An adaptive learning method for the fusion information of electronic nose and hyperspectral system to identify the egg quality

IF 4.1 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Sensors and Actuators A-physical Pub Date : 2022-10-16 DOI:10.1016/j.sna.2022.113824

Qinglun Zhang , Siyuan Kang , Chongbo Yin , Ziyang Li , Yan Shi

{"title":"An adaptive learning method for the fusion information of electronic nose and hyperspectral system to identify the egg quality","authors":"Qinglun Zhang , Siyuan Kang , Chongbo Yin , Ziyang Li , Yan Shi","doi":"10.1016/j.sna.2022.113824","DOIUrl":null,"url":null,"abstract":"<div>Data fusion technology based on the multi-sensor system can obtain the holistic properties of samples. However, multi-sensor data fusion will bring more redundant information, which will lead to low classification performance. In this work, a multi-data-fusion-attention network (MDFA-Net) is proposed, combined with the electronic nose (e-nose) and hyperspectral system to identify the egg quality. Firstly, the gas information and spectral information of eggs are obtained under different feeding conditions. Secondly, a feature adaptive learning (FAL) unit is designed to select effective information and enhance the ability of feature expression. Thirdly, based on the FAL unit, a decision network is formed to identify the fusion information of e-nose and hyperspectral system. Finally, compared with other deep learning network models, the accuracy of MDFA-Net is 99.88%, the precision is 99.87%, the recall is 99.88%, and the F1-score is 99.90%, which shows better classification performance and stability.</div>","PeriodicalId":21689,"journal":{"name":"Sensors and Actuators A-physical","volume":null,"pages":null},"PeriodicalIF":4.1000,"publicationDate":"2022-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"19","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Sensors and Actuators A-physical","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0924424722004599","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}

引用次数: 19

Abstract

Data fusion technology based on the multi-sensor system can obtain the holistic properties of samples. However, multi-sensor data fusion will bring more redundant information, which will lead to low classification performance. In this work, a multi-data-fusion-attention network (MDFA-Net) is proposed, combined with the electronic nose (e-nose) and hyperspectral system to identify the egg quality. Firstly, the gas information and spectral information of eggs are obtained under different feeding conditions. Secondly, a feature adaptive learning (FAL) unit is designed to select effective information and enhance the ability of feature expression. Thirdly, based on the FAL unit, a decision network is formed to identify the fusion information of e-nose and hyperspectral system. Finally, compared with other deep learning network models, the accuracy of MDFA-Net is 99.88%, the precision is 99.87%, the recall is 99.88%, and the F₁-score is 99.90%, which shows better classification performance and stability.

查看原文本刊更多论文

一种融合电子鼻和高光谱系统信息的自适应学习方法用于鸡蛋品质识别

基于多传感器系统的数据融合技术可以获得样本的整体特性。然而，多传感器数据融合会带来更多的冗余信息，导致分类性能下降。本文提出了一种多数据融合关注网络(MDFA-Net)，结合电子鼻(e-nose)和高光谱系统来识别鸡蛋质量。首先，获取不同饲养条件下鸡蛋的气体信息和光谱信息;其次，设计特征自适应学习单元，选择有效信息，增强特征表达能力;第三，以FAL单元为基础，构建决策网络，对电子鼻与高光谱系统的融合信息进行识别。最后，与其他深度学习网络模型相比，MDFA-Net的准确率为99.88%，精密度为99.87%，召回率为99.88%，f1分数为99.90%，表现出更好的分类性能和稳定性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Sensors and Actuators A-physical 工程技术-工程：电子与电气

CiteScore

8.10

自引率

6.50%

发文量

630

审稿时长

49 days

期刊介绍： Sensors and Actuators A: Physical brings together multidisciplinary interests in one journal entirely devoted to disseminating information on all aspects of research and development of solid-state devices for transducing physical signals. Sensors and Actuators A: Physical regularly publishes original papers, letters to the Editors and from time to time invited review articles within the following device areas: • Fundamentals and Physics, such as: classification of effects, physical effects, measurement theory, modelling of sensors, measurement standards, measurement errors, units and constants, time and frequency measurement. Modeling papers should bring new modeling techniques to the field and be supported by experimental results. • Materials and their Processing, such as: piezoelectric materials, polymers, metal oxides, III-V and II-VI semiconductors, thick and thin films, optical glass fibres, amorphous, polycrystalline and monocrystalline silicon. • Optoelectronic sensors, such as: photovoltaic diodes, photoconductors, photodiodes, phototransistors, positron-sensitive photodetectors, optoisolators, photodiode arrays, charge-coupled devices, light-emitting diodes, injection lasers and liquid-crystal displays. • Mechanical sensors, such as: metallic, thin-film and semiconductor strain gauges, diffused silicon pressure sensors, silicon accelerometers, solid-state displacement transducers, piezo junction devices, piezoelectric field-effect transducers (PiFETs), tunnel-diode strain sensors, surface acoustic wave devices, silicon micromechanical switches, solid-state flow meters and electronic flow controllers. Etc...