病害胁迫是造成小麦减产及危及世界粮食安全的主要因素之一。如何准确区分相似病害并科学诊断病害严重度, 成为国内外研究热点。文中针对中国冬小麦种植区常见的两种真菌疾病——白粉病和条锈病, 采用高光谱成像系统获取两种病害侵染的小麦叶片图谱合一数据, 通过主成分分析法对影像数据进行降维、密度分割法对病害面积进行分割后, 得到识别病斑准确率达到97%; 进一步分析侵染白粉病和条锈病的叶片病斑区域的光谱特征差异, 选择第二主成分图像筛选两种病害的敏感波段, 得到识别白粉病的敏感波段为519、643、696、764、795、813 nm, 条锈病的敏感波段为494、630、637、698、755、805 nm。最后对筛选出的敏感波段建立白粉病和条锈病支持向量机(SVM)判别模型并验证, 得到两种病害的区分精度为92%。综上, 利用高光谱图像协同解析可在叶片尺度实现小麦白粉病和条锈病的有效判别, 这为开发病害区分仪器提供了重要的理论基础。

Disease stress is one of the main factors causing a reduction in wheat production and threatening food security. How to distinguish similar diseases accurately and diagnose disease severity scientifically is becoming a hot topic worldwide. The objective of this study is to discriminate powdery mildew and yellow rust of winter wheat, two common fungal diseases in the Chinese wheat-growing region. In the study, a high-resolution hyperspectral imaging system(ImSpector V10E) was utilized to capture spectral and imagery information of wheat leaves infected by two diseases. The dimensionality reduction of hyperspectral images was done by using principal component analysis(PCA), and with the density slice method, the recognition accuracy for the disease area at leaf level can be 97%. On this basis, the spectral difference of two diseases was analyzed, and 12 disease-sensitive bands were selected in the light of the second principal component(PC-2) images. The bands for powdery mildew were at 519, 643, 696, 764, 795 and 813 nm, while those for yellow rust were at 494, 630, 637, 698, 755 and 805 nm. Furthermore, a support vector machine(SVM) discriminant model was established based on selected sensitive wavebands, and its accuracy reached 92%. The results revealed that the hyperspectra combined with feature extraction of high-resolution imagery could effectively achieve discrimination of powdery mildew and yellow rust at leaf level, which will provide a theoretical foundation for developing a portable recognition device.