【摘要】 采用近红外高光谱成像技术结合化学计量学方法建立注胶肉的快速无损检测模型。首先通过近红外高光谱成像系统获取含有不同浓度梯度卡拉胶的猪里脊肉高光谱图像,然后提取图像中的光谱数据,使用偏最小二乘法(Partial least square,PLS)探究光谱信息与不同掺假比例卡拉胶之间的定量关系。结果表明全波段光谱(900～1700 nm)所构建的PLS校正集模型均方根误差(Root mean square error,RMSE)为1.74%,预测模型RMSE为3.16%。表明基于全波段所建立的PLS模型具有较优的预测性能。利用连续投影算法(Successive projection algorithm,SPA)筛选获得11个特征波长,并优化全波长PLS模型,将预测集样品带入,以验证模型的预测效果,结果表明SPA算法结合PLS建模方法所建立的模型预测效果更优,预测集相关系数(R_P)为0.93,均方根误差(Root mean square error of prediction,RMSEP)为3.51%,预测偏差(Residual predictive deviation,RPD)为2.66。试验表明利用高光谱成像技术可实现对注胶猪肉的快速无损检测。更多还原

【Abstract】 A rapid and non-destructive testing model of gum-injected meat was established by using near-infrared hyperspectral imaging technique combined with chemometrics method. First,hyperspectral images of gum-injected meat with different concentrations of carrageenan were obtained,and then the spectral data were extracted from the images. The quantitative relationship between spectral information and carrageenan with different adulteration ratios was investigated by partial least squares(PLS). As a result,the root mean square error(RMSE)of the PLS model using the full spectrum for calibration and prediction set was 1.74% and 3.16%,respectively. The results showed the PLS model with full spectrum had high accuracy. The 11 optimal wavelengths were selected from the full spectra by successive projection algorithm(SPA). The new SPA-PLS model was established to simplify the PLS model and the prediction performance was verified. It indicated that the PLS model combined with SPA showed the better predictive performance. The SPA-PLS had correlation coefficient of 0.93,root mean square error of 3.51% and residual predictive deviation(RPD)of 2.66. The experiment demonstrated that it was feasible to conduct rapid and non-destructive detection of gum-injected pork by hyperspectral imaging technology.更多还原