卷积神经网络模型结合近红外光谱对特种乳掺水的鉴别

doi:10.7506/rykxyjs1671-5187-20260205-011

乳业科学与技术 ›› 2026, Vol. 49 ›› Issue (3): 41-47.DOI: 10.7506/rykxyjs1671-5187-20260205-011

• 分析检测 • 上一篇

卷积神经网络模型结合近红外光谱对特种乳掺水的鉴别

邵子祥, 托尔坤・买买提

新疆大学智慧农业学院(研究院),新疆乌鲁木齐 830017

发布日期:2026-07-06
基金资助:
新疆战略人才培养计划一流科技领军人才项目（XJRC-2025-KJ-PY-KJLJ-126）

Identification of Water Adulteration in Non-bovine Milks Using Near-Infrared Spectroscopy Combined with a Convolutional Neural Network Model

SHAO Zixiang, Torkun·MAMET*

To address the increasingly severe issue of water adulteration in the non-bovine milks market, this study proposed a rapid, non-destructive method for identifying multiple types of adulterated non-bovine milks based on portable near-infrared spectroscopy combined with convolutional neural networks (CNN). Spectral data were collected for four types of non-bovine milks, camel, horse, donkey, and yak milks, as well as samples of these milks diluted with water at varying concentrations (2%–80%, V/V). We compared the effects of standard normal variate transformation (SNV) and Savitzky-Golay filter smoothing preprocessing on the performance of partial least squares discriminant analysis (PLS-DA) and CNN models. The results showed that non-bovine milks exhibited significant characteristic absorption of moisture at 6 882 and 10 200 cm?1. The CNN model with SNV preprocessing achieved the best performance, with a classification accuracy of 98.71% on the test set, and its precision and recall were superior to those of traditional PLS-DA model. This model not only accurately distinguished between milks from different species but also exhibited high sensitivity in detecting water adulteration as low as 2%. The present study demonstrates that the discrimination system based on a portable near-infrared spectrometer and CNN can effectively reduce matrix interference in minor species milks.

Online:2026-07-06

摘要/Abstract

摘要： 针对特种乳市场日益严重的掺水欺诈问题，本研究提出一种基于便携式近红外光谱结合卷积神经网络的多种类特种乳掺假快速、无损鉴别方法。采集驼乳、马乳、驴乳及牦牛乳4种典型特种乳及其2%~80%梯度掺水样本的光谱数据，对比标准正态变量变换、Savitzky-Golay滤波平滑两种预处理方式，探究其对偏最小二乘判别分析与卷积神经网络模型鉴别性能的影响。结果表明，特种乳在6882 cm?1及10200 cm?1波段存在显著水分特征吸收峰；经标准正态变量变换预处理的卷积神经网络模型性能最优，测试集分类准确率可达98.71%，精确率与召回率均优于传统偏最小二乘判别分析模型，能够精准区分不同乳种来源，对低至2%的微量掺水比例也具备良好识别能力。该研究构建的便携式近红外光谱与深度学习结合的鉴别体系，可有效消除不同特种乳间的基质干扰，为特种乳掺水无损快速筛查、保障特种乳市场质量安全提供了高效可靠的技术方案。

中图分类号:

TS252.2

邵子祥, 托尔坤・买买提. 卷积神经网络模型结合近红外光谱对特种乳掺水的鉴别[J]. 乳业科学与技术, 2026, 49(3): 41-47 https://doi.org/10.7506/rykxyjs1671-5187-20260205-011

College of Smart Agriculture (Research Institute), Xinjiang University, Ürümqi 830017, China. Identification of Water Adulteration in Non-bovine Milks Using Near-Infrared Spectroscopy Combined with a Convolutional Neural Network Model[J]. Journal of Dairy Science and Technology, 2026, 49(3): 41-47 https://doi.org/10.7506/rykxyjs1671-5187-20260205-011

[1]	刘玉茹，李欣怡，陈晓民，张淑丽，张德喜. 饲料种类与生牛乳豆腥味相关性[J]. 乳业科学与技术, 2025, 48(2): 20-29.
[2]	贾宏信，苏米亚，陈文亮，曲也直. VC与婴幼儿健康及母乳中VC含量研究进展[J]. 乳业科学与技术, 2025, 48(1): 41-45.
[3]	沈雨菲，张艳，同芬芬，王爽爽，胡启胜，徐立清，宋宇轩，葛武鹏. 不同泌乳期绵羊乳理化指标与酸度变化关联度[J]. 乳业科学与技术, 2024, 47(6): 8-14.
[4]	乌兰君, 杨梅, 王满霞, 米书慧, 吴永锋, 武俊瑞, 乌日娜, 刘彪, 岳喜庆. 牛初乳与牛常乳中乳脂肪球膜蛋白质的组成及功能的对比研究[J]. 乳业科学与技术, 2016, 39(5): 7-12.
[5]	王钦博, 杨焱, 齐晓彦. 桑黄乙醇提取物对乳制品有害菌的抑制作用[J]. 乳业科学与技术, 2016, 39(1): 1-4.
[6]	于晓平, 李娜, 王大为. 香菇肽乳饮料的工艺优化[J]. 乳业科学与技术, 2015, 38(5): 16-20.
[7]	张薇, 宋春春, 徐玉娟, 于雷. 米糠蛋白复合乳饮料制备及其乳化稳定性[J]. 乳业科学与技术, 2013, 36(4): 7-10.
[8]	李良玉, 曹荣安, 于伟, 贾鹏禹. 小麦胚乳饮料加工技术[J]. 乳业科学与技术, 2013, 36(4): 11-13.
[9]	赵瑾德，张淑丽，贺凯茹，陈晓民，刘玉茹，张艳良，乌日娜，张德喜，武俊瑞. 灭菌乳中枯草芽孢杆菌危害与控制研究进展[J]. 乳业科学与技术, 2024, 47(4): 59-65.

卷积神经网络模型结合近红外光谱对特种乳掺水的鉴别

Identification of Water Adulteration in Non-bovine Milks Using Near-Infrared Spectroscopy Combined with a Convolutional Neural Network Model

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 9

编辑推荐

Metrics