Research And Implement Of Sows’ Cough Monitoring System Based On Voice Recognition Technology

China is the largest pork producer and consumer country in the world.Pigs’ health is directly related to food security,human health,environmental pollution and other issues.In the process of breeding pigs,sows’ healthily breeding is not only related to their own reproductive rate and productivity,but also piglets’ birth and health.For the current pig large-scale,intensively breeding way,the breeding density of sows is high but the welfare is poor.In addition,management conditions and air quality are not good.The above factors will lead to a lot of sows’ illness and even death.Respiratory system disease with higher prevalence rate is a kind of sows’ diseases,and cough is the main symptom of it.Obtaining the health information of sows,including cough conditions,mainly depends on continuous observation of feeders in traditional cultivation.Human burden is too heavy and subjectivity is strong,so it is easy for feeders to fail to timely discover and handle sick sows because of negligence.Thus,real-time monitoring the sow cough state accurately to discover and control sows’ disease,has practical significance in enhancing the health degree of sows and welfare breeding level.With the rapid development of computer and voice processing technology,speech recognition technology has been widely applied in the field of livestock welfare breeding.Growth and health information of livestock is obtained from their sounds.According to the actual demand in sows breeding and production,a sow cough monitoring system is designed based on speech recognition technology in this paper,which can provide a judgment gist for the early diagnosis of sows’ respiratory diseases.This paper mainly studied the following aspects.(1)Sow sound samples acquisition is studied based on Wireless Multimedia Sensor Networks.And a sound acquisition chamber is built on the basis of existing conditions of the pig farm,to acquire sow sound samples.(2)This paper studies the elementary process of sow sound signal processing,to realize sow sound sample signals’ filtering,framing,endpoint detection and other operations.Besides,the power spectral density of sow sound signals is analyzed after preprocessed.An algorithm to train the power spectral density characteristic of sow sound signals is designed based on objective optimization thought.An sow cough recognition algorithm is designed with the core of improved fuzzy C-means clustering algorithm,to recognize sow sound samples acquired.(3)A sows’ cough monitoring system is designed and realized,including obtaining sows’cough information,developing a WEB monitoring platform and an Android management terminal.It is convenient for breeders to record,query and manage related information of sows.The test is carried on in pig farm of Jiangsu Jurong agriculture museum.Ten secondary hybrid Meishan sows from production date for about one week are selected as subject.Among them,5 is with obvious cough symptom,and another 5 is normal.The study shows,sow sound samples could be obtained quickly,steadily and automatically.Effective classification characteristics are get to identify sow cough and scream sounds.Ignoring individual differences,the overall recognition accuracy of sow cough sounds reaches about 83.4%.The recognition algorithm works effectively.The monitoring system basically meets actual requirements,and provides technical support for the automatic diagnosis of sows’ respiratory system disease.