论文部分内容阅读
[目的]对蚕豆病害叶进行鉴别研究。[方法]用傅里叶变换红外光谱技术结合统计分析对蚕豆锈病、茎基腐病、轮纹病、黄化卷叶病和正常叶片进行鉴别研究。[结果]病害叶片和正常叶片的红外图谱相似,仅有几个吸收强度比存在差异;对光谱进行二阶导数分析发现,病害叶片和正常叶片的二阶导数光谱在1200～700cm-1范围差异明显,对该区域内的光谱数据进行相关分析、主成分分析和聚类分析,结果显示,正常叶片之间、同种病害样品之间相关系数都在0.928以上,而病害叶片和正常叶片以及不同病害样品之间的相关系数降低;45个样品聚类分析和主成分分析正确率分别为73.3%和82.2%。[结论]傅里叶变换红外光谱技术具有快速、无损、准确等优点,有望为蚕豆病虫害诊断提供新的方法与途径。
[Objective] The research aimed to identify the leaves of Vicia faba. [Method] The differential diagnosis of broad bean rust, stem rot, ring rot, yellow curl leaf and normal leaf were studied by Fourier transform infrared spectroscopy and statistical analysis. [Result] The infrared spectra of diseased leaves and normal leaves were similar, and only a few absorption intensity ratios were different. The second derivative analysis of the spectra showed that the second derivative spectra of diseased leaves and normal leaves were different in the range of 1200 ~ 700cm-1 Obviously, the correlation analysis, principal component analysis and cluster analysis of the spectral data in the region showed that the correlation coefficient between the normal leaves and the same kinds of diseases was above 0.928, while the diseased leaves and the normal leaves were different The correlation coefficient between disease samples was decreased. The correct rates of cluster analysis and principal component analysis of 45 samples were 73.3% and 82.2% respectively. [Conclusion] Fourier transform infrared spectroscopy was rapid, nondestructive and accurate. It is expected to provide a new method and method for the diagnosis of Vicia faba.