关键词: 麦冬, 图像处理, 主成分分析, 支持向量机, 病害识别

Abstract: The images of three diseases of black spot, anthracnose and leaf blight of Sichuan wheat and winter leaves were taken the research object, the bimodal method, Otsu threshold segmentation method and K-means clustering segmentation algorithm were compared and analyzed on the image of ophiopogon japonicus. The segmentation effect showed that the K-means clustering algorithm combined with the mathematical morphology processing method can meet the segmentation requirements; then, the color, shape and texture information of the lesion image was extracted into the lesion feature vector; then, the variance analysis and principal component analysis method was used to eliminate the characteristic parameters with poor disease characterization ability and reduce the eigenvector dimension to 10 dimensions. Finally, the classifier for disease identification was designed by the support vector machine, which recognition rate reached 90% after the experiment. The method has the advantages of low cost, simple algorithm and high efficiency, and basically meets the requirements of practical applications.

Key words: ophiopogon japonicus, image processing, PCA, SVM, disease recognition