基于可见/近红外光谱的乙烯利催熟西瓜与正常成熟西瓜分类试验研究

Study on Classification of Ethylene Treated and Non-Ethylene Treated Watermelons by Visible/Near Infrared Spectroscopy

针对国内常有瓜农采摘远离成熟期的西瓜,采用高浓度乙烯利处理,诱导其快速成熟的现象,对未经乙烯利处理和经乙烯利处理的西瓜进行了可见/近红外漫透射光谱的分类试验研究,在判别分析中,提出一种简单的透过率之比判别分析法,对未经乙烯利处理样品的误判率为32.5%,经乙烯利处理样品的误判率为20%;采用马氏距离判别分析和偏最小二乘法判别分析都可以取得理想的判别结果,一阶微分光谱经Norris微分滤波处理后,利用马氏距离判别,判别结果为未经乙烯利处理样品校正集误判率1.67%,经乙烯利处理样品校正集、预测集均没有出现误判的情况;二阶微分光谱利用偏最小二乘法判别,没有误判情况发生。在使用这两种判别方法进行判别分析时,光谱预处理方法影响判别效果,应根据采用的判别方法决定所需采用的光谱预处理方法。

According to the fact that farmers often picked unripe watermelon and treated them with high concentration ethylene to quicken ripeness, classification experiments on the two classes of watermelon mentioned above were conducted based on the Vis/NIR spectroscopy diffuse transmittance technique. In the discriminant analysis, a method to classify them by diffuse transmittance ration at two wavelengths was adopted to discriminate them. Result of mistake ratio 32.5% for samples without ethylene treatment and 20% for ethylene treatment samples indicated that this method could discriminate the two classes of watermelons roughly. Mahalanobis distance and partial least square methods were also used here for discriminant analysis and satisfied results were obtained. The first derivative spectra with Norris derivative filtering of samples without being ethylene-treated using Mahalanobis distance discriminant analysis got the result of mistake ratio 1.67% for calibration set, no mistake for prediction set and no mistake for samples being ethylene treated. No mistake took place for the second derivative spectra using partial least square method. In discriminant analysis, spectral data pretreatment methods influence the discriminant results and it should be selected according to the analysis methods.