The Method Of Seafood Quality Safety Early-warning Under The Architecture Of Internet Of Things

Seafood quality safety issues have threatened the lives of people and our country’s economic development seriously. In order to reduce the losses caused by seafood quality safety issues, and reduce the risk of people eating seafood, studying the effective seafood quality safety early-warning method has great theoretical value and practical significance.Signs have existed before seafood quality safety issues occurs, and these signs will be showed by different data indexes, such as seafood sensory indexes, heavy metal content, fishery drugs residues, toxins, and etc. By the Internet of Things technology, the data indicators during farming, processing and transportation of the seafood are monitored in real-time to form surveillance data sets, and then appropriate methods are used to process and analyze the monitored data,and early-warning information can be sent when seafood quality safety issues occurs.The method of seafood quality safety early-warning under the architecture of Internet of Things is the core of the thesis. In the thesis, firstly proposed data filling method based on rough sets for incomplete information system of seafood under the architecture of Internet of Things, and then using attribute reduction the filled complete information system is processed, seafood quality safety evaluation indexes are streamlined, the important reduction indexes are taken as input training samples, and using Radial-Basis Function neural network to predict and realize seafood quality safety early-warning. Finally, seafood quality safety early-warning system is researched based on the architecture of Internet of Things.