论文部分内容阅读
肉牛的行为是其健康状态的外在表现,目前主要依赖于饲养员的目测判断。大规模肉牛饲养采用人工观察的方法带来繁重人力负担的同时,也会造成误判。为了能自动识别肉牛是否生病,在肉牛的2个角上安装无线传感器节点,通过传感器获取肉牛运动加速度,采用卡尔曼算法对提取的各参数进行分析,可以识别出肉牛的采食行为和行走行为。这种方法识别出采食行为的正确率为78%,行走行为的正确率为62.3%。同时,这种方法也可以应用到其他动物上,对畜牧业发展具有积极意义。
The behavior of beef cattle is the external manifestation of their state of health and currently relies mainly on the visual judgment of breeders. Large-scale beef cattle feeding using manual observation method brings heavy human burden, it will also cause miscarriage of justice. In order to automatically identify whether the beef cattle are sick, wireless sensor nodes are installed on the two corners of the beef cattle, the acceleration of the beef cattle is obtained by the sensor, and the extracted parameters are analyzed by using a Kalman algorithm to identify the feeding behavior and walking behavior of the beef cattle . This method recognizes that the correct rate of eating behavior is 78% and the correct walking behavior is 62.3%. At the same time, this method can also be applied to other animals and has positive significance for the development of animal husbandry.