基于光谱技术和BP神经网络的苹果汁中柠檬黄检测与分析  

Detection and Analysis of Lemon Yellow in Apple Juice Based on Spectroscopic Technique and BP Neural Network

作　　者：孟德龙 顾慈勇 刘振鲁 MENG De-long;GU Ci-yong;LIU Zhen-lu(College of Physics,Nanjing University of Aeronautics and Astronautics,Nanjing 210016;College of Physics and Electronic Information Engineering,Guilin University of Technology,Guilin 541004)

机构地区：[1]南京航空航天大学物理学院,南京210016 [2]桂林理工大学物理与电子信息工程学院,桂林541004

出　　处：《电光系统》2024年第2期10-13,共4页Electronic and Electro-optical Systems

摘　　要：合成色素的大量使用给食品安全和人们的身体健康带来了严重的危害,因此,建立快速有效的合成色素检测方法具有重要意义。在文章中,提出了一种基于紫外可见光谱和反向传播(BackPropagation,BP)神经网络快速检测苹果汁中柠檬黄的方法。首先,检测了苹果汁中不同浓度柠檬黄的紫外可见光谱;然后,采用Savitzky-Golay(S-G)平滑消除了光谱数据的噪声;最后,建立了基于BP神经网络的预测模型。同时,采用粒子群优化(Particle SwarmOptimization,PSO)算法对BP神经网络模型进行了优化与分析。结果表明,采用紫外可见光谱结合PSO-BP神经网络模型方法有效实现了苹果汁中柠檬黄的快速检测。文中研究为食品中合成色素的检测提供了一种简单有效的方法。The extensive use of synthetic pigments has posed serious hazards to food safety and human health.Therefore,it is important to establish a rapid and effective detection method for synthetic pigments.In this pa-per,a method for rapid detection of lemon yellow in apple juice based on ultraviolet-visible spectroscopy and back propagation(BP)neural network is proposed.First,the ultraviolet-visible spectra of different concentra-tions of lemon yellow in apple juice are detected.Then,the noise of the spectral data is eliminated using Savitz-ky-Golay(S-G)smoothing.Finally,a prediction model based on the BP neural network model is developed.Meanwhile,the BP neural network model is optimized and analyzed using the particle swarm optimization(PSO)algorithm.The results show that the UV-visible spectroscopy combined with PSO-BP neural network model method effectively realizes the rapid detection of lemon yellow in apple juice.This study provides a sim-ple and effective method for the detection of synthetic pigments in food.

关 键 词：紫外可见光谱 BP神经网络 粒子群优化 食品安全 

分 类 号：TS202[轻工技术与工程—食品科学]