电子鼻
模式识别(心理学)
样品(材料)
二进制数
特征(语言学)
计算机科学
数学
人工智能
色谱法
化学
算术
语言学
哲学
作者
Hui-Rang Hou,Qing‐Hao Meng,Pei-Feng Qi,Jing Tao
标识
DOI:10.1109/tim.2021.3112789
摘要
The aim of this study is to identify types and grades of Chinese liquors. For this, a hand-held electronic nose (e-nose) system is designed and a triangular difference-based binary coding (TDBC) recognition method is proposed. For a test sample of liquors, features extracted from five gas sensors of the e-nose are converted to binary codes (0 and 1) for each liquor category. Specifically, for each liquor category, if a feature value of a test sample is within the feature value range of all training samples, we mark it as 1, otherwise 0. Subsequently, for each liquor category, the sum of binary codes of the test sample are calculated, and the category corresponding to the maximum sum value is determined as the predicted label of the test sample. Using the e-nose-based TDBC method, average recognition accuracies of 97.5% and 99.0% for liquor type identification and grade evaluation were achieved, which were considerably higher than those obtained using four traditional recognition methods. These results indicate that, as a novel approach, the e-nose-based TDBC method allows the recognition of Chinese liquors accurately and quickly, which is of great significance for liquor detection and industrial quality assurance methods.
科研通智能强力驱动
Strongly Powered by AbleSci AI