Rapid determination of farinograph parameters of wheat flour using data fusion and a forward interval variable selection algorithm,Analytical Methods

当前位置： X-MOL 学术 › Anal. Methods › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Rapid determination of farinograph parameters of wheat flour using data fusion and a forward interval variable selection algorithm
Analytical Methods ( IF 2.7 ) Pub Date : 2017-10-19 00:00:00 , DOI: 10.1039/c7ay02065a
Jia Chen _{1,

2,

3,

4} , Fayin Ye _{1,

2,

3,

4} , Guohua Zhao _{1,

2,

3,

4,

5}

Affiliation

Farinograph tests are used to predict the functional properties and quality of wheat flour. However, these tests are time-consuming and labor-intensive. Conventional rapid determination methods based on near-infrared (NIR) spectra showed a limited ability to predict farinograph parameters. The potential of combining NIR and mid-infrared (MIR) spectral regions to predict wheat flour farinograph quality properties (water absorption, dough development time, dough stability, and degree of softening) was investigated. Partial least squares models based on NIR, MIR and fused spectra were calibrated and compared. Two data fusion strategies (low- and mid-level) have been applied to take advantage of the synergistic effect of information obtained from MIR and NIR. Low-level data fusion models showed inferior performance compared to the corresponding MIR and NIR models, whereas mid-level data fusion models combined with a forward interval variable selection algorithm were validated to show good performance. Fusion of the previously selected variables from MIR and NIR spectra improved the prediction accuracy of farinograph parameters, which indicates the superiority of the forward interval variable selection algorithm that will be helpful for the cereal and baking industries.

中文翻译：

基于数据融合和前向区间变量选择算法的小麦面粉粉质参数快速测定

粉质仪测试用于预测小麦粉的功能特性和质量。但是，这些测试既费时又费力。基于近红外（NIR）光谱的常规快速确定方法显示了预测粉质仪参数的能力有限。研究了结合NIR和中红外（MIR）光谱区域预测面粉粉质仪质量特性（吸水率，面团发育时间，面团稳定性和软化程度）的潜力。校准并比较了基于NIR，MIR和融合光谱的偏最小二乘模型。已应用两种数据融合策略（低级和中级）来利用从MIR和NIR获得的信息的协同效应。与相应的MIR和NIR模型相比，低级数据融合模型显示出较差的性能，而经验证的中级数据融合模型与前向间隔变量选择算法相结合，则显示出良好的性能。融合了从MIR和NIR光谱中先前选择的变量，提高了粉质仪参数的预测准确性，这表明前向区间变量选择算法的优越性将对谷物和烘焙行业有所帮助。

更新日期：2017-11-23

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11