电子鼻
质量(理念)
计算机科学
质量评定
人工智能
工程类
模式识别(心理学)
机器学习
作者
Wentian Zhang,Taoping Liu,Amber Brown,Maiken Ueland,Shari L. Forbes,Steven W. Su
出处
期刊:IEEE Sensors Journal
[Institute of Electrical and Electronics Engineers]
日期:2022-01-01
卷期号:: 1-1
标识
DOI:10.1109/jsen.2022.3147185
摘要
As with any profitable industry, the whisky market is subject to fraudulent activity, including adulteration. An expert can identify the differences between whiskies, but it is difficult for the majority of consumers to differentiate fraudulent beverages. Complex chemical and analytical analyses have been able to detect the differences between whiskies; however, this type of analysis is time-consuming, complex, requires trained professionals, and can only be conducted in the laboratory. A rapid and real-time assessment of whisky quality could prove beneficial to wholesalers and consumers. The odour of whiskies can be used to identify their brands, regions and styles, as thus has the potential for quality assessment and fraudulent detection. One type of technology used for real-time odour analysis is an electronic nose (e-nose). This study investigates the capability of a new e-nose prototype (called NOS.E) developed by our team to identify the differences between six whiskies with respect to their brand names, regions, and styles. This study investigates the capability of a new e-nose prototype (called NOS.E) developed by our team in identifying the differences among whiskies. Ensemble of several classifiers is adopted to improve the classification accuracy of the system. The proposed e-nose solution was verified by a field testing displayed at the CEBIT Australia 2019 trade show, by reaching an accuracy of 96.15%, 100%, and 92.31% in brand name, region, and style classification, respectively. Confirmation of the NOS.E findings was further carried out using comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GC $\times $ GC-TOFMS).
科研通智能强力驱动
Strongly Powered by AbleSci AI