Study On Identification Method Of Movement Behaviors Of The Facility Cultured Stud Rams

Animal behavior is an external manifestation of animal's psychological and physiological conditions,which can be used to study the psychological state of animals by analyzing it.Three kinds of movement behaviors,including standing,walking,and running,can reflect their status of health and adaptability to the environment.Therefore,a study on identification methods of the movement behavioral characteristics of the facility breeding rams was conducted,and three kinds of rams' behaviors: standing,walking,and running wer recognised and classified in this thesis.This provided a basis for welfare breeding of rams.Based on MATLAB software,a kind of rams' movement behavioral identification system with an acceleration sensor was designed,in which,an atriaxial acceleration sensor was employed to collect the rams' movement data in real time,and the data was wirelessly transmited.An optimal deployment scheme of the sensor was determined through the experimental analysis,and the classification of movement behaviors were performed using the interval threshold classification algorithm and the BP neural network algorithm.The experimental results show that the data obtained from the the sensor fixed on the back of the ram closing to the front leg is the most stable and the data collection effect is the best.By using the interval threshold classification algorithm and BP neural network algorithm,the rams' movement behaviors were classified and recognized.The recognition accuracy of the two methods is 88.72% and 98.17% respectively,but the BP neural network algorithm has a better recognition effect on the rams' motor behaviors.Therefore,the system can realize the identification of the rams' movement behaviors.