论文部分内容阅读
小麦的网腥、印度腥与矮腥黑穗病危害小麦生产与人体健康,是出入境检验检疫的重要对象。该文利用小麦腥黑穗病害显微图像,采用图像分析与识别技术进行了小麦的网腥、印度腥及矮腥3类病害的分类识别。在分离出单个病害孢子图像的基础上,提取了3类病害孢子图像的16个形状和纹理特征,通过分析,从中选择小麦病害孢子的6个典型特征,并分别用最小距离法、BP神经网络和支持向量机分类器对提取的96个小麦腥黑穗病害孢子图像进行了分类试验,结果表明:支持向量机法对小麦腥黑穗病的分类识别能力优于最小距离法和BP神经网络,总体识别率达到82.9%。因此,采用图像分析技术和支持向量机识别方法进行小麦腥黑穗病害诊断的方法具有可行性。
Netting of wheat, Indian fishy and dwarfed smut endanger wheat production and human health, is an important object of entry-exit inspection and quarantine. In this paper, microscopic images of wheat-tipped head smut were used to classify and identify the three diseases of fishweed, fishy and dwarf in wheat using image analysis and recognition technology. Sixteen spore images of three types of diseases were extracted based on the morphological features of six spores and six typical spore morphological characteristics were selected. The minimum distance method, BP neural network And support vector machine (SVM) classifier were used to classify the extracted 96 spike images of wheat head smut. The results showed that the support vector machine method was better than the minimum distance method and BP neural network in the classification of wheat head smut, The overall recognition rate reached 82.9%. Therefore, it is feasible to use the method of image analysis and SVM to diagnose wheat head scarlet disease.