运用连续小波变化-支持向量回归模型预测肉制品的各成分含量

摘要

Continual wavelet transform (CWT), as an application direction of the wavelet analysis, is keener to the slight signal change. Near-infrared spectroscopy (NIR) analytical technique is simple, fast and low cost, making neither pollution nor damage to the samples, and can determine many components simultaneously. Support vector machine (SVM) is based on the principle of structural risk minimization, which makes SVM has better generalization ability than other traditional learning machines that are based on the learning principle of empirical risk minimization. In this paper, we use CWT-SVM model to predict meat's component. Compared with Partial Least Squares (PLS) and SVR, more satisfactory results were obtained.%作为小波分析的一个应用方向,连续小波变换对于信号的变化非常灵敏.近红外光谱技术是一种简单,快速,无损,价格低廉的方法,可以进行多组分同时分析.支持向量机基于结构风险最小化原理替代了传统方法中的的经验风险最小化原理,使得它具有更好的泛化能力.把连续小波变换-结合支持向量回归模型用于肉制品的成分测定,取得了令人满意的效果.