近红外光谱快速分析技术及其在动物饲料和产品品质检测中的应用

近红外光谱是近年来发展最快、最引人注目的光谱学技术。主要介绍了近红外光谱技术的基本原理和发展, 包括近红外光谱预处理技术如微分处理、信号平滑等技术的发展和近红外光谱分析模型如多元线性回归、主成分分析、偏最小二乘法和人工神经网络等的发展。综述了国内外近几年来此技术在动物饲料和产品品质检测中的应用。文献调查显示, 近红外光谱分析技术以其快速、无损、不污染环境等诸多优点在国内外饲料和动物产品检测方面得到广泛应用。在饲料分析方面，近红外不仅能用于其常量成分干物质、粗蛋白、粗纤维、粗脂肪等的测定，而且能用于微量成分、有毒有害成分的测定。在动物产品分析方面，该技术已用于禽蛋、牛肉、羊肉、猪肉等的各种物理和化学指标的测定。文中详细给出了已经报道的利用近红外光谱技术测定饲料和动物产品测定指标和光谱处理以及模型建立的情况, 并讨论了近红外光谱快速检测技术的在饲料分析和动物产品分析领域的应用新趋势和局限性。

Near Infrared Reflectance Spectroscopy (NIRS) and Its Application in the Determination for the Quality of Animal Feed and Products

Near-infrared reflectance spectroscopy (NIRS) has been the most rapidly developing and noticeable spectrographic analytical technique in recent years. The determining principle and progresses of near-infrared reflectance spectroscopy are presented briefly. It mainly includes the progresses in pre-processing technique and analyzing model of near-infrared reflectance spectroscopy. Two pre-processing techniques, including differential coefficient-dealt with technique, the signal-smoothing technique, and four analyzing models of near-infrared spectroscopy, including the multiplied lined regression (MLR), principal component analysis (PCA), partial least squares (PLS), and artificial nerve network (ANN). The application of near-infrared reflectance spectroscopy to the first time. The investigation of reviewed papers shows that the near- infrared reflectance spectroscopy is widely applied in feed analysis and animal products analysis because of its rapidness, non-destruction and non-pollution. The near infrared reflectance spectroscopy has been used to determine the feed common ingredient, such as dry matter, crude protein, crude fiber, crude fat and so on, micro-components including amino acid，vitamin, and noxious components, and to determine the physical and chemical properties of animal products which including egg, mutton, beef and pork. Details of the analytical characteristics of feed and animal products described in the reviewed papers are given. New trends and limits to the application of near-infrared reflectance spectroscopy in these fields are also discussed.