Automation Research of the Red sea bream fry Appearance-Form sorting system

2015 Fiscal Year Final Research Report

• Project/Area Number: 25420230
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Intelligent mechanics/Mechanical systems
• Research Institution: Kinki University
• Principal Investigator: KUBOTA Hitoshi 近畿大学, 先端技術総合研究所, 講師 (80551704)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 水産養殖 / 種苗生産 / 画像処理 / 選別

Outline of Final Research Achievements

In this research, I have developed mainly image processing technique of detecting the shape of the mouse and the contour shape of the red sea bream. In the mouse detection, I have applied to the red sea bream the method using Active Appearance Model which is popular for human face recognition. In the contour shape detection, I have applied to the red sea bream the binary processing method, edge detection method and line detection method. The result of our research was as follows. In the mouse detection, the method using Active Appearance Model is weak for extreme deformed mouse and for an improper inclination angle of the face area to the camera. In the contour shape detection,the capability of detecting the deformed fly has been largely enhanced and the tact time for sorting the deformed fly has been highly shortened.

Free Research Field

ロボット工学　画像処理