采用近红外光谱法结合偏最小二乘法构建蕨菜中总黄酮含量的快速无损测定方法。取蕨菜样品140份，采用傅里叶变换近红外光谱仪采集4 000~11 500 cm^-1波段内近红外光谱，以一阶导数预处理原始光谱，设置主因子数为10，在6 100~7 500 cm^-1和5 400~6 000 cm^-1波段内建模。结果表明：校正集定量分析模型的校正均方根误差（RMSEC）为0.078，交叉验证决定系数（R²）为0.991 9；验证集定量分析模型的预测均方根误差（RMSEP）为0.125，R²为0.984 1，说明所建模型性能较优。分别以定量分析模型和紫外-可见（UV-Vis）分光光度法分析完全外部验证集样品，预测回收率（预测值和测定值比值的百分数）接近100%，说明所建模型的预测准确度较高，可用于蕨菜中总黄酮的快速、准确测定。

A rapid and non-destructive method for the determination of total flavonoids in Pteridium aquilinum was proposed by near infrared spectroscopy combined with partial least square method. The 140 samples of Pteridium aquilinum were taken, and their near infrared spectra were collected in the waveband of 4 000-11 500 cm^-1 by Fourier transform near infrared spectrometer. The original spectra obtained were pretreated by the first-order derivative, and the model was established in the waveband of 6 100-7 500 cm^-1 and 5 400-6 000 cm^-1 with principal factor number of 10. It was shown that RMSEC of the quantitative analysis model established with calibration set was 0.078, and R² was 0.991 9. RMSEP of the quantitative analysis model established with varification set was 0.125, and R² was 0.984 1, indicating that the performance of the established model was superior. The completely external validation set samples were analyzed by the quantitative analysis model and UV-Vis spectrophotometry, giving predicted recoveries (percentage ratios of predicted values to determined values) closed to 100%, indicating that the established model had high prediction accuracy, and could be used for the rapid and accurate determination of total flavonoids in Pteridium aquilinum.