Research On High Quality Breeding Goose System Based On Internet Of Things

At present,only the pedigree breeding method,rather than th e individual breeding method,can be used for geese breeding in China.However,individual breeding method is still based on traditional manual recording method,which is inefficient,artificial strength and low accuracy.Therefore,it is difficult to realize the accurate correspondence between goose and eggs in this way.It is easy to lead to the missed collection and confusion of characterization data of egg production performance as well.Aiming at the urgent need of intelligent and efficient breeding of high-quality breeding geese,the key technology system of goose breeding based on Internet of things was considered,providingtechnical support for intelligent breeding and efficient breeding of breeding geese.The main contents of this paper are as follows:(1)Aiming at the current situation that the laying time of breeding goose is uncertain and it is difficult to monitor the laying behavior and law accurately by manual crouching.The method based on monitoring video tracing was adopted to monitor the laying behavior and law.The egg laying process of breeding geese was clarified.The laying time preference and laying interval of geese were analyzed.The size of breeding goose and goo se egg was counted as well.It provided the design basis for the integrated device of goose individual automatic weighing,identification and egg body marking.(2)According to the demand of production performance statistics of breeding geese,a high-quality breeding system based on Internet of things was built.Meanwhile,an integrated device of goose individual automatic weighing,identification and egg body marking was designed.This system used the radio frequency technology for information intergration.Dynamic weighing method was used to measure the weight of goose in delivery room as well.Through designing the goose egg posture adjustment device and conveyor belt,combined with individual identification results based on RFID,the technology of non-contact inkjet coding was used to realize the accurate marking of goose individual information and laying performance performance.The data edge calculation method of breeding goose breeding system was presented The data transmission mode and MySQL database of goose breeding system were designed,which provided accurate data source for statistical analysis of goose feed conversion rate,egg production and other individual phenotypic indicators.(3)Aiming at the detection requirements of breeding goose and goose egg in the delivery room,the enhancement method of target monitoring data set of breeding goose and goose egg was studied.An improved SSD target detection algorithm was proposed.In the algorithm,the RsNet101 residual network was used to replace the VGG-16 basic Network,the feature pyramid network was used to realize the image fusion of low-level features and high-level features and the goose and goose eggs in the output image were classified by regression finally.(4)A high quality breeding goose system based on Internet of things was developed,the prototype of the integrated device was debugged andthe remote monitoring of goose breeding system was realized.The RFID identification distance was tested.The results showed that the goose identification information can be effectively identified in the delivery room,which verified the feasibility of RFID antenna design.The experiment on laying eggs monitoring performance of breeding goose breeding system was carried out.The results showed that the accuracy of the system was 97.8%and monitoring of egg production data was accurate.The weight data acquisition performance of breeding goose was tested.The results showed that the weighing error of the system was 1.5%and the accuracy was high.