基于多分类SVM的石榴叶片病害检测方法

Detection of Pomegranate Leaf Disease Based on Multi-Classification SVM

摘要:

石榴是陕西临潼广泛种植的水果作物之一。石榴的生产力由于其果实、茎和叶中各种类型的疾病引起的感染而降低，叶片病害主要由细菌、真菌、病毒等引起。疾病是限制水果生产的一个主要因素，疾病往往难以控制，如果没有准确的疾病诊断，就不能在适当的时间采取适当的控制行动。图像处理技术是植物叶片病害检测和分类中广泛应用的技术之一，旨在利用支持向量机分类技术对石榴叶片病害进行检测和分类。首先用K均值聚类法分割出病变区域，然后提取颜色和纹理特征，最后采用LSVM（线性支持向量机）分类技术对叶片病害类型进行检测。所提出的系统可以成功地检测和分类所检查的疾病，准确率为89.55%。

Abstract:

Pomegranate is one of the widely grown fruit crops in Lintong, Shaanxi. The productivity of pomegranate is reduced by infection caused by various types of diseases in its fruit, stem and leaves, and leaf diseases are mainly caused by bacteria, fungi, viruses and so on. Disease is a major factor limiting fruit yield, and disease is often difficult to control, and without an accurate diagnosis of disease, appropriate control action cannot be taken at the appropriate time. Image processing technology is one of the widely used techniques in plant leaf disease detection and classification, aiming at using support vector machine classification technology to detect and classify pomegranate leaf disease. Firstly, K-means clustering method was used to segment the lesion area domain, and then extract the color and texture features. Finally, the LSVM (Linear Support Vector Machine) classification technique was used to detect the leaf disease types. The proposed system can successfully detect and classify the examined diseases with an accuracy of 89.55%.