Impact of Drift Factor on Feature Optimization in Electronic Nose Detection Systems
Author(s): Amadou F. Diop¹, Ndeye A. Sarr², and Mame A. Mbaye³
Affiliation: 1,2,3Department of Computer Science 1,2,3Cheikh Anta Diop University, Dakar, Senegal
Page No: 11-21-
Volume issue & Publishing Year: Volume 1 Issue 7 ,Dec-2024
Journal: International Journal of Modern Engineering and Management | IJMEM
ISSN NO: 3048-8230
DOI:
Abstract:
The electronic nose, a vital tool for olfactory detection in non-destructive testing (NDT), plays a critical role in mimicking human olfaction. However, its performance is frequently hampered by drift phenomena, including sensor degradation due to environmental variations and olfactory fatigue from prolonged usage. While the impact of drift on electronic nose performance is well-documented, its influence on feature optimization remains underexplored. This study introduces a novel perspective: drift factors not only disrupt sensor readings but also significantly influence the feature optimization process and subsequent drift compensation. To investigate, we focused on temperature and humidity—two predominant environmental drift factors. Experimental results confirmed that drift factors substantially affect feature optimization, demonstrating a positive correlation between sensor score concentration and classification accuracy. We employed an innovative quadratic feature optimization method to mitigate drift effects during feature optimization, enhancing the electronic nose's drift resilience. The unweighted quadratic feature optimization method emerged as the most effective, achieving a 100% recognition rate on the training set and 96% on the test set. These results highlight the method's potential to significantly enhance drift resistance in electronic nose systems. This study provides insights into the interplay between drift factors and feature optimization, offering a robust framework for advancing electronic nose technology.
Keywords:
Electronic Nose; Drift Compensation; Feature Optimization; Random Forest; Artificial Neural Networks
Reference:
[1] W. Qiao, Y. Zhang, and J. Liu, “Drift compensation in electronic noses: A review,” Sensors, vol. 21, no. 10, p. 3320, 2021.
[2] B. Li, X. Zhao, and Y. Wang, “Electronic nose technology in food and agricultural products quality evaluation,” Sensors, vol. 20, no. 15, p. 4269, 2020.
[3] G. Cheng and C. Lee, “Advances in electronic nose technology: From sensors to applications,” J. Food Sci. Technol., vol. 58, no. 3, pp. 767–777, 2021.
[4] J. Kim and K. Sung, “Effects of temperature and humidity on the electronic nose sensor response in agricultural product quality evaluation,” Sens. Actuators B Chem., vol. 295, pp. 112–119, 2019.
[5] V. Nenova and K. Bratanov, “Artificial neural networks for drift compensation in gas sensor data,” Sensors, vol. 20, no. 8, p. 2301, 2020.
[6] H. Liu and L. Shi, “A review of drift compensation methods for gas sensors in electronic noses,” J. Sensors, vol. 2018, p. 1215470, 2018.
[7] L. Xie and Z. Zhao, “Improving electronic nose performance in agricultural product detection through feature optimization,” Food Control, vol. 123, p. 107762, 2021.
[8] Q. Yan, X. Zhang, and M. Yang, “Machine learning methods for drift compensation in gas sensor arrays,” Sensors, vol. 19, no. 13, p. 2974, 2019.
[9] X. Wu and Y. Liu, “A novel approach to drift compensation in electronic noses using neural networks,” J. Agric. Food Chem., vol. 68, no. 19, pp. 5138–5145, 2020.
[10] C. Lai and X. Liu, “Sensor drift compensation based on domain adaptation for electronic noses,” Sensors, vol. 20, no. 4, p. 1002, 2020.
[11] X. Pang and J. Wei, “Nonlinear feature extraction for drift compensation in electronic nose applications,” IEEE Sensors J., vol. 21, no. 13, pp. 14529–14535, 2021.
[12] S. Feng and H. Liu, “Optimization of gas sensor features for agricultural product detection using random forest,” Sensors, vol. 21, no. 18, p. 6002, 2021.
[13] Z. Zhang and M. Zhang, “Enhancing the reliability of electronic noses through data fusion and feature optimization,” Food Bioprocess Technol., vol. 13, no. 4, pp. 563–572, 2020.
[14] X. Liu and J. Zhang, “Compensation methods for gas sensor arrays in electronic noses: A comprehensive review,” IEEE Access, vol. 7, pp. 95471–95488, 2019.
[15] D. Liu and Y. Zhou, “Improved random forest algorithm for drift compensation in electronic noses,” Sens. Actuators B Chem., vol. 318, p. 128218, 2020.
[16] L. Peng and Q. Chen, “Electronic noses for food and agricultural product detection: A review,” Food Res. Int., vol. 104, pp. 191–199, 2018.
