Design Of Moldy Corn Kernels Sampling Detect System Based On Image Technology

Corn,as an important coarse grain in grain crops,has high edible value and economic value.In the first place,corn has a higher plant protein and heat.About 30% of the world's population regards corn as the main grain.Corn can not only be a direct food food,but also can be processed into other types of food products and condiments.For maize grading detection and evaluation,effective means of efficient utilization of resources can be achieved.At present,the detection of maize quality is mainly based on artificial sorting.However,the weight,size,color and defects of corn are the main criteria for judging the corn quality.Because of the diversity of the judgment standards,there is a low efficiency and high error rate in this process.Therefore,the application of advanced technology to automatic detection and sorting of maize quality is the current development trend.This paper presents a design scheme of corn seed detection and classification system based on genetic algorithm and neural network.Through the embedded technology,the image information collection and transmission function of corn seed is realized,and the information is transmitted to the industrial control server for the classification function of corn seed,and the efficiency of classification and sorting is improved.The main contents of this paper are as follows:1)corn seed quality detection and classification scheme based on embedded technology and web network communication technology is designed.The scheme uses advanced embedded technology to realize the information collection and network transmission of corn sampling image.Based on the RBG model,the method of image processing is proposed,and key information such as hue,size and color ratio is extracted from corn seed image.2)corn seed quality classification model based on genetic algorithm and neural network is proposed,based on the genetic algorithm and neural network.Based on the initial selection of the characteristics of maize classification,the genetic algorithm is introduced to select the characteristics of the primary selection,and the accuracy rate of the classification identification is used as a suitable method.The degree function is used as feedback to guide the gradual optimization of the characteristic quantity,so as to improve the efficiency of analysis and identification of corn seed grading.3)based on the selected characteristic parameters,the neural network algorithm was applied to train the classification standard of corn seeds.The rapid identification and classification of corn seeds are achieved.Through this software system,we can conveniently realize the quick classification function based on the sampled image of corn.