Compression strength prediction of Xylosma racemosum using a transfer learning system based on near-infrared spectral data

Abstract

A transfer learning system was designed to predict Xylosma racemosum compression strength. Near-infrared (NIR) spectral data for Acer mono and its compression strength values were used to resolve the weak generalization problem caused by using a X. racemosum dataset alone. Transfer component analysis and principal component analysis are domain adaption and feature extraction processes to enable the use of A. mono NIR spectral data to design the transfer learning system. A five-layer neural network relevant to the X. racemosum dataset, was fine-tuned using the A. mono dataset. There were 109 A. mono samples used as the source dataset and 79 X. racemosum samples as the target dataset. When the ratio of the training set to the test set was 1:9, the correlation coefficient was 0.88, and mean square error was 8.84. The results show that NIR spectral data of hardwood species are related. Predicting the mechanical strength of hardwood species using multi-species NIR spectral datasets will improve the generalization ability of the model and increase accuracy.

Keywords

Xylosma racemosum / Compression strength prediction / Near-infrared spectroscopy / Transfer learning system / TCA–PCA

References

[1] Borgwardt KM, Gretton A, Rasch MJ, Kriegel HP, Schölkopf B, Smola AJ. Integrating structured biological data by kernel maximum mean discrepancy. Bioinformatics, 2006, 22(14): e49-e57.

[2] Cook D, Feuz KD, Krishnan NC. Transfer learning for activity recognition: a survey. Knowl Inf Syst, 2013, 36(3): 537-556.

[3] Li J. Wood Science, 2002, Beijing: Higher Education Press 304 341

[4] Liang H, Cao J, Tu W, Lin X, Zhang Y. Nondestructive determination of the compressive strength of wood using near-infrared spectroscopy. BioResources, 2016, 11: 7205-7213.

[5] Lu J, Behbood V, Hao P, Zuo H, Xue S, Zhang G. Transfer learning using computational intelligence: a survey. Knowl-Based Syst, 2015, 80: 14-23.

[6] Pan SJ, Tsang IW, Kwok JT, Yang Q. Domain adaptation via transfer component analysis. IEEE Trans Neural Netw, 2010, 22(2): 199-210.

[7] Rakotovololonalimanana H, Chaix G, Brancheriau L, Ramamonjisoa L, Ramananantoandro T, Thevenon MF. A novel method to correct for wood MOE ultrasonics and NIRS measurements on increment cores in Liquidambar styraciflua L. Ann For Sci, 2015, 72(6): 753-761.

[8] Satoru T, Hikaru K. A review of recent application of near infrared spectroscopy to wood science and technology. J Wood Sci, 2015, 61: 213-220.

[9] Schimleck LR, Mora C, Daniels RF. Estimation of the physical wood properties of green Pinus taeda radial samples by near infrared spectroscopy. Can J For Res, 2003, 33(12): 2297-2305.

[10] Todorović N, Popović Z, Milić G. Estimation of quality of thermally modified beech wood with red heartwood by FT-NIR spectroscopy. J Wood Sci Technol, 2015, 49(3): 527-549.

[11] Wang C, Mahadevan S (2011) Heterogeneous domain adaptation using manifold alignment// IJCAI 2011. In: Proceedings of the 22nd international joint conference on artificial intelligence, Barcelona, Catalonia, Spain, July 16–22, 2011. DBLP, 2011

[12] Watanabe K, Kobayashi I, Matsushita Y, Saito S, Kuroda N, Noshiro S. Application of near-infrared spectroscopy for evaluation of drying stress on lumber surface: a comparison of artificial neural networks and partial least squares regression. Dry Technol, 2014, 32(5): 590-596.

[13] Wold S, Esbensen K, Geladi P. Principal component analysis. J Chemom Intell Lab Syst, 1987, 2(1–3): 37-52.

[14] Yosinski J, Clune J, Bengio Y, Lipson H (2014) How transferable are features in deep neural networks? In: Advances in neural information processing systems 27 (NIPS’ 14), NIPS Foundation, 2014, 3320–3328. https://papers.nips.cc/paper/5347-how-transferable-are-features-in-deep-n%E2%80%A6.