Design And Implementation Of Wheat Breeding Assistant System

China has the highest total wheat production and consumption in the world.Our wheat has made great achievements in the past few decades,ensuring the safety of our national rations.With the development of information technology and artificial intelligence technology,wheat breeding assisted by technology has become a trend.(1)This paper designs and develops a wheat breeding assistant system to assist related breeding experts to carry out wheat breeding and cultivate excellent wheat varieties.In this paper,the system function modules are designed in detail through demand analysis,and the system is developed according to the function modules of demand analysis and design.The system adopts B/S architecture,Python language,and the database uses MySQL 6.0.The main work of this paper includes the following three aspects.Wheat breeding database construction.Data collection of wheat varieties is the first step,so it is designed for data collection.There are national and provincial wheat variety data,There are two main ways to collect data,one is to capture the network and grab the wheat varieties on specific websites,the other is to import them in a standardized way according to the breeding data accumulated by the Henan Institute of Science and Technology wheat breeding center all the year round.For the web page,this paper uses Scrapy framework,according to the characteristics of web pages,designed a crawler,Then the Climbed data are cleaned,filtered,and then used the word segmenter for segmentation According to the characteristics of breeding data,a custom dictionary was added to improve the effect of word segmentation.For the second method of importing data tables,using Excel tables as templates,data can be imported by entering variety information based on the fields set by the template.The database of wheat varieties breeding is constructed to provide unified information storage and management functions for wheat variety data,which is convenient for wheat breeding researchers to query,manage,and analyze the use data..(2)Wheat breeding model construction based on BP neural network.For the construction of wheat breeding model,it is difficult to construct a complete breeding model because of the complexity of wheat breeding model construction.This paper adopts the method of divide and conquer to build multiple wheat breeding models,such as the yield resistance model and the cold resistance model are used to process wheat breeding data from many perspectives.The user needs to predict a certain shape of the wheat,and only needs to select the corresponding model.For the model construction,this paper mainly uses Bayes,Support Vector Machine and BP neural network.The three methods are trained and tested respectively.The results show that the model constructed with BP neural network has the highest accuracy rate and the best effect.(3)Designed and implemented a wheat assistant breeding system.According to the functional requirements of the system,the design and implementation functions mainly include wheat breeding data collection,wheat breeding data pre-processing,wheat breeding data import,data management,wheat variety pedigree display,wheat variety pedigree query,genetic relationship map,breeding model,breeding variety recommendation.Test results show that the system development requirements are met,the system performance is tested,the system performance is good,and the system development has achieved the expected goal successfully.