为了实现水稻病害的自动检测,设计并实现了一种基于支持向量机的水稻纹枯病识别方法. 首先利用R分量和中值滤波进行图像预处理,然后利用改进的图切割方法进行病斑分割,再提取病斑的颜色和纹理特征,最后利用支持向量机方法对水稻纹枯病进行分类识别. 结果表明:识别准确率达到95%,能够满足实际应用的需求. 本研究结果可以为水稻病害的自动识别提供参考依据.%Recognition method of rice sheath blight based on SVM was presented for the purpose of achieving the auto -matic detection of the rice diseases .Firstly, R component and median filter are used for image pre-processing .Second-ly, the improved graph cut method is used to segment the lesion .Thirdly, the color and texture features of lesions are ex-tracted .Finally , the rice sheath blight are classified by support vector machine .The results show that the first two meth-ods are more suitable for the evaluation of segmentation of crop disease images in the four methods .The results show that the recognition accuracy rate reaches 95%, which meet the needs of practical applications .The results of the paper lay a foundation for realization of the automatic diagnosis of rice diseases .
展开▼