麻雀搜索算法优化BP算法结合高光谱预测小米米粉糊化特性

doi:10.7506/spkx1002-6630-20210806-074

食品科学 ›› 2022, Vol. 43 ›› Issue (19): 65-70.doi: 10.7506/spkx1002-6630-20210806-074

麻雀搜索算法优化BP算法结合高光谱预测小米米粉糊化特性

王国梁，王文俊，成锴，刘鑫，赵建贵，李洪，郭二虎，李志伟

（1.山西农业大学农业工程学院，山西太谷 030801；2.山西农业大学谷子研究所，山西长治 046000）

出版日期:2022-10-15 发布日期:2022-10-26
基金资助:
山西省重点研发计划项目（201903D211005）；国家现代农业产业技术体系建设专项（CARS-06-13.5-A21）；山西农业大学科技创新基金项目（2017YJ12）；山西省优秀博士来晋工作奖励资金项目（SXYBKY2019018）

Hyperspectral Imaging Combined with Back Propagation Neural Network Optimized by Sparrow Search Algorithm for Predicting Gelatinization Properties of Millet Flour

WANG Guoliang, WANG Wenjun, CHENG Kai, LIU Xin, ZHAO Jiangui, LI Hong, GUO Erhu, LI Zhiwei

(1. College of Agricultural Engineering, Shanxi Agricultural University, Taigu 030801, China;2. Millet Research Institute, Shanxi Agricultural University, Changzhi 046000, China)

Online:2022-10-15 Published:2022-10-26

摘要/Abstract

摘要： 为了实现小米米粉糊化特征指标的批量、快速检测，探索计算机深度学习结合高光谱成像技术在小米米粉糊化特征指标预测方面的应用方法，本研究运用高光谱数据提取、预处理分步运算程序获得小米米粉平均光谱数据，并以该数据矩阵为基础，运用麻雀搜索算法（sparrow search algorithm，SSA）优化误差反向传播（error back propagation，BP）算法进行待测样品糊化特征指标回归、预测。结果表明，光谱数据预处理程序能够标准化并简化光谱数据提取、预处理过程，该程序在粉末及小颗粒样本光谱数据的提取、预处理过程中具有普遍适用性；运用BP算法及SSA优化BP算法对小米米粉糊化各特征指标进行预测，从预测值与测试值间均方误差（mean squared error，MSE）可以看出，各指标MSE均下降，以峰值黏度（peak viscosity，PV）为例，其MSE从0.026 6降为0.017 5，可知运用SSA优化BP算法能够提高小米米粉糊化特征指标预测精度，降低MSE。本研究结论可以为高光谱成像结合计算机深度学习在小米米粉糊化特性预测方面应用提供理论支撑。

关键词: 小米米粉糊化特征指标；高光谱成像；数据预处理；麻雀搜索算法

Abstract: For large-scale rapid detection of the gelatinization parameters of millet flour, a method to predict the gelatinization characteristics of millet flour was explored using hyperspectral imaging combined with deep learning. The average spectral data of millet flour were obtained through successive hyperspectral data feature extraction and pre-processing, and based on the data matrix obtained, a regression model to predict the gelatinization parameters of millet flour samples was developed using a back propagation (BP) neural network optimized by sparrow search algorithm (SSA). The results showed that the spectral data pre-processing program used in this study could standardize and simplify the process of spectral data extraction and pre-processing, and this program was generally applicable to spectral data extraction and pre-processing for powder and fine particle samples. BP algorithm and SSA-optimized BP algorithm were used to predict the gelatinization parameters of millet flour. The mean square error (MSE) between the prediction value and the tested value of each parameter decreased after optimization of BP algorithm, from 0.026 6 to 0.017 5 for peak viscosity. Therefore, the SSA optimized BP algorithm could predict the gelatinization properties of millet flour more accurately. This study can provide theoretical support for the application of hyperspectral imaging coupled with deep learning in the prediction of the gelatinization properties of millet flour.

