Breeding pig abnormal state detection method and device based on PSO-SVM

Abstract

The invention discloses a breeding pig abnormal state detection method and device based on PSO-SVM, which can effectively detect suspected sick pigs and precisely recognize breeding pig oestrus behaviors in time, so that the manual observation cost is reduced, and the benefits of a farmer are improved. The method comprises the steps that (1) individual information of breeding pigs is collected, wherein the individual information consists of triaxial acceleration, body temperature and foraging information; (2) a feature matrix is built through the feature extraction of the individual information; (3) on the basis of a prebuilt breeding pig knowledge base, with the feature matrix, whether a breeding pig gets sick or not is judged by a method based on the rules; (4) breeding pig oestrus behaviors are recognized and classified through inputting the feature matrix to a prebuilt breeding pig oestrus behavior recognition model based on a support vector machine, and two outputs of whether oestrus happens or not and oestrus probability are obtained, wherein the parameters of the support vector machine are optimized through a particle swarm optimization when the breeding pig oestrus behavior recognition model is built.

Description

technical field [0001] The invention relates to the field of data processing and analysis of the Internet of Things for livestock and poultry breeding, in particular to a PSO-SVM-based abnormal state detection method and device for breeding pigs. Background technique [0002] In recent years, with the extensive promotion of breeding technology in my country, the scale of breeding pigs has been continuously improved. The health status of breeding pigs is the guarantee of pig production, but at present, most domestic animals still use manual observation to monitor livestock behavior. This not only consumes a lot of time and energy, but also the data observed by manual is highly subjective, which is not conducive to accuracy and stability. , continuous recording; timely and accurate identification of normal estrus of sows is the technical key to obtain the mating time of sows and improve their conception rate, but in actual production, most breeding farms currently only rely on...

AI Technical Summary

Problems solved by technology

[0005] At present, there are few studies on the simultaneous detection of disease and estrus abnormalities in breeding pigs, and most of the research on the detection of disease and estrus in breeding pigs is based on a certain physiolog

Images