Monitoring and predicting the health condition of fish using image processing technology

The paper analyzes actuality and existing problems of aquiculture in China and presents the distributed control system which uses two personal industrial computers as a spot host computer and a remote supervision and control computer respectively and many S7-224 programmable logic controllers as well as many single chip processors as secondary computers. The system can automatically supervise and control many environmental factors. Especially when fishes are unwell or dead, they will turn side and the white area of their bellies will be seen. According to features then using the technology of digital image processing, situation of fishes' growth in aquafarm can be real-timely supervised and predicted in time. First, the field image shot by video camera was grayed, afterwards, all kinds of noise interference were removed to gained good-quality image, finally the image was statistically calculated and recognized. The result indicates that the system works well.