基于近红外漫反射光谱技术判别柿子品种和 贮藏期的研究

Identification of different varieties and storage time of persimmon by near infrared diffuse reflectance spectroscopy

中文关键词:  近红外漫反射光谱  柿子  品种  贮藏期  判别

英文关键词:near infrared diffuse reflectance spectroscopy  persimmon  different varieties  storage time  identification

中文摘要:

      目的 通过近红外漫反射光谱技术建立了柿子不同品种和贮藏期的快速判别方法。方法 实验对贮藏冷库(0±0.5 ℃)20 d、40 d、60 d的磨盘柿和阳丰甜柿的近红外光谱(400~2500 nm)进行平滑、一阶倒数和标准正常化处理(SNV)处理, 采用主成分分析法(PCA)建立判别模型。结果 在全波长范围内, 不同品种定标模型的正确分类率达到100%; 阳丰甜柿不同贮藏期的正确分类率达到97.78%; 磨盘柿不同贮藏期的正确分类率达到98.89%。3个预测模型的累积准确率达到96.67%。结论 通过近红外漫反射光谱技术, 判别不同品种的柿子并预测其贮藏期具有应用价值。

英文摘要:

      Objective A rapid method was developed for identification of different varieties and storage time of persimmon by near infrared diffuse reflectance spectroscopy(NIRS). Methods ‘Mopan’ permissions and ‘Yangfeng’ sweet persimmons were collected and refrigerated at 0±0.5 ℃for 20 d、40 d and 60 d storage, and their spectra were acquired respectively by NIRS. After smooth processing, first derivative and standard normal variate (SNV), principal component analysis(PCA) was used to establish identification models. Results The calibration model for different varieties of persimmon, the correct classification rate was 100%. The calibration model for different storage time of ‘Yangfeng’ sweet persimmon, the correct classification rate was 97.78%. The calibration model for different storage time of mopan persimmon, the correct classification rate was 98.89%. Three prediction models cumulative accuracy rate was 96.67%. Conclusion The qualitative identification of different kinds of persimmon and their storage time by near infrared diffuse reflectance spectroscopy has application value.