Design And Implementation Of Corn Nutrition Monitoring System

The rapid development of computer information technology to drive the development and application of information technology in agricultural production,using the digital image processing technology was carried out on the field crops grown fast effective to estimate nutritional status is an important subject in the process of the agricultural informatization.At present through image processing technique to detect crop nutrition research mostly focus on cotton,wheat,fruit,vegetables,and other fields,and are mostly in the laboratory research,under the particular environment for detecting field corn nutrition research is less.In this paper,based on the digital image processing technology,to the environment in the field of corn plant for timely detection,and nutritional status of the collected experimental design of field environment under the three varieties of four nitrogen nutrition level of farm corn leaf images and the corresponding related crop nutrition data,Through the test image for image denoising,image segmentation and morphological processing,processing of corn leaf color information and with the test data of crop nutrition information combining color characteristics of selecting best correlation algorithm,total nitrogen content of the test carried out on the blade at the same time,based on this algorithm and the correlation between content nitrogen of crop maize nutrition estimate model is established.Test is completed in the related model and through accurate rate under the premise of using Java Web related technology was designed and implemented a field can be real-time shooting corn leaf images to estimate nutritional status and gives the corresponding fertilizer management means recommended Web system.In this paper,the main research contents and results are as follows:1.Based on the existing digital image processing technology,the Lab color model,using the method of Otsu the between-cluster variance to the field of corn leaf to unsupervised image segmentation.The segmentation of image morphology open after operation processing,realize the segmentation of the image noise removal and its separation with background environment.The processing method solves the long field environment of crop image segmentation accuracy is not high,target recognition and difficult problems,has high generality and practical application value.2.Based on the segmentation,image processing and conversion of the color channel after extraction of leaf color of normalized HSV feature parameters,and calculate the correlation between corn leaf nutritional status and the best H/(H + S)color features,based on the established models for predicting the relationship between color features and the nutritional status of corn,the estimating model of precision test reached the significant level,the relative error of the model estimated at 91.25%,with high precision.The establishment of the model for real-time nondestructive estimation field corn nitrogen nutrition laid a foundation.3.Based on the field of corn leaf image segmentation method adopted in the experiment,and establish the related color features,SPAD value and nitrogen content and crop nutrition model,this study was designed and implemented a field of corn leaf can be for users to upload images to estimate nutritional status of the Web system.In addition to nutrition estimate function,the system can also correlation analysis was carried out on the relative growth of corn and returned to the user the corresponding means of fertilizer recommendation,the user in the fields of crops can detect agricultural expert knowledge and gain a lot.