用遗传算法快速提取近红外光谱特征区域和特征波长

Methods of Characteristic Wavelength Region and Wavelength Selection Based on Genetic Algorithm

提出了一种遗传区间偏最小二乘法(GA-iPLS),并用该方法快速提取苹果糖度近红外光谱的特征区域,在此基础上采用遗传偏最小二乘法(GA-PLS)提取苹果糖度近红外光谱的特征波长,进行苹果糖度预测。结果表明,整个光谱等分为40个子区间,遗传区间偏最小二乘法能快速寻找出5个特征子区间(第4,6,8,11,18号);在5个特征子区间的基础上用遗传偏最小二乘法继续优化,从中提取44个特征波长。建立在5个特征子区间和44个特征波长上的偏最小二乘法模型精度均优于全光谱偏最小二乘法模型,对预测集的预测相关系数提高了近10%;且模型得到了很大的简化,用于建模的主因子数减少了7个。这些结果表明,用这两种方法不但可以建立简洁、数据运算量少的模型,还可以快速地提取近红外光谱的特征区域和特征波长。

Genetic algorithm interval partial least square （GA-iPLS） and genetic algorithm partial least square （GA-PLS） were proposed to select the characteristic wavelength region and characteristic wavelength of sugar content against apple near-infrared spectra for sugar content prediction. The apple near-infrared spectra data were divided into 40 intervals. Consequently, 5 subsets （No. 4,6,8,11,18） and 362 data points were selected quickly by GA-iPLS, and 44 characteristic wavelengths were selected by GA-PLS based on the 5 subsets. Compared with the whole spectra data model, the GA-iPLS and GA-PLS models could not only improve precision with the coefficients of determination for prediction set improved by 10 %, but also simplify the model with 7 primary factors decreased in the model. With the proposed methods, a concise easily computed model can be built to select the characteristic reigon and wavelength of near-infrared spectra.