Electronic nose sensor drift compensation based on deep belief network

Luo, Yu; Wei, Shanbi; Chai, Yi; Sun, Xuefeng

doi:10.1109/chicc.2016.7553969

Cited by 14 publications

(11 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Third, two feature calibration methods, namely, the OSC [22] and DS [24] methods, are also considered. In addition, to verify the superiority of the GDBCN in deep learning models, we select an RBM, a 3-layer DBN [34][35] and a multi-layer perceptron (MLP) as the contrast model, where the DBN and RBM are utilized for initializing the perceptron coupled with a softmax classifier. For the GDBCN, the learning rate is = 0.…”

Section: ) Experimental Resultsmentioning

confidence: 99%

“…1: → 1 2: → 2 Several contrast methods, including SVM, softmax classifier, OSC [22], DS [24], neighbourhood component analysis (NCA) coupled with softmax, DRCA [28], CDSL [27], RBM [39] and MLP, are leveraged for evaluating the effectiveness and competitiveness of the proposed GDBCN model. Moreover, different from our GDBCN model, we establish another DBN model [34][35] for feature extraction and then collect the extracted features from the Master dataset for training the softmax layer, which is called DBN-softmax, to test the recognition performance. For the GDBCN, the learning rate is = 0.…”

Section: ) Experimental Resultsmentioning

confidence: 99%

“…Although previous studies have proven the feasibility and validity of deep learning in drift compensation of gas sensor data, in general, the actual performances of these deep learning models are less impressive compared with some existing methods. DRCA and CDSL perform better than DBN and sSAE for the same dataset [27][28][34][35]. It is important and promising to further investigate the application of deep learning in this field.…”

Section: Related Workmentioning

confidence: 96%

“…It adopted an sSAE and a DBN to extract features automatically, and then the extracted features were used to train a gas classifier. Luo et al [35] applied DBN to pre-process gas sensor data to associate and combine with characteristics of sensor data and strengthen the coupling between each characteristic. Then, the DBN model combined with an SVM demonstrated effectiveness in improving gas recognition under drift.…”

Section: Related Workmentioning

confidence: 99%

See 3 more Smart Citations

A Drift-Compensating Novel Deep Belief Classification Network to Improve Gas Recognition of Electronic Noses

et al. 2020

View full text Add to dashboard Cite

Electronic nose (E-nose) systems have a good effect on the identification of distinct odours. However, the properties of chemical gas sensors indicate that ageing, poisoning, fluctuation of environmental conditions (moisture, temperature, etc.) and a lack of fabrication repeatability, etc. have a large impact on the sensitivity and accuracy of sensors, which leads to sensor data drift. Although previous studies have indicated the feasibility and validity of deep learning in drift compensation of gas sensor data, the actual performances of these deep learning models are less impressive compared with some existing methods. Thus, we intend to further explore a novel deep learning model for drift compensation for E-noses. In this paper, we investigate the drift compensation effect of E-nose data based on a deep belief network (DBN) and constructed a Gaussian deep belief classification network (GDBCN) model by cascading a Gaussian-Bernoulli restricted Boltzmann machines based DBN with a softmax classifier layer to compensate for sensor drift at the decision level. The merits of our method are as follows: 1) it is a unified classification model for drift auto-compensation at the decision level rather than a feature extractor; 2) it couples unsupervised and supervised techniques by modelling the intrinsic distribution of the data from different domains in an unsupervised manner and fine-tunes the model parameters by leveraging the label information of the source domain; 3) the supervised fine-tuning process for the coupled GDBCN model fits well with the nature of the supervised task and guarantees that the parameters of the DBN will be useful for classification; 4) the GDBCN model is a classification model and thus automatically compensates for drift without manually setting specific model rules for domain alignment before classification. Experimental results on real sensor datasets demonstrate the effectiveness and superiority compared with several existing control methods.

show abstract

Section: ) Experimental Resultsmentioning

confidence: 99%

Section: ) Experimental Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 96%

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

A Drift-Compensating Novel Deep Belief Classification Network to Improve Gas Recognition of Electronic Noses

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Hence, new machine-learning paradigms have been adopted to deal with this challenge. Luo et al utilized a deep-learning neural network to abstract the drift-independent features for recognition in [21], eliminating the drift of the features. Zhang et al focused on the situation in which the labels for the drifted instances are unknown [11,22].…”

Section: Introductionmentioning

confidence: 99%

Active Learning on Dynamic Clustering for Drift Compensation in an Electronic Nose System

Liu

Chen

et al. 2019

Sensors

View full text Add to dashboard Cite

Drift correction is an important concern in Electronic noses (E-nose) for maintaining stable performance during continuous work. A large number of reports have been presented for dealing with E-nose drift through machine-learning approaches in the laboratory. In this study, we aim to counter the drift effect in more challenging situations in which the category information (labels) of the drifted samples is difficult or expensive to obtain. Thus, only a few of the drifted samples can be used for label querying. To solve this problem, we propose an innovative methodology based on Active Learning (AL) that selectively provides sample labels for drift correction. Moreover, we utilize a dynamic clustering process to balance the sample category for label querying. In the experimental section, we set up two E-nose drift scenarios—a long-term and a short-term scenario—to evaluate the performance of the proposed methodology. The results indicate that the proposed methodology is superior to the other state-of-art methods presented. Furthermore, the increasing tendencies of parameter sensitivity and accuracy are analyzed. In addition, the Label Efficiency Index (LEI) is adopted to measure the efficiency and labelling cost of the AL methods. The LEI values indicate that our proposed methodology exhibited better performance than the other presented AL methods in the online drift correction of E-noses.

show abstract

Diagnostic Performance of Electronic Noses in Cancer Diagnoses Using Exhaled Breath

et al. 2022

View full text Add to dashboard Cite

IMPORTANCE There has been a growing interest in the use of electronic noses (e-noses) in detecting volatile organic compounds in exhaled breath for the diagnosis of cancer. However, no systematic evaluation has been performed of the overall diagnostic accuracy and methodologic challenges of using e-noses for cancer detection in exhaled breath. OBJECTIVETo provide an overview of the diagnostic accuracy and methodologic challenges of using e-noses for the detection of cancer.

show abstract

Electronic nose sensor drift compensation based on deep belief network

Cited by 14 publications

References 13 publications

A Drift-Compensating Novel Deep Belief Classification Network to Improve Gas Recognition of Electronic Noses

A Drift-Compensating Novel Deep Belief Classification Network to Improve Gas Recognition of Electronic Noses

Active Learning on Dynamic Clustering for Drift Compensation in an Electronic Nose System

Diagnostic Performance of Electronic Noses in Cancer Diagnoses Using Exhaled Breath

Contact Info

Product

Resources

About