|本期目录/Table of Contents|

[1]张 芸,宋 刚*,刘 军,等.用于奶油色素定量分析的注意力残差网络设计与验证[J].武汉工程大学学报,2024,46(04):410-416.[doi:10.19843/j.cnki.CN42-1779/TQ.202310012]
 ZHANG Yun,SONG Gang*,LIU Jun,et al.Design and verification of attention residual network for quantitative analysis of cream pigments[J].Journal of Wuhan Institute of Technology,2024,46(04):410-416.[doi:10.19843/j.cnki.CN42-1779/TQ.202310012]
点击复制

用于奶油色素定量分析的注意力残差
网络设计与验证
(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
46
期数:
2024年04期
页码:
410-416
栏目:
生物与环境工程
出版日期:
2024-08-28

文章信息/Info

Title:
Design and verification of attention residual network for quantitative analysis of cream pigments
文章编号:
1674 - 2869(2024)04 - 0410 - 07
作者:
张 芸1宋 刚*2刘 军1谭正林3黄晓彤1
1. 武汉工程大学计算机科学与工程学院,
智能机器人湖北省重点实验室(武汉工程大学),湖北 武汉 430205;
2. 武汉工程大学艺术设计学院,湖北 武汉 430205;
3. 湖北经济学院烹饪与营养学系,湖北 武汉 430205
Author(s):
ZHANG Yun1 SONG Gang*2 LIU Jun1 TAN Zhenglin3 HUANG Xiaotong1
1. School of Computer Science and Engineering ,
Hubei Key Laboratory of Intelligent Robot (Wuhan Institute of Technology),Wuhan 430205, China;
2. School of Art and Design,Wuhan Institute of Technology, Wuhan 430205, China;
3. Department of Cuisine and Nutrition,Hubei University of Economics, Wuhan 430205, China
关键词:
近红外光谱温度注意力机制残差网络奶油色素
Keywords:
near-infrared spectroscopy temperature attention mechanism residual network cream pigment
分类号:
TP391
DOI:
10.19843/j.cnki.CN42-1779/TQ.202310012
文献标志码:
A
摘要:
针对样品温度变化问题,由于近红外光谱对温度等物理条件变化十分敏感,以奶油中的靛蓝色素作为光谱定量分析数据,提出了一种变温注意力残差网络解决方案。变温注意力残差网络融合温度以及光谱特征,其主干结构使用并发空间和通道挤压和激励注意力机制对残差块处理后的特征进行整合增强。随后采用最大池化和随机丢弃层进行特征降维和模型正则化。将去掉注意力模块的网络与六种深度学习常用的回归分析网络对比,验证其在领域的高适用性。将变温注意力残差网络与6种网络中最佳模型的3种优化形式对比,验证其高性能。最后对模型调优,训练和测试损失差缩小至0.000 5,决定系数和相对分析误差达到了最佳值0.929 3和3.703 1,表明该模型能在实践中对变温条件下的光谱定量分析。
Abstract:
Temperature change of sample causes fluctuation to its spectrum. As near-infrared spectroscopy is very sensitive to changes in physical conditions such as temperature, we took the indigo pigment in cream as the spectral quantitative analysis data and proposed a variable temperature attention residual network. This network integrates temperature and spectral features, and its backbone structure adopts a concurrent spatial and channel squeeze and excitation attention mechanism to integrate and enhance the features processed by the residual block. Subsequently, we used maximum pooling and random dropout layers for feature dimensionality reduction and model regularization. By comparing the network without the attention module with six commonly used regression analysis networks in deep learning, we verified its high applicability in this field; by comparing the variable temperature attention residual network with three optimization forms of the best model among the six networks, we verified its high performance. After we tuned the model, the difference between the training and test losses was reduced to 0.000 5, and the coefficient of determination and the relative analysis error reached the best values of 0.929 3 and 3.703 1, indicating that the model can perform quantitative analysis of spectra under variable temperature conditions in practice.

参考文献/References:

[1] 刘军,吴梦婷,谭正林,等.近红外光谱无损检测技术中数据的分析方法概述[J].武汉工程大学学报,2017,39(5):496-502.

