Estimation of Fish Feed Pellet Scattering Distribution From Automatic Feeder By Utilizing Particle Kinematics and Digital Image Processing Fajar Mukharom Darozat
Institut Teknololgi Bandung, PT. Multidaya Teknologi Nusantara
Abstract
Measuring the distribution of feed pellets throwed by an automatic feeder manually is technically difficult and requires a lot of energy and costs. This research aims to create an estimator of feed scattering distribution using kinematics of motion of the ejected feed particles based on the position of the particles obtained using digital image processing. Six different types of pellets, 150 gr each were used as samples. A screen was placed in front of the pellet thrower. Each pellet that hitting the screen will leave a mark. Picture of the screen with collision marks then captured using smartphone camera. Each image then digitally modified by applying perspective correction and color manipulation to increase contrast between collision marks and background. The coordinate points of collision between the pellet and the screen are mapped using blob detection technique. By utilizing the coordinates of the point of impact and the speed of the feed throw, an estimation of pellet location when reaching the surface of the pond can be made.
Keywords: Aquaculture, Digital Image Processing, Particle Kinematics
