Fig 9 - uploaded by Wenwen Zhang
Content may be subject to copyright.
Source publication
Conventional gas recognition algorithms used in artificial olfaction rely on manually selected steady-state features of the sensor array signal and ignore the dynamic process signal response features of the sensor array; hence, a large amount of dynamic feature information, which can improve the gas recognition accuracy, is lost. To address this pr...
Contexts in source publication
Context 1
... to 100%. Similarly, the loss function curves for both rapidly decrease at the beginning of training and then fluctuate downward with increasing global steps until they are close to 0. It can be concluded that WTCMCapsNet does not underfit or overfit during the training and testing process. In addition, the final confusion matrix, as shown in Fig. 9, exhibits no misidentifications of the unknown CO, H 2 , and CO and H 2 mixture gases, where the horizontal coordinate represents the true category of the gas and the vertical coordinate represents the predicted type of the ...
Context 2
... to 100%. Similarly, the loss function curves for both rapidly decrease at the beginning of training and then fluctuate downward with increasing global steps until they are close to 0. It can be concluded that WTCMCapsNet does not underfit or overfit during the training and testing process. In addition, the final confusion matrix, as shown in Fig. 9, exhibits no misidentifications of the unknown CO, H 2 , and CO and H 2 mixture gases, where the horizontal coordinate represents the true category of the gas and the vertical coordinate represents the predicted type of the ...
Similar publications
To extract any information from a complex data vibration, the time-frequency technique represents a suitable tool and a better indicator of the overall asset health of a rotating machine. Fluid influences on a cracked driveshaft and dynamic analysis of a rotating Cardan shaft are discussed in this paper. The study of the forced vibration signals of...
Citations
... extraction is aimed to extract relevant information from sensor responses to fully characterize the responses and has significant influences on the performance of feature selection and pattern recognition. 20,47 Compared with other feature extraction methods, such as coefficients of transform domain, 51 and exponentially moving average method, 52 features directly extracted from the original responses have special physiochemical meanings and reflect the reaction kinetics information. Besides, some papers have proved that the introduction of dynamic features, such as derivatives and integrals of the response curves, can improve the performance of pattern recognition. ...
Many studies focus on feature extraction and selection of gas sensor arrays for gas identification. In this work, we intended to find a feature subset obtained by selecting the most important features for simultaneously improving component and concentration detection performance of a gas sensor array to three harmful volatile organic compounds (VOCs, toluene, methanol, and ethanol) and their mixtures. First, 30 features were extracted from six sensors’ responses to form a multi-feature set. Then, two feature selection methods based on Wilks’ Λ-statistic and random forest were employed to obtain the best feature combination. Seven out of 30 features were finally selected to form the optimal feature set. The gas identification accuracy is 94.3%, and the concentration estimation error is 0.79 ppm (RMSE). Through feature selection, not only qualitative and quantitative analyses performance of VOCs mixtures are significantly improved, but also system complexity (6 to 4 sensors) and computation cost (by about 15%) are effectively reduced.
... An RNN is used for time-series data in natural language processing (NLP), speechemotion recognition (SER), speech-to-text (STT), and other processes [150]. Since wavelengths are measured at regular intervals, the wavelength axis is regarded as the time-series axis in NIR data. ...
The analysis of infrared spectroscopy of substances is a non-invasive measurement tech�nique that can be used in analytics. Although the main objective of this study is to provide a review
of machine learning (ML) algorithms that have been reported for analyzing near-infrared (NIR)
spectroscopy from traditional machine learning methods to deep network architectures, we also
provide different NIR measurement modes, instruments, signal preprocessing methods, etc. Firstly,
four different measurement modes available in NIR are reviewed, different types of NIR instruments
are compared, and a summary of NIR data analysis methods is provided. Secondly, the public
NIR spectroscopy datasets are briefly discussed, with links provided. Thirdly, the widely used
data preprocessing and feature selection algorithms that have been reported for NIR spectroscopy
are presented. Then, the majority of the traditional machine learning methods and deep network
architectures that are commonly employed are covered. Finally, we conclude that developing the
integration of a variety of machine learning algorithms in an efficient and lightweight manner is a
significant future research direction.
In this paper, a novel multi-sensor detection system framework for the rapid screening of bacterial infection is proposed. To capture the dynamic information of sensor response curves, eight features, such as time-domain and frequency-domain features, are extracted for each sensor. Additionally, a novel feature selection algorithm based on adaptive similarity and latent semantics (ASLSFS) is employed to eliminate irrelevant features in the initial feature set. Due to the redundant information and noise introduced by the sensors’ broad-spectrum response characteristics and hardware circuit interference, a dynamic information change weighting array optimization algorithm (DICWAO) is developed, which leverages the impact of adding candidate sensor features on the shared information among previously selected sensor features, candidate sensor features, and class label. The experimental results validate the effectiveness of the designed system. Comparative analysis with existing algorithms verifies the effectiveness of the developed feature selection algorithm and array optimization framework.
Gas-sensing technology that is ubiquitous has progressively gained significance in our daily lives. There is a growing requirement for the real-time, dependable and low-concentration gases detection to monitor toxic and...
