An Integrated Handheld Electronic Nose for Identifying Liquid Volatile Chemicals Using Improved Gradient-Boosting Decision Tree Methods

Abstract

The main ingredients of various odorous products are liquid volatile chemicals (LVC). In human society, identifying the type of LVC is the inner logic of many applications, such as exposing counterfeit products, grading food quality, diagnosing interior environments, and so on. The electronic nose (EN) can serve as a cost-effective, time-efficient, and safe solution to LVC identification. In this paper, we present the design and evaluation of an integrated handheld EN, namely SMUENOSEv2, which employs the NVIDIA Jetson Nano module for running the LVC identification method. All components of SMUENOSEv2 are enclosed in a handheld case. This all-in-one structure makes it convenient to use SMUENOSEv2 for quick on-site LVC identification. To evaluate the performance of SMUENOSEv2, two common odorous products, i.e., perfumes and liquors, were used as the samples to be identified. After sampling data preprocessing and feature generation, two improved gradient-boosting decision tree (GBDT) methods were used for feature classification. Extensive experimental results show that SMUENOSEv2 is capable of identifying LVC with considerably high accuracies. With previously trained GBDT models, the time spent for identifying the LVC type is less than 1 s.

Using a Light-Weight CNN for Perfume Identification with An Integrated Handheld Electronic Nose

“…Based on the experimental results, the performance of LwCNN was compared with the four benchmark methods. As tested in the author's previous works [3,4], the hyperparameters of XGBoost, LightGBM, BPNN were automatically tunned using the hyperopt Python module [32], which models the parameter tuning process as solving a multi-variant function optimization problem. The time spent for automatic parameter tuning was included in the model training time, since the parameter values are also part of the trained model.…”

“…Existing electronic techniques that can be used for PI mainly include EN [3,4], fluctuationenhanced sensing (FES) [5,6], gas chromatography (GC) [7], and GC with mass spectrometry (GC-MS) [8,9]. Typical ENs are mainly composed of the measurement acquisition hardware and data processing software.…”

“…Then, the sensing data of the gas sensors is denoised and stored. Based on the denoised sensing measurements, the perfume type is identified by the data processing software [3]. In contrary to ENs, FES exploits the information contained in the low measurement noises [5] or micro-fluctuations of measurements [5,6].…”

“…Although existing EN solutions have rarely been reported separating the gaseous ingredients, the gas identification capability of ENs can help in exposing counterfeit perfume products. For example, the integrated handheld EN designed in the author's laboratory, namely SMUENOSEv2, which utilizes an embedded NVIDIA Jetson nano module as the computation kernel, is suitable for quick on-site PI applications [3].…”