Abstract
The characteristics of abnormal-water-quality-change detection were investigated using image analysis based on an activity decrease in medaka (Oryzias latipes). The activity of the medaka was quantified by analyzing the blocks' number at which the change in the life span of medaka was observed and its duration, and the alarm of accidental water quality change could be accurately sent in a short time for practical use on the basis of an activity decrease in medaka, although the detection time was extended slightly compared with that of the conventional method based on an abnormal reaction of medaka. The alarm condition was investigated for typical contaminants at various concentrations. From the results, this image analysis method was effective, particularly for the abnormal-water-quality-change detection of industrial wastewater containing a high concentration ammonia or phenol and of municipal sewage containing LAS.
