Study on the Discrimination of Stall-Fed and Free-Range Yak Meat Using Near-Infrared Spectroscopy Combined with the SIMCA Algorithm
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
维普
万方数据
为保护牦牛肉在市场中的独特性和真实性,维护消费者的合法权益,本研究以近红外光谱技术(near infrared spectroscopy,NIR)对不同饲养模式的牦牛肉样品在全波近红外Whole-NIR(400～2 500 nm)光谱范围内结合簇类独立软模式法(Soft Independent Modeling of Class Analogies,SIMCA)建立模型来判别牦牛肉舍饲和放养来源的可行性.结果表明:在Whole-NIR(400～2 500 nm)光谱范围内原始光谱差异不明显,当因子数为5 时预测误差平方加和(PRESS)趋于平稳,最终无限趋近于0,NIR结合SIMCA模式识别方法得到Q-T2 分布图,类与类之间界限明显,能够将牦牛肉样品按照不同饲养模式分开聚类,模型性能优良,预测准确率高达 100%.综上所述,利用NIR结合SIMCA可实现对不同饲养模式牦牛肉真实性进行鉴别的目的.
To protect the uniqueness and authenticity of yak meat in the market and safeguard the legitimate rights and interests of consumers,this study employs near-infrared spectroscopy(NIR)technology to analyze yak meat samples from different feeding regimes within the Whole-NIR(400～2 500 nm)spectral range,in conjunc-tion with Soft Independent Modeling of Class Analogies(SIMCA),to establish a model for discriminating between stall-fed and free-range yak meat sources.The results indicate that the original spectral differences within the Whole-NIR(400～2 500 nm)spectral range are not significant,but when the number of factors is set to 5,the predictive error sum of squares(PRESS)tends to stabilize and eventually approaches zero.The combination of Near-Infrared Spectroscopy(NIR)with SIMCA pattern recognition method yields a Q-T2 distribution plot,where clear boundaries between classes are evident.This enables the clustering of yak meat samples based on dif-ferent feeding modes,with excellent model performance and a high prediction accuracy of 100%.In summary,NIR combined with SIMCA can effectively discriminate the authenticity of yak meat from different feeding regimes.
Yak MeatNear infrared spectroscopySoft independent modeling of class analogiesHusbandry pattern
维普
万方数据
