相关实验视频

Updated: May 6, 2026

Electronic Tongue Generating Continuous Recognition Patterns for Protein Analysis

Electronic Tongue Generating Continuous Recognition Patterns for Protein Analysis

Published on: September 16, 2014

化学测量方法,感官评估和智能感官技术与基于GAN的集成深度学习框架相结合,以区分化品种

Che Shen¹, Ran Wang¹, Qi Jin¹

¹Engineering Research Center of Bio-process, Ministry of Education, Hefei University of Technology, Hefei 230009, China.

|August 24, 2024

概括

这项研究开发了一个先进的深度学习框架,使用多种分析技术来验证盐品种的真实性. 综合模型在区分不同盐品种方面实现了95%的准确性.

科学领域:

食品科学食品科学食品科学
分析化学分析化学
人工智能的人工智能

背景情况:

消费者对盐真实性的信心是一个重大问题.
需要准确的方法来区分各种盐品种.

研究的目的:

开发和验证一个集成的深度学习框架,用于可靠的盐品种歧视.
结合各种分析数据,以提高真实性评估.

主要方法:

使用集成的生成对抗网络 (GAN) 与头空间固体相微提取/气色谱-质谱 (HS-SPME/GC-MS),头空间气色谱-离子移动谱 (HS-GC-IMS),电子鼻子 (E-nose),电子舌头 (E-tongue),定量描述性分析和自由氨基酸/5'-核酸分析.
使用InfoGAN和训练的基础分类器生成一个复合数据集.
通过对集成模型的动态权重进行融合预测结果.

主要成果:

综合框架实现了高性能指标:95%的准确性,0.080RMSE,0.9450精度,0.9470回忆和0.9460F1分数.
一项废除研究证实了集合学习对于模型概括的不可或缺.

结论:

开发的深度学习框架有效地区分了四种盐的种.
多个分析数据源和集体学习的整合显著提高了真实性评估的可靠性.

关键词:

食品的真实性食品的真实性气色谱-离子移动性光谱法 (GC-IMS) 是一种气色谱法.生成性的对抗性网络.头部空间固相微提取/气相染色学质谱学 (HS-SPME/GCMS)综合深度学习框架综合深度学习框架盐的子盐的子

更多相关视频

New Methods to Study Gustatory Coding

New Methods to Study Gustatory Coding

Published on: June 29, 2017

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Integration of Animal Behavioral Assessment and Convolutional Neural Network to Study Wasabi-Alcohol Taste-Smell Interaction

Published on: August 16, 2024

相关概念视频

Gustation

Gustation

Gustation is a chemical sense that, along with olfaction (smell), contributes to our perception of taste. It starts with the activation of receptors by chemical compounds (tastants) dissolved in the saliva. The saliva and filiform papillae on the tongue distribute the tastants and increase their exposure to the taste receptors.