Multi-source Information Comprehensive Monitoring Of Pigs' Abnormal Behavior Based On Fuzzy Reasoning

In the process of livestock and poultry breeding,the abnormal behavior of pigs is common and varied.Through the artificial observation and recording of abnormal behavior of pigs,it takes a lot of manpower and material resources,and it has certain subjective human factors that cannot meet the current rapid development of pig raising,which is large-scale and intensive.In the 21 st century,digital livestock and poultry breeding technology is gradually applied to pig breeding.Due to the complexity of digital monitoring of pigs,it is very difficult to realize the accurate monitoring of abnormal behavior of pigs by a single monitoring method.The four kinds of monitoring techniques including machine vision,acoustic,ultrasonic and micro-inertial sensors are described to comprehensively monitor the behavior of pigs from different perspectives in this paper.Aiming at the complex uncertainties in the pig digital information,the abnormal behavior of pigs is judged by the four monitoring methods based on their own algorithms respectively,and then the interval-valued fuzzy reasoning technique based on the comprehensive weighted similarity is used to judge the abnormal behavior of pigs comprehensively.Such digital monitoring and judging the abnormal behavior of pigs can find abnormalities in pig breeding as early as possible,which can help to improve economic and promote the upgrade conversion of the monitoring of pigs from artificial to digitalization and intelligence.The main contents of this paper are:(1)The research background and significance of the judgment of abnormal behavior of pigs in livestock and poultry breeding are expounded,the current situation and development trend of overseas and domestic livestock and poultry breeding as well as the main existing problems are summarized,which bring forward the significance and necessity of comprehensive judgment of digital monitoring of pigs' behavior.(2)Due to the complexity of digital monitoring of pigs,the four kinds of monitoring techniques including video,audio,ultrasonic and micro-inertial sensors,are described to comprehensively monitor the behavior of pigs from different perspectives in this paper,then the abnormal behavior of pigs is judged by their own algorithms.(3)The computer-monitored system based on four monitoring methods is designed and implemented in this paper.In the host computer,establish a convenient,intuitive and friendly man-machine interface,real-time display,record,storage of different monitoring methods of detection data,trends and results of the treatment to achieve the upgrade conversion of the monitoring of pigs from artificial to digitalization and intelligence.At the same time,it also provides a basis and guarantee for the design of monitoring system of other automation industries in the follow-up.(4)Aiming at the complex uncertainties in the pig digital information,the abnormal behavior of pigs is judged by the four monitoring methods based on their own algorithms respectively,the rules base of pigs' abnormal behavior based on fuzzy reasoning was established based on the breeding experience of breeder.The interval-valued fuzzy reasoning technique based on the comprehensive weighted similarity is used to judge the abnormal behavior of pigs,which can find abnormalities in pig breeding as early as possible and abnormal warning information,assist the breeder's decision-making,and lay the foundation for the intelligent judgment of abnormal behavior of pigs.