A Model Transfer Method Based on Transfer Component Analysis and
Direct Correction
LI Ling-qiao1, WANG Zhuo-jian1, CHEN Jiang-hai1, LU Feng1, HUANG Dian-gui2, YANG Hui-hua3, LI Quan2*
1. School of Computer Science and Information, Guilin University of Electronic Technology, Guilin 541004, China
2. Guangxi Zhuang Autonomous Region Center for Analysis and Test Research, Nanning 530022, China
3. School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China
Abstract：Near-infrared spectroscopy has been widely used due to its high efficiency and non-destructive properties advantages. However, the consistency phenomenon between near-infrared spectrometers can lead to insufficient accuracy when the master model predicts the spectra of its slave instruments. If the calibration model is rebuilt based on the offset spectrum, it will lead to higher consumption. This paper proposes a transfer component analysis direct standardization (TCADS) algorithm to address the above issues.The algorithm initially employs an enhanced TCA algorithm to convert master and enslaved person spectra, which adhere to distinct distributions, by projecting them into high-dimensional reproducing kernel Hilbert space. Subsequently, it reduces the dimensionality of their spectral matrices. Finally, a direct standardization algorithm is reapplied to the master and slave spectra post-TCA transformation, further enhancing the model's transfer performance. This algorithm combines nonlinear correction with linear correction, effectively alleviating the problem of overcorrection compared to traditional linear correction algorithms, and is robust. To verify the effectiveness of the algorithm, experiments were conducted on public datasets and compared with traditional direct standardization (DS), piecewise direct standardization (PDS), and slope and bias correction (SBC) methods. The experiment demonstrates that the TCADS algorithm proposed in this article efficiently minimizes spectral disparities between the master instrument and the slave instrument. This enhancement notably outperforms traditional model transfer algorithms, facilitating the effective sharing of near-infrared spectral models established on the master instrument to the slave instrument.
Key words：Near-infrared spectroscopy; Model transfer; Transfer learning; Transfer component analysis; Direct standardization
李灵巧，王卓健，陈江海，卢 丰，黄殿贵，杨辉华，李 泉. 迁移成分分析结合直接校正的模型传递方法[J]. 光谱学与光谱分析, 2024, 44(12): 3399-3405.
LI Ling-qiao, WANG Zhuo-jian, CHEN Jiang-hai, LU Feng, HUANG Dian-gui, YANG Hui-hua, LI Quan. A Model Transfer Method Based on Transfer Component Analysis and
Direct Correction. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3399-3405.
[1] Reda R, Saffaj T, Bouzida I, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2023, 303: 123213.
[2] Díaz E O, Iino H, Koyama K, et al. Food Chemistry, 2023, 429: 136907.
[3] Ojala K, Myrskyranta M, Liimatainen A, et al. International Journal of Pharmaceutics, 2020, 577: 119028.
[4] Santos F D, Vianna S G T, Cunha P H P, et al. Microchemical Journal, 2022, 181: 107696.
[5] Mishra P, Roger J M, Rutledge D N, et al. Postharvest Biology and Technology, 2020, 170: 111326.
[6] Huang G, Chen X, Li L, et al. Chemometrics and Intelligent Laboratory Systems, 2020, 201: 103986.
[7] Li L, Huang W, Wang Z, et al. Postharvest Biology and Technology, 2022, 183: 111720.
[8] Wang Y, Veltkamp D J, Kowalski B R. Analytical Chemistry, 1991, 63(23): 2750.
[9] Tan H W, Brown S D. Journal of Chemometrics, 2001, 15(8): 647.
[10] Bouveresse E, Hartmann C, Massart D L, et al. Analytical Chemistry, 1996, 68(6): 982.
[11] Sun X D, Wu H L, Chen Y, et al. Chemometrics and Intelligent Laboratory Systems, 2019, 194: 103869.
[12] Munnaf M A, Mouazen A M. Soil and Tillage Research, 2023, 225: 105545.
[13] Qi Y, Liu Y, Luo J. PhotoniX 4, 22(2023). https://doi.org/10.1186/s43074-023-00098-0.
[14] LIU Cui-ling, ZHOU Zi-yan, LI Tian-rui, et al(刘翠玲, 周子彦, 李天瑞, 等). Journal of Food Science and Technology(食品科学技术学报), 2019, 37(4): 95.
[15] Pan S J, Tsang I W, Kwok J T, et al. IEEE Transactions on Neural Networks, 2010, 22(2): 199.
[16] Kennard R W, Stone L A. Technometrics, 1969, 11(1): 137.
[17] Cui X. Molecules, 2022, 27(18): 6069.
[18] Zhang J, Zhou X, Li B. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2023, 301: 122978.
