Abstract
A new method is presented for measuring microfibril angles (MFAs) from transverse tree sections by using polarized optical