论文部分内容阅读
利用高光谱图像技术研究了诊断温室黄瓜病害的方法,以提高诊断的准确性和效率。试验以黄瓜霜霉病、白粉病为研究对象,利用高光谱图像采集系统获取黄瓜病叶的高光谱图像数据,在450～900nm范围内的高光谱图像数据中,选出特征波长下的图像;然后,对该图像进行去除噪声的滤波处理,并提取黄瓜病叶的色度矩纹理特征向量;最后采用支持向量机分类方法对黄瓜病害进行诊断。研究结果表明,采用高光谱图像新技术与线性核函数对黄瓜霜霉病、白粉病的正确诊断率达100%,采用高光谱图像技术可以实现对温室黄瓜病害进行快速、精确的分类诊断。
Using hyperspectral image technology to study the method of diagnosing cucumber diseases in greenhouse, in order to improve the diagnostic accuracy and efficiency. In the experiment, cucumber downy mildew and powdery mildew were taken as research objects. Hyperspectral image acquisition system was used to obtain the hyperspectral image data of cucumber diseased leaves. In the hyperspectral image data of 450 ~ 900nm, the images with characteristic wavelengths were selected. Then, the image is filtered to remove noise and the chromatic moment texture feature vector of cucumber leaf is extracted. At last, the diagnosis of cucumber diseases is carried out by the support vector machine classification method. The results showed that the correct diagnosis rate of cucumber downy mildew and powdery mildew was 100% using hyperspectral imagery and linear kernel function, and hyperspectral image technology could be used to diagnose cucumber diseases in greenhouse rapidly and accurately.