基于机器视觉的点带石斑鱼异常行为识别方法研究

应用机器视觉技术对点带石斑鱼(Epinephelus malabaricus)异常行为进行识别并及时报警有助于提高点带石斑鱼存活率,减少养殖过程中人力损耗。在养殖水体中氨氮、温度、p H等不变的条件下,通过人工调节水体溶氧浓度以获取点带石斑鱼正常与异常状态下的图像,对采集到的图像使用前景提取、二值化、开运算、中值滤波等方法进行处理后,选取正常状态与异常状态下2张图像进行差分运算,以得到2种状态下鱼口面积差图。利用鱼眼形状获取鱼眼中心坐标,由此剪裁出只包含鱼口的目标图像。根据鱼口面积判断鱼口张、闭状态与时长,若鱼口持续张开2 min,视为鱼类出现异常行为并进行报警。利用50条大小相近的点带石斑鱼对实验方法进行验证,结果表明,实验方法能有效识别点带石斑鱼的异常行为。该方法的提出有利于提高水产养殖效益,促进水产养殖现代化发展。

Study on the identification of abnormal behavior of Epinephelus malabaricus based on computer vision

The application of computer vision technology on identification and warning of Epinephelus malabaricus’ abnormal behavior is helpful to improve the Epinephelus malabaricus’ survival rate, reduce the loss of human resources in the culture process . Under the condition of constant ammonia nitrogen , temperature, pH value, through artificially adjusted dissolved oxygen concentration in culture water , Epinephelus malabaricus’ images under normal and abnormal conditions were obtained .The collected images were processed using foreground extraction , binarization, open operation, median filtering method , then two images under normal condition and abnormal condition were selected for differential operation to get the differential image of mouth area under the two conditions .The fisheye center coordinate was obtained based on the fish eye shape.So that, the image only contained the fish mouth based on the fisheye center coordinates could be cut out .The fish mouth state ( open or close ) and lasting time were judged according to mouth areae . If the fish mouth was open for 2 minutes, it was regarded that the fish appeared abnormal behavior and would give warning.50 Epinephelus malabaricus with uniform size were used to verify this method , and the results indicated that the method could effectively identify the abnormal behavior of Epinephelus malabaricus.In conclusion , the proposed method is beneficial for improving the efficiency of the aquaculture , and promoting the development of modern aquaculture .