The guppies in the multiple-choice apparatuses spent more time prior to decision making and made more ‘wrong decisions’ than those in the two-choice apparatus. Manipulation of fish movements at our will constitutes a key element in the development of a device for automatically sorting fish by computer vision.
[1]
Laurence T. Kell,et al.
A potential method for the differentiation between haddock fish stocks by computer vision using canonical discriminant analysis
,
1995
.
[2]
R. Abbott.
Induced Aggregation of Pond‐Reared Rainbow Trout (Salmo gairdneri) Through Acoustic Conditioning
,
1972
.
[3]
L.-F F. Pau,et al.
Fish Quality Control by Computer Vision
,
1991
.
[4]
K. Tsukamoto,et al.
Stock enhancement in Japan : Review and perspective
,
1998
.
[5]
Boaz Zion,et al.
Sorting fish by computer vision
,
1999
.
[6]
Norval J. C. Strachan,et al.
Length measurement of fish by computer vision
,
1993
.
[7]
N.J.C. Strachan.
Sea trials of a computer vision based fish species sorting and size grading machine
,
1994
.
[8]
Boaz Zion,et al.
In-vivo fish sorting by computer vision
,
2000
.
[9]
L. F. Pau,et al.
PDL-HM: morphological and syntactic shape classification algorithm
,
2005,
Machine Vision and Applications.
[10]
C. R. Savage,et al.
Underwater fish-video images: Image quality and edge detection techniques
,
1994
.
[11]
N. Liley,et al.
A sexual pheromone in the guppy, Poecilia reticulata (Peters)
,
1979
.
[12]
K. Tsukamoto,et al.
Effect of fish size, handling stresses and training procedure on the swimming behavior of hatchery-reared striped jack: implications for stock enhancement
,
2000
.