计算机视觉在香芋病害检测中的应用研究

The Applied Research of Computer Vision in the Taro Disease Detection

针对传统的人工识别香芋病害的局限性、主观性的特点,基于计算机视觉技术设计了香芋病害识别系统,由病害图像采集装置、图像处理系统、形态特征提取系统和缺陷目标检测系统组成。通过提取香芋叶片的颜色特征和纹理特征,建立香芋病害数据库,并利用支持向量机(SVM)的方法对香芋病害识别。试验结果表明:基于计算机视觉的香芋病害检测系统能够准确识别香芋的病害类别,满足使用要求,能够为后续的香芋病害防治工作提供准确信息。

Aiming at the artificial recognition of taro disease which has such characteristic as limitation and subjectivity.This paper designed the taro disease detection system based on computer vision.The system is constituted of imaging collection system,image processing system,the extraction of morphology characteristic system,and defect detection system.Through extraction color feature and texture feature of taro leaf,database of taro disease is established.The paper uses the support vector machines(SVM)to recognize taro disease.The experiment results show that the taro disease detection system based on computer vision may recognize the taro disease category accurately.It can meet the usage requirement of users and provide accurate information to the follow-up disease control.