提出了一种利用可见-近红外反射光谱技术对婆婆纳、波斯婆婆纳、直立婆婆纳等3种入侵植物和本地杂草宝盖草的植物辨别方法,可以对外表相似度极高的这4种植物进行有效鉴别。研究在对光谱曲线进行预处理和聚类分析后,随机采用30×4个...提出了一种利用可见-近红外反射光谱技术对婆婆纳、波斯婆婆纳、直立婆婆纳等3种入侵植物和本地杂草宝盖草的植物辨别方法,可以对外表相似度极高的这4种植物进行有效鉴别。研究在对光谱曲线进行预处理和聚类分析后,随机采用30×4个样本作为建模样本,其余的20×4个样本作为预测样本,应用独立软模式法SIMCA(soft independent models of class analogy)进行分类,在显著性水平为5%下,其预测分辨率为78.75%,去除婆婆纳后的预测分辨率为90%。根据变量建模能力(modeling power)值,找到敏感波段496～521,589～626和789～926nm,并将相应的波段的光谱值作为最小二乘的支持向量机LS-SVM(least squares support vector machine)的输入,进行建模预测,并以预测结果作为目标函数值,进行遗传算法GA(genetic algorithm)优化,结果发现,预测分辨率达95.35%,辨识效果好,能快速正确区分外来入侵植物。展开更多
To select the suitable plant species controlling Eupatorium adenophorum in hilly area, ecological adaptability, competitiveness and control efficacy of many replacement plants with economic value in Guizhou Province o...To select the suitable plant species controlling Eupatorium adenophorum in hilly area, ecological adaptability, competitiveness and control efficacy of many replacement plants with economic value in Guizhou Province of China were studied. The results showed that the coverage of paspalum wetsfeteini and Dolichos lablab were 100% and 93%, which increased by 3.70 and 3.44 times compared with Lolium perenne, respectively; their relative crowding coefficient with E. adenophorum were 7.09 and 22.78, which increased by 2.43 and 7.80 times compared with L perenroe, respectively. Using excavation method, the control efficacies of replacement plants were 99.3 % and 96.9%, respectively, while the control efficacy of replacement plants using mowing method was lower than 66.4%. The overwintering rate of P. wetsfeteini in the following year was 95% ; its coverage was still 100% and its control efficacy against E. adenophorum remained over 99%. D. lablab was difficult to survive the winter in the north region beyond 26.2°N, so it could only be applied as annual replacement plant. The coverage of Setaria anceps, Cajanus cajan and other test plants were less than 90%, with poor control efficacy against E. adenophorum.展开更多
文摘提出了一种利用可见-近红外反射光谱技术对婆婆纳、波斯婆婆纳、直立婆婆纳等3种入侵植物和本地杂草宝盖草的植物辨别方法,可以对外表相似度极高的这4种植物进行有效鉴别。研究在对光谱曲线进行预处理和聚类分析后,随机采用30×4个样本作为建模样本,其余的20×4个样本作为预测样本,应用独立软模式法SIMCA(soft independent models of class analogy)进行分类,在显著性水平为5%下,其预测分辨率为78.75%,去除婆婆纳后的预测分辨率为90%。根据变量建模能力(modeling power)值,找到敏感波段496～521,589～626和789～926nm,并将相应的波段的光谱值作为最小二乘的支持向量机LS-SVM(least squares support vector machine)的输入,进行建模预测,并以预测结果作为目标函数值,进行遗传算法GA(genetic algorithm)优化,结果发现,预测分辨率达95.35%,辨识效果好,能快速正确区分外来入侵植物。
基金Supported by Special Fund for Agro-scientific Research in the Public Interest"Research and Demonstration of Comprehensive Prevention Technology against Invasive Plants"(201103027)
文摘To select the suitable plant species controlling Eupatorium adenophorum in hilly area, ecological adaptability, competitiveness and control efficacy of many replacement plants with economic value in Guizhou Province of China were studied. The results showed that the coverage of paspalum wetsfeteini and Dolichos lablab were 100% and 93%, which increased by 3.70 and 3.44 times compared with Lolium perenne, respectively; their relative crowding coefficient with E. adenophorum were 7.09 and 22.78, which increased by 2.43 and 7.80 times compared with L perenroe, respectively. Using excavation method, the control efficacies of replacement plants were 99.3 % and 96.9%, respectively, while the control efficacy of replacement plants using mowing method was lower than 66.4%. The overwintering rate of P. wetsfeteini in the following year was 95% ; its coverage was still 100% and its control efficacy against E. adenophorum remained over 99%. D. lablab was difficult to survive the winter in the north region beyond 26.2°N, so it could only be applied as annual replacement plant. The coverage of Setaria anceps, Cajanus cajan and other test plants were less than 90%, with poor control efficacy against E. adenophorum.