[2] ZHANG S P, TAN Z L, LIU J, et al. Determination of the food dye indigotine in cream by near-infrared spectroscopy technology combined with random forest model[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2020, 227: 117551.
[3] LIU J, SUN S Q, TAN Z L, et al. Nondestructive detection of sunset yellow in cream based on near-infrared spectroscopy and interval random forest[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy,2020,242:118718.
[4] LIU J, ZHANG J X, TAN Z L, et al. Detecting the content of the bright blue pigment in cream based on deep learning and near-infrared spectroscopy [J]. Spectrochimica Acta. Part A, Molecular and Biomolecular Spectroscopy, 2021, 270:120757.
[5] JOSHI I, TRUONG V K, CHAPMAN J, et al. The use of two-dimensional spectroscopy to interpret the effect of temperature on the near infrared spectra of whisky[J]. Journal of Near Infrared Spectroscopy, 2020, 28(3): 148-152.
[6] GON?ALVES M, PAIVA N T, FERRA J M, et al. Effect of temperature and age on near infrared spectra of amino resins[J]. Journal of Near Infrared Spectroscopy, 2021, 29(2):84-91.
[7] SHENG R, CHENG W, LI H H, et al. Model development for soluble solids and lycopene contents of cherry tomato at different temperatures using near-infrared spectroscopy[J]. Postharvest Biology and Technology, 2019,156:110952
[8] LIU J J, LI B Q, XU M L, et al. Simultaneous quantitative analysis of three components in mixture samples based on NIR spectra with temperature effect[J]. Analytical Methods, 2017, 9(13): 2076-2081.
[9] 郑游,王磊,杨紫文.基于多尺度深度图自适应融合的单目深度估计[J].武汉工程大学学报,2024,46(1):85-90.
[10] 陈江川,吴云韬,孔权.基于CBAM-Res2Net的人群计数算法[J].武汉工程大学学报,2022,44(6):664-669.
[11] SARADHI V V M, RAO V P, KRISHNAN G V, et al. Prediction of alzheimer’s disease using LeNet- CNN model with optimal adaptive bilateral filtering[J].International Journal of Communication Networks and Information Security, 2023,15(1):52-58.
[12] 金守峰,侯一泽,焦航,等.基于改进AlexNet模型的抓毛织物质量检测方法[J].纺织学报,2022,43(6):133-139.
[13] 井奚月,乔婕,么秀华,等.ZFNet模型在胶质瘤MRI诊断中的应用[J].中国现代神经疾病杂志,2021,21(3):156-161.
[14] 孟森,黄海松,朱云伟,等.基于U~2-Net和VGG16的天然气泄露检测[J].激光与红外,2023,53(3):386-393.
[15] 王溢琴,董云云,刘慧玲.基于GoogLeNet和空间谱变换的高光谱图像超分辨率方法[J].光学技术,2022,48(1):93-101.
[16] 蔡嘉辉,王琨,董康,等.基于DenseNet和随机森林的电力用户窃电检测[J].计算机应用,2021,41(增刊1):75-80.
[17] 吴永清,唐娜,黄璐瑶,等.可见近红外光谱结合多元统计分析的面粉吸水率检测模型构建[J].光谱学与光谱分析,2023,43(9):2825-2831.
[18] 谭正林,张芸,刘军.一种近红外光谱优化方法、装置、系统以及存储介质: CN202310688655.1 [P]. 2024-01-05.
[19] ROY A G, NAVAB N, WACHINGER C. Concurrent spatial and channel ‘squeeze & excitation’in fully convolutional networks[C]//Medical Image Computing and Computer Assisted Intervention-MICCAI 2018: 21st International Conference,September 16-20,2018,Proceedings,Part I. Granada, Spain:Springer International Publishing, 2018: 421-429.
[20] 黄海霞,李耀翔,张哲宇.基于ResNet的森林土壤碳含量近红外预测模型[J]. 森林工程,2023,39(6):164-171.

相似文献/References:

