论文标题
使用不同的深度学习模型和支持向量机的快速检测方法来增强电子鼻系统性能的验证
Validation of the rapid detection approach for enhancing the electronic nose systems performance, using different deep learning models and support vector machines
论文作者
论文摘要
例如，实时气体分类是食品和饮料质量控制，工业环境中的事故预防等应用中的一个必要问题和挑战。近年来，即使在电子鼻子（E-Nose）领域，深度学习（DL）模型也显示出在各种问题中分类和预测数据的巨大潜力。在这项工作中，我们使用了支持向量机（SVM）算法和三种不同的DL模型来验证不同测量条件的快速检测方法（基于处理原始信号的早期部分和上升窗口协议）。我们使用五个不同的电子鼻子数据库进行了一组试验，其中包括15个数据集。根据结果，我们得出的结论是，所提出的方法具有很高的潜力，并且可以用于电子鼻技术，从而减少了进行预测并加速响应时间的必要时间。因为在大多数情况下，它仅使用前30％或更少的测量数据（在气体注入开始后计数）实现可靠的估计。）调查结果表明，快速检测方法使用不同的分类方法产生可靠的预测模型。尽管如此，SVM似乎可以获得最佳的准确性，正确的窗口尺寸和更好的训练时间。
Real-time gas classification is an essential issue and challenge in applications such as food and beverage quality control, accident prevention in industrial environments, for instance. In recent years, the Deep Learning (DL) models have shown great potential to classify and forecast data in diverse problems, even in the electronic nose (E-Nose) field. In this work, we used a Support Vector Machines (SVM) algorithm and three different DL models to validate the rapid detection approach (based on processing an early portion of raw signals and a rising window protocol) over different measurement conditions. We performed a set of trials with five different E-Nose databases that include fifteen datasets. Based on the results, we concluded that the proposed approach has a high potential, and it can be suitable to be used for E-nose technologies, reducing the necessary time for making forecasts and accelerating the response time. Because in most cases, it achieved reliable estimates using only the first 30% or fewer of measurement data (counted after the gas injection starts.) The findings suggest that the rapid detection approach generates reliable forecasting models using different classification methods. Still, SVM seems to obtain the best accuracy, right window size, and better training time.
