SMART-NIR: A Multi-Kernel Vision Transformer with Kolmogorov-Arnold Network Classifier for Near-Infrared Spectral Classification

Nguyen Thi Hoang Phuong, Phan Minh Nhat, Nguyen Van Hieu

Abstract


Near-infrared (NIR) spectroscopy is a powerful analytical tool widely applied in classification and quality control. However, it faces significant challenges when handling complex, multidimensional, and nonlinear data. This paper introduces SMART-NIR, an innovative framework designed to enhance NIR spectral analysis by integrating multi-kernel feature extraction, an improved Vision Transformer architecture with Dual-MLP, and Kolmogorov-Arnold Networks (KAN) for superior classification performance. The proposed framework achieved an impressive classification accuracy of 99.24%, surpassing traditional Transformer models. The multi-kernel feature extraction technique allows adaptive extraction of key spectral features, while KAN integration improves classification accuracy by 5% over standard MLP classifiers. Additionally, the Dual-MLP architecture significantly reduces the number of parameters and floating point operations (FLOPs) compared to conventional feed-forward networks used in Transformers. SMART-NIR demonstrates exceptional capability in capturing complex, nonlinear data relationships, making it highly effective for real-time inspection and analysis in dynamic, noisy environments. The framework offers a promising solution for enhancing the accuracy and efficiency of NIR spectroscopy across various practical applications.

Full Text:

PDF PDF

References


Huang, Y. (2022). Chemometric methods in analytical spectroscopy technology. In Chemometric Methods in Analytical Spectroscopy Technology (pp. 1-29). Singapore: Springer Nature Singapore.

Ozaki, Y., Huck, C., Tsuchikawa, S., & Engelsen, S. B. (Eds.). (2021). Near-infrared spectroscopy: Theory, spectral analysis, instrumentation, and applications. Singapore: Springer.

Moghaddam, H. N., Tamiji, Z., Lakeh, M. A., Khoshayand, M. R., & Mahmoodi, M. H. (2022). Multivariate analysis of food fraud: A review of NIR based instruments in tandem with chemometrics. Journal of Food Composition and Analysis, 107, 104343.

Béc, K. B., Grabska, J., & Huck, C. W. (2022). Miniaturized NIR spectroscopy in food analysis and quality control: Promises, challenges, and perspectives. Foods, 11(10), 1465.

Cruz-Tirado, J. P., dos Santos Vieira, M. S., Correa, O. O. V., Delgado, D. R., Angulo-Tisoc, J. M., Barbin, D. F., & Siche, R. (2024). Detection of adulteration of Alpaca (Vicugna pacos) meat using a portable NIR spectrometer and NIR-hyperspectral imaging. Journal of Food Composition and Analysis, 126, 105901.

Meena, D., Chakraborty, S., & Mitra, J. (2024). Geographical origin identification of red chili powder using NIR spectroscopy combined with SIMCA and machine learning algorithms. Food Analytical Methods, 1-19.

Zhu, R., Wu, X., Wu, B., & Gao, J. (2024). High-accuracy classification and origin traceability of peanut kernels based on near-infrared (NIR) spectroscopy using Adaboost-Maximum uncertainty linear discriminant analysis. Current Research in Food Science, 8, 100766.

Yuan, L., Meng, X., Xin, K., Ju, Y., Zhang, Y., Yin, C., & Hu, L. (2023). A comparative study on classification of edible vegetable oils by infrared, near infrared and fluorescence spectroscopy combined with chemometrics. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 288, 122120.

Quintelas, C., Rodrigues, C., Sousa, C., Ferreira, E. C., & Amaral, A. L. (2024). Cookie composition analysis by Fourier transform near infrared spectroscopy coupled to chemometric analysis. Food Chemistry, 435, 137607.

Jiang, Q., Zhang, M., Mujumdar, A. S., & Wang, D. (2023). Non-destructive quality determination of frozen food using NIR spectroscopy-based machine learning and predictive modelling. Journal of Food Engineering, 343, 111374.

Zhao, S., Jiao, T., Adade, S. Y. S. S., Wang, Z., Wu, X., Li, H., & Chen, Q. (2024). Based on vis-NIR combined with ANN for on-line detection of bacterial concentration during kombucha fermentation. Food Bioscience, 60, 104346.

Zuo, E., Sun, L., Yan, J., Chen, C., Chen, C., & Lv, X. (2022). Rapidly detecting fennel origin of the near-infrared spectroscopy based on extreme learning machine. Scientific Reports, 12(1), 13593.

Zhang, W., Kasun, L. C., Wang, Q. J., Zheng, Y., & Lin, Z. (2022). A review of machine learning for near-infrared spectroscopy. Sensors, 22(24), 9764.

Deng, Z., Wang, T., Zheng, Y., Zhang, W., & Yun, Y. H. (2024). Deep learning in food authenticity: Recent advances and future trends. Trends in Food Science & Technology, 104344.

