论文部分内容阅读
紫外可见多波长透射光谱包含了细菌微生物对光的吸收和前向散射等信息,能反映细菌细胞的组分、大小以及形态等特征,具有细菌种属的特异性,可应用于细菌微生物的快速种类鉴别。以水体中常见细菌微生物为研究对象,实验测量了大肠埃希氏菌、金黄色葡萄球菌、鼠伤寒沙门氏菌以及肺炎克雷伯菌的紫外可见多波长透射光谱,简要分析了不同种类细菌微生物的多波长透射光谱特征;研究了透射光谱与支持向量机多向量分析方法相结合的水体细菌微生物快速识别方法,利用基于网格搜索法的训练集内部交叉验证获取建模所需最佳惩罚因子C和核函数参数g,根据最优参数和LibSVM一对一多分类法建立细菌快速分类鉴别模型。利用不同株实验细菌的透射光谱作为测试集对所建模型进行识别正确率的验证,结果表明,所建立的快速分类鉴别模型可以对选取的大肠埃希氏菌、金黄色葡萄球菌、鼠伤寒沙门氏菌以及肺炎克雷伯菌进行快速种类识别,识别正确率为100%;分类鉴别模型对不同大肠杆菌亚种的测试集识别正确率为100%,证明该模型对细菌属间鉴别具有较好的稳定性。不仅可为饮用水源细菌微生物的快速识别预警提供方法,而且可在生物医学方面作为细菌微生物鉴别的一种简便、快速、准确的手段。
UV-visible multi-wavelength transmission spectroscopy contains information on the absorption and forward scattering of light by bacterial microorganisms and can reflect the composition, size and morphology of the bacterial cells. It has the specificity of bacterial species and can be applied to bacterial microorganisms rapidly Type identification. The common bacteria in the water as the research object, the experimental measurement of Escherichia coli, Staphylococcus aureus, Salmonella typhimurium and Klebsiella pneumoniae UV-visible multi-wavelength transmission spectrum, a brief analysis of different types of bacteria and more microorganisms Wavelength transmission spectroscopy. The rapid identification of bacteria in water by a combination of transmission spectra and support vector machine (Multivariable) multi-vector analysis methods was studied. The best penalty factor C Kernel parameter g, according to the optimal parameters and LibSVM one-to-one multi-tax classification bacteria rapid classification model. The transmission spectra of different strains of bacteria were used as the test set to validate the correctness of the model. The results showed that the rapid differential identification model could be used to detect the selected strains of Escherichia coli, Staphylococcus aureus, Salmonella typhimurium And Klebsiella pneumoniae for rapid species identification, recognition rate was 100%; classification identification model for different E. coli subspecies test set identification accuracy rate of 100%, proves that the model for the identification of bacterial species has better stability Sex. It not only provides a method for rapid identification and early warning of bacterial microorganisms in drinking water sources, but also can be used as a simple, rapid and accurate method for identification of bacterial microorganisms in biomedicine.