中文摘要：

利用高光谱成像技术实现快速估计农田大棚EVA棚膜拉伸强度。采用高光谱仪获取EVA棚膜波段范围在978.37~1 676.30 nm之间的特征光谱曲线,利用拉伸强度分析仪获取样本拉伸强度保持率,利用化学计量学方法建立反映样本内部有机分子交联特性的特征光谱曲线与拉伸强度保持率之间的耦合模型,根据建立的化学计量学模型对未知的EVA棚膜拉伸强度进行估计,其中采用支持向量机（SVM）回归方法获得最优预报效果,预报相关系数平方（r2）的百分比达到86.63%,预报均方根误差（RMSE）为1.049。结果表明,利用高光谱技术结合化学计量学方法用于快速估计农田大棚棚膜拉伸强度是可行的。

英文摘要：

A novel method based on the hyperspectral imaging techniques is introduced to evaluate the tensile strength rate of agricultural greenhouse EVA films. The hyperspectral imaging equipment is used to collect the characterisitc spectral curves of EVA films. The gathering band region is between 978.37 nm and 1 676.30 nm. The tensile strength rate of EVA film is measued by using the tensile strength analysis meter. A correlation model between the characterisitc spectral curves which reflect the cross-linking properties of the inner organic molecules and the tensile strength rate is set up by using the Chemometrics. Then, the tensile strength rate of unknown EVA films can be estimated by using the established correlation models.The support vector machine（SVM） regression model achieves the optimal forecasting performance, where the percentage of correlation coefficient（r2） reaches 86.63% and the root mean square error（RMSE） declines to 1.049. The results indicated that the methods of combination of the Chemometrics with the hyperspectral imaging techniques can be used to rapidly determine the tensile strength rate of agricultural greenhouse EVA films.