Title :
Automatic Capturing System of cage traps for medium-sized destructive animals: Reduction Method of Unintentionally Catch
Author :
Shimizu, Yuta ; Yamada, Yasuhiro
Author_Institution :
Graduate School of Engineering, University of Fukui, 910-8507, Japan
Abstract :
In recent years, there has been widespread damage caused by medium-sized animals, such as civets, weasels, and raccoons. Cage traps for catching destructive animals are commercially available, however these traps could unintentionally catch a pet such as a cat. This study proposes two systems in order to achieve a trap that can catch medium-sized destructive animals which is inexpensive and does not unintentionally catch other animals. The first capture system was sensitive to brightness and would therefore operate only at night, thereby making use of the nocturnal properties of the destructive animals. The second capture system used image processing and would determine which destructive animals is in the trap by means of camera images of the animal´s face. The capture system which was sensitive to brightness, succeeded in verifying the capture of several weasels. The capture system which used image processing, successfully identified raccoons and cats using features based on the Histogram of Oriented Gradients (HOG), and was therefore shown to be effective in animal species determination.
Keywords :
Cameras; Computers; Graphics; MATLAB; Positron emission tomography; Support vector machines; Vectors; Cage trap; HOG; Histogram of Oriented Gradients; Mammalian pest control; Unintentionally catch;
Conference_Titel :
SICE Annual Conference (SICE), 2013 Proceedings of
Conference_Location :
Nagoya, Japan