Key words: gelatinization characteristics of millet flour; hyperspectral imaging; data pre-processing; sparrow search algorithm

中图分类号:

O657.3

王国梁，王文俊，成锴，刘鑫，赵建贵，李洪，郭二虎，李志伟. 麻雀搜索算法优化BP算法结合高光谱预测小米米粉糊化特性[J]. 食品科学, 2022, 43(19): 65-70.

WANG Guoliang, WANG Wenjun, CHENG Kai, LIU Xin, ZHAO Jiangui, LI Hong, GUO Erhu, LI Zhiwei. Hyperspectral Imaging Combined with Back Propagation Neural Network Optimized by Sparrow Search Algorithm for Predicting Gelatinization Properties of Millet Flour[J]. FOOD SCIENCE, 2022, 43(19): 65-70.

[1]	王国梁王文俊成锴刘鑫赵建贵李洪郭二虎李志伟. 基于麻雀搜索优化BP算法小米米粉糊化特性的高光谱预测[J]. 食品科学, 0, (): 0-0.
[2]	乐薇，孔京华，刘诗诗，段凤琪. 基于碳点分子印迹纸芯片法测定金银花绿原酸含量[J]. 食品科学, 2022, 43(16): 253-260.
[3]	关琦颜贤仔余莉莉王翠萍曾佩张欢王纯荣. 基于表面增强拉曼光谱和偶氮偶联反应对组氨酸的灵敏检测[J]. 食品科学, 0, (): 0-0.
[4]	蔡晓郭文川曾硕崇曹梦柯阳尚宏. CPPU对生长期‘徐香’猕猴桃光学参数及内部品质的影响及其关系分析[J]. 食品科学, 0, (): 0-0.
[5]	夏启何天伦黄志轩陈达. 流判别神经网路驱动的奶粉真实性拉曼成像非定向检测方法[J]. 食品科学, 0, (): 0-0.
[6]	邵帅斌，刘美含，石宇晴，郝朝龙，韩宙，张伟，陈达. 基于卷积神经网络的乳粉掺杂物拉曼光谱分类方法[J]. 食品科学, 2022, 43(14): 296-301.
[7]	邵帅斌刘美含石宇晴郝朝龙韩宙张伟陈达. 基于卷积神经网络的乳粉掺杂物拉曼光谱分类方法[J]. 食品科学, 0, (): 0-0.
[8]	翟文磊韦迪哲王蒙. 光催化自清洁表面增强拉曼光谱基底用于食品污染物可循环检测的研究进展[J]. 食品科学, 0, (): 0-0.
[9]	卢春霞. 基于氯血红素/G四聚体DNA酶比色检测苏丹红Ⅲ[J]. 食品科学, 0, (): 0-0.
[10]	曾少甫郭盼章凯倩胡长鹰刘宜奇王志伟. 基于紫外-可见分光光度法测定食品接触橡胶中高锰酸钾消耗量[J]. 食品科学, 0, (): 0-0.
[11]	李爱阳，黄建华. 氢化物发生-电感耦合等离子体串联质谱法分析大米中的无机砷[J]. 食品科学, 2022, 43(6): 331-336.
[12]	李爱阳黄建华. 应用氢化物发生-电感耦合等离子体串联质谱分析大米中的无机砷[J]. 食品科学, 0, (): 0-0.
[13]	夏启黄志轩鲍蕾卜汉萍陈达. 基于自编码收缩神经网络的奶粉掺杂快速拉曼成像检测方法[J]. 食品科学, 0, (): 0-0.
[14]	李爱阳，黄建华. 应用氢化物发生-微波等离子体原子发射光谱分析食用菌中总砷含量[J]. 食品科学, 2021, 42(20): 260-265.
[15]	李爱阳黄建华. 应用氢化物发生-微波等离子体原子发射光谱分析食用菌中总砷含量[J]. 食品科学, 0, (): 0-0.

麻雀搜索算法优化BP算法结合高光谱预测小米米粉糊化特性

Hyperspectral Imaging Combined with Back Propagation Neural Network Optimized by Sparrow Search Algorithm for Predicting Gelatinization Properties of Millet Flour

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价