Research on Sunflower Leaf Disease Diagnosis Based on Imaging Identification

With computer imaging identification technology as the major research means and sunflower leaf disease as the object of study, it is to comprehensively use the knowledge in the aspects of image processing, pattern recognition and plant pathology, etc to diagnose for sunflower leaf disease. Three sunflower diseases(sclerotiniose，rust disease and bacterial leaf spot) are chosen as the research object . In the research to accurately separate leaf speck image, it combines separation methods of OTSU and region growing method; further to extract speck color and textural features and take those two feature elements as input condition of sunflower leaf disease diagnosis; finally to use support vector machine fitting for small sample classification realizes sunflower leaf disease diagnosis identification.