Maize Leaf Disease Identification Technique Study Based On Image Recognition

Crop disease, insect pest and weed affect the output and quality of crop badly. Funded by the National Natural Science Fund(30360047),the paper mainly researched the theories and methods of recognition and diagnosis of main maize disease during maize growing by using computer vision technology, aimed at the problem of low precision and untimely diagnosis about crop disease. The study took common maize leaf diseases as study object and brought forward feasible method to improve diagnosis precision. Thereby it can provide theory support to the correlative research of the auto recognition and diagnosis of crop disease, insect pest and weed.According to the characteristics of maize leaf disease, threshold method was adopted to do image segmentation; area-marking and Freeman link code method were used calculating the num of disease well as wiping off redundancy dots. And then color feature and form feature were calculated and saved in the database. Direct discriminance, naive Bayesian classifier method and fuzzy pattern recognition based on weighted feature were used to recognize five kinds of maize leaf disease. And then it integrated the results of the three methods and got the identification of diagnostic results finally. Software for maize disease recognition was developed by use of Visual C++. It can provide software and technology support for developing crop disease intelligent recognition system. According to the property and demand, the research analyzed the factors affecting the performance of the system. On the basis of considering the ratio of performance and price, the rule of image catching was made.The study made two aspects of progress in the following. First, it realized automatic identification of the common five kinds of maize leaf disease, and the recognition accuracy rate of over 90%. Secondly, Bayesian method and fuzzy pattern recognition method are used in the diagnosis of maize leaf disease. Because it integrated the advantages of the three methods and got the identification of diagnostic results, it improved the accuracy and reliability of recognition.The research applied computer vision technology in maize disease recognition, which extends the application area of machine vision and also provide a use for reference in the field of agriculture technology for machine vision.