[17] C. Lu and L. Zhang, “Impact of environmental factors on sensor drift in electronic noses for agricultural applications,” Sensors, vol. 21, no. 10, p. 3364, 2021.
[18] Z. Wang and C. Li, “Artificial neural networks for electronic nose drift compensation,” IEEE Trans. Ind. Electron., vol. 66, no. 6, pp. 4531–4539, 2019.
[19] E. González and A. Buitrago, “Analysis of drift effects in gas sensor arrays for electronic noses,” Sens. Actuators B Chem., vol. 274, pp. 238–244, 2018.
[20] H. Zhang and D. Lee, “Enhancing performance of electronic noses in food detection through advanced sensor calibration,” Sens. Actuators B Chem., vol. 290, pp. 132–140, 2019.
[21] X. Cui and J. Liu, “The application of neural network algorithms in drift compensation for electronic noses,” Sensors, vol. 20, no. 3, p. 742, 2020.
[22] Y. Chen and X. Wang, “A comparison of drift compensation techniques in electronic noses for agricultural applications,” Food Control, vol. 108, p. 106881, 2020.
[23] W. Qiao and J. Liu, “Using BP neural network to compensate for drift in electronic nose sensors,” J. Food Eng., vol. 242, pp. 36–42, 2019.
[24] J. Kim and J.-H. Ji, “The effects of environmental humidity on the performance of an electronic nose used in tea quality evaluation,” Sensors, vol. 21, no. 12, p. 4175, 2021.
[25] R. Zhou and X. Liu, “Improving drift compensation performance of electronic noses through multi-layer neural networks,” Sens. Actuators B Chem., vol. 314, p. 128080, 2020.
[26] Z. Wang and S. Zhang, “Use of random forest for data classification and drift compensation in electronic noses,” J. Agric. Food Chem., vol. 69, no. 11, pp. 3129–3137, 2021.
[27] C. Li and L. Zhou, “Addressing drift in gas sensors: Methods and challenges,” Chem. Eng. J., vol. 366, pp. 321–331, 2019.
[28] P. Zhang and X. Zhang, “Review on the compensation of drift in gas sensors for electronic noses,” Sensors, vol. 21, no. 10, p. 3463, 2021.
[29] X. Zhang and L. Yang, “Data preprocessing and compensation methods for electronic nose sensors,” Food Bioprocess Technol., vol. 13, no. 9, pp. 1560–1570, 2020.
[30] L. Yang and Y. Zhang, “Optimization of sensor performance in electronic noses through drift compensation,” Sens. Actuators B Chem., vol. 315, p. 128231, 2021.
[31] Y. Zhang and H. Huang, “Application of electronic noses in food industry: Advances and challenges,” Trends Food Sci. Technol., vol. 92, pp. 41–50, 2019.
[32] L. Xie and M. Wei, “Application of neural networks for drift compensation in electronic nose systems,” IEEE Sensors J., vol. 18, no. 12, pp. 4673–4681, 2018.
[33] J. Xu and J. Li, “Feature selection and optimization of sensors in electronic noses: A review,” Sensors, vol. 21, no. 16, p. 5432, 2021.
[34] Y. Wei and Y. Zhang, “Improving classification performance in electronic noses through sensor optimization,” IEEE Access, vol. 7, pp. 110513–110520, 2019.
[35] L. Zhao and S. Yang, “Effectiveness of electronic nose systems for fruit and vegetable quality detection,” Food Chem., vol. 307, p. 125513, 2020.
[36] Z. Chen and D. Liu, “A review on sensor drift compensation methods in electronic noses,” Sens. Actuators B Chem., vol. 273, pp. 232–239, 2018.
[37] Y. Jin and Y. Wu, “Application of drift compensation algorithms in food quality monitoring using electronic noses,” Food Res. Int., vol. 115, pp. 239–246, 2019.
[38] S. Zhao and J. Li, “Recent advances in electronic nose sensor array systems for food quality detection,” Sensors, vol. 21, no. 17, p. 5805, 2021.
[39] M. Zhang and P. Zhang, “Sensor drift analysis in electronic noses and methods for improving accuracy,” Sens. Actuators B Chem., vol. 318, p. 128223, 2020.
[40] Y. Liu and M. Wei, “Feature extraction techniques for drift compensation in gas sensors,” Sensors, vol. 20, no. 13, p. 3835, 2020.
[41] S. Liu and X. Wang, “Enhancing the anti-drift performance of electronic noses through multi-sensor fusion,” Sens. Actuators B Chem., vol. 315, p. 128245, 2021.
[42] L. Zhang and X. Liu, “Improving the robustness of electronic nose sensors for fruit classification,” Comput. Electron. Agric., vol. 178, p. 105801, 2020.
[43] H. Wu and Z. Chen, “Impact of environmental variations on the performance of gas sensors in electronic noses,” Food Control, vol. 127, p. 108131, 2021.