论文部分内容阅读
介绍了一套基于计算机视觉技术,可以实时自动监测鱼群游动特性的系统,对获取的动态图片序列进行分析,提取出指示生物的行为参数,如平均游动速率、平均游动高度、平均鱼间距及平均转身次数等。并通过对斑马鱼在模拟突发污染胁迫下行为状态的监测分析,指出了利用鱼类的行为反应与污染物的关系可对水质安全状况进行初步评价,对水体污染做出预警。该系统可应用于关键领域对水体的自动连续监测,达到预防突发性污染事故的目的。
A set of system based on computer vision technology that can automatically monitor fish swimming characteristics in real time is introduced. The acquired dynamic picture sequences are analyzed to extract biological behavior parameters such as average swimming velocity, average swimming height, average Fish spacing and the average number of turns and so on. And by monitoring and analyzing the behavior of zebrafish under the simulated sudden pollution stress, it is pointed out that using the relationship between fish ’s behavioral response and pollutants, preliminary assessment of water quality and safety can be carried out to make early warning of water pollution. The system can be applied to automatic and continuous monitoring of water bodies in key areas to achieve the purpose of preventing unexpected pollution accidents.