Online Image Acquisition And Counting Of Wheat Stripe Rust Spores

Wheat stripe rust has always been a major threat to the wheat in our country,whose yield and quality has been highly influenced by this disease.Because of the lack of early accurate forecasting technology,the wheat stripe rust was very common and serious.The number of summer spore bacterium source directly influences the incidence and spreading speed of wheat stripe rust.To determinate the number of the summer wheat stripe rust spores in the air in real-time and rapidly,this study researched the real-time online image acquisition technology and microscopic image processing technology.The main research contents and conclusions are as follows:(1)Reprocessing,segmentation and morphology processing of wheat stripe rust microscopic images.It includes image resizing,color space conversion,image segmentation and morphology processing.Several image segmentation methods were compared,including the K-means clustering algorithm,Otsu threshold segmentation algorithm,Canny edge detection method and watershed segmentation method.For different color components and their combination of three color spaces(RGB,L*a*b*,HSV),segmentations were executed using K-means clustering algorithm.The segmentation results were analyzed and the best one(the Canny edge detection method)was found.After the segmentation,a series of morphological image processing was carried out.(2)Recognition and counting of wheat stripe rust spores.For single spores,they have simpler edges and bigger shape factors than adhesion spores.Using the shape factor,the single and adhesion spores were found and separated.Using average area method and corner detection method,the number of spores was got.127 microscopic images of 400 times magnification and 342 microscopic images of 200 times magnification were used to test these methods.The experimental results showed that the average accuracies of the average area method and the corner detection method were no smaller than 84%,and using these two methods we could count the spores accurately.(3)Online obtaining of wheat stripe rust spore microscopic images.Methods of wheat strip rust spore capturing,microscopic image acquisition and image wireless transmission were proposed.By using vacuum pump,51 microcontroller,digital microscope,Intel microcomputer,4G wireless router,solar power supply system,the online image acquisition of wheat stripe rust spores was realized.(4)Designing of the wheat stripe rust spore counting software.Using Matlab compiler,the designing of wheat stripe rust spore counting software was completed.Its function and direction for use are described in detail in this paper.This software can use four different segmentation methods,two different counting methods.Its operation is simple and clear,and it can be run with Matlab.