Mishra, P., Passos, D., Marini, F., Xu, J., Amigo, J. M., Gowen, A. A., ... & Nordon, A. (2022). Deep learning for near-infrared spectral data modelling: Hypes and benefits. TrAC Trends in Analytical Chemistry, 157, 116804.

Zhao, M., Cang, H., Chen, H., Zhang, C., Yan, T., Zhang, Y., ... & Xu, W. (2023). Determination of quality and maturity of processing tomatoes using near-infrared hyperspectral imaging with interpretable machine learning methods. LWT, 183, 114861.

Castro, W., Saavedra, M., Castro, J., Tech, A. R. B., Chuquizuta, T., & Avila-George, H. (2024). Using recurrent neural networks to identify broken-cold-chain fish fillet from spectral profiles. Neural Computing and Applications, 36(8), 4377-4386.

Dong, F., Bi, Y., Hao, J., Liu, S., Yi, W., Yu, W., ... & Wang, S. (2024). A new comprehensive quantitative index for the assessment of essential amino acid quality in beef using Vis-NIR hyperspectral imaging combined with LSTM. Food Chemistry, 440, 138040.

Ozturk, S., Bowler, A., Rady, A., & Watson, N. J. (2023). Near-infrared spectroscopy and machine learning for classification of food powders during a continuous process. Journal of Food Engineering, 341, 111339.

Jo, S., Sohng, W., Lee, H., & Chung, H. (2020). Evaluation of an autoencoder as a feature extraction tool for near-infrared spectroscopic discriminant analysis. Food Chemistry, 331, 127332.

Yang, B., Chen, C., Chen, F., Chen, C., Tang, J., Gao, R., & Lv, X. (2021). Identification of cumin and fennel from different regions based on generative adversarial networks and near infrared spectroscopy. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 260, 119956.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. Proceedings of the 31st International Conference on Neural Information Processing Systems (pp. 6000–6010).

Jin, X., Zhou, J., Rao, Y., Zhang, X., Zhang, W., Ba, W., ... & Zhang, T. (2023). An innovative approach for integrating two-dimensional conversion of Vis-NIR spectra with the Swin Transformer model to leverage deep learning for predicting soil properties. Geoderma, 436, 116555.

Wang, Z., Zhang, J., Zhang, X., Chen, P., & Wang, B. (2022). Transformer model for functional near-infrared spectroscopy classification. IEEE Journal of Biomedical and Health Informatics, 26(6), 2559-2569.

Li, Y., Sun, H., Zheng, Y., Wei, Q., Chen, Z., Zhang, J., ... & Chen, F. (2024). Combined gramian angular difference field image coding and improved mobile vision transformer for determination of apple soluble solids content by Vis-NIR spectroscopy. Journal of Food Composition and Analysis, 131, 106200.

Liao, L., Lu, J., Wang, L., Zhang, Y., Gao, D., & Wang, M. (2024). CT-Net: an interpretable CNN-Transformer fusion network for fNIRS classification. Medical & Biological Engineering & Computing, 1-15.

Liu, Z., Wang, Y., Vaidya, S., Ruehle, F., Halverson, J., Soljacic, M., ... & Tegmark, M. (2024). Kan: Kolmogorov-Arnold networks. arXiv preprint arXiv:2404.19756.

Yu, R., Yu, W., & Wang, X. (2024). Kan or MLP: A fairer comparison. arXiv preprint arXiv:2407.16674.

Genet, R., & Inzirillo, H. (2024). Tkan: Temporal Kolmogorov-Arnold networks. arXiv preprint arXiv:2405.07344.

Vaca-Rubio, C. J., Blanco, L., Pereira, R., & Caus, M. (2024). Kolmogorov-Arnold networks (KANs) for time series analysis. arXiv preprint arXiv:2405.08790.

Xu, K., Chen, L., & Wang, S. (2024). Kolmogorov-Arnold Networks for Time Series: Bridging Predictive Power and Interpretability. arXiv preprint arXiv:2406.02496.

Hou, Y., Wu, J., & Feng, X. (2024). A Comprehensive Survey on Kolmogorov Arnold Networks (KAN). arXiv preprint arXiv:2407.11075.

Seydi, S. T. (2024). Unveiling the Power of Wavelets: A Wavelet-based Kolmogorov-Arnold Network for Hyperspectral Image Classification. arXiv preprint arXiv:2406.07869.

Malounas, I., Vierbergen, W., Kutluk, S., Zude-Sasse, M., Yang, K., Zhao, M., ... & Fountas, S. (2024). SpectroFood dataset: A comprehensive fruit and vegetable hyperspectral meta-dataset for dry matter estimation. Data in Brief, 52, 110040.




DOI: https://doi.org/10.31449/inf.v49i13.7417

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.