A Research of Full-Automatic Experiment Platform Using Machine Vision for Water Quality Monitoring

Japanese medaka (Oryzias latipes) is highly valuable in the field of monitoring the quality of drinking water. But as a result of the limitation of experiment platform and realization technology, the previous study cannot extract the characteristics which can reflect the toxicity of water real-time and accurately. This paper discusses a new experiment platform which can replace the fish tank, adjust the focal length of the camera and measure the swimming speed of medaka automatically. In order to achieve these functions, a precision positioning system using feedback technology, a pair of autofocus mechanism, a new observation method and a set of machine vision algorithm were adopted in this platform. The preliminary test results show that the designed platform and machine vision algorithm for extracting the characteristic information of medaka are effective and feasible. The platform can be widely applied to study some toxic substances' effects on medaka and to assess the degree of risk of the drinking water.