[1]赵振华.橡胶硫化温度的模糊控制[J].武汉工程大学学报,2008,(04):93.
 ZHAO Zhen hua.Fuzzy control of vulcanization temperature[J].Journal of Wuhan Institute of Technology,2008,(04):93.
[2]袁江,胡明辅*,毕二朋,等.湿空气饱和水蒸气压数学计算式的拟合与优选[J].武汉工程大学学报,2011,(10):25.
 YUAN Jiang,HU Ming fu,BI Er peng,et al.Formulas fitting and optimizing of saturated watervapor pressure of moist air[J].Journal of Wuhan Institute of Technology,2011,(04):25.
[3]许钢,林园胜,胡天水,等.虚拟仪器技术在温度采集系统中的应用[J].武汉工程大学学报,2013,(07):81.[doi:103969/jissn16742869201307016]
 XU Gang,LIN Yuan sheng,HU Tian shui,et al.Application of virtual instrument technologyin temperature acquisition system[J].Journal of Wuhan Institute of Technology,2013,(04):81.[doi:103969/jissn16742869201307016]
[4]周剑秋,叶志雄,邱奇,等.温度与应变率对Cu70 Zn30孪晶变形的影响[J].武汉工程大学学报,2014,(05):42.[doi:103969/jissn16742869201405010]
 ZHOU Jian qiu,YE Zhi xiong,QIU Qi,et al.Effects of strain rate and temperature on deformation twinning in Cu70Zn30 alloy[J].Journal of Wuhan Institute of Technology,2014,(04):42.[doi:103969/jissn16742869201405010]
[5]刘 岑,杨 帆,刘 兵,等.室温与超低温时奥氏体不锈钢S30408的屈强比[J].武汉工程大学学报,2018,40(02):228.
 LIU Cen,YANG?Fan,LIU Bing,et al.Austenitic Stainless Steel S30408 Yield Ratio at Room Temperature and Ultra-Low Temperature[J].Journal of Wuhan Institute of Technology,2018,40(04):228.
[6]王振鹏,黄民水*,卢海林.基于振动监测的工字钢梁温度-频率关系模型[J].武汉工程大学学报,2020,42(03):321.[doi:10.19843/j.cnki.CN42-1779/TQ.201912012]
 WANG Zhenpeng,HUANG Minshui*,LU Hailin.Model of Frequency Versus Temperature of Steel I-Beam Based on Vibration Monitoring[J].Journal of Wuhan Institute of Technology,2020,42(04):321.[doi:10.19843/j.cnki.CN42-1779/TQ.201912012]
[7]吕全红,肖莲珍*.基于水化动力学模型的水泥基材料温度效应[J].武汉工程大学学报,2020,42(04):434.[doi:10.19843/j.cnki.CN42-1779/TQ.201910011]
 Lu Quanhong,XIAO Lianzhen*.Temperature Effect of Cement-Based Materials Based on Hydration Kinetics Model[J].Journal of Wuhan Institute of Technology,2020,42(04):434.[doi:10.19843/j.cnki.CN42-1779/TQ.201910011]
[8]程凯旋,杨加美,丁珮珊,等.高密度聚乙烯垫片的非线性压缩-回弹性能测试[J].武汉工程大学学报,2021,43(04):468.[doi:10.19843/j.cnki.CN42-1779/TQ. 202105019]
 CHENG Kaixuan,YANG Jiamei,DING Peishan,et al.Nonlinear Compression-Resilience Performance Test of HDPE Gaskets[J].Journal of Wuhan Institute of Technology,2021,43(04):468.[doi:10.19843/j.cnki.CN42-1779/TQ. 202105019]
[9]雷 德,蔡 璐*.压缩二氧化碳和甲基吡咯烷酮剥离石墨烯的分子动力学模拟[J].武汉工程大学学报,2023,45(01):48.[doi:10.19843/j.cnki.CN42-1779/TQ.202201006]
 LEI De,CAI Lu*.Molecular Dynamics Simulation of Graphene Exfoliation in MixedSolvent of Compressed Carbon Dioxide and Methylpyrrolidon[J].Journal of Wuhan Institute of Technology,2023,45(04):48.[doi:10.19843/j.cnki.CN42-1779/TQ.202201006]
[10]江志豪,文小玲*,舒李俊.水产养殖水域的溶解氧浓度检测方法研究[J].武汉工程大学学报,2023,45(02):196.[doi:10.19843/j.cnki.CN42-1779/TQ.202210023]
 JIANG Zhihao,WEN Xiaoling*,SHU Lijun.Detection Method of Dissolved Oxygen Concentration in Aquaculture Waters[J].Journal of Wuhan Institute of Technology,2023,45(04):196.[doi:10.19843/j.cnki.CN42-1779/TQ.202210023]

备注/Memo

备注/Memo:
收稿日期:2023-10-18
基金项目:湖北省自然科学基金(2022CFC001);浙江省生物标志物与体外诊断转化重点实验室开放基金(KFJJ 2023006);武汉工程大学第十四届研究生教育创新基金(CX2022331、CX2022348、CX2022365)
作者简介:张 芸,硕士研究生。Email: yunzhangbdd@163.com
*通信作者:宋 刚,博士研究生,讲师。Email: 16681419@qq.com
引文格式:张芸,宋刚,刘军,等. 用于奶油色素定量分析的注意力残差网络设计与验证[J]. 武汉工程大学学报,2024,46(4):410-416,423.
更新日期/Last Update: 2024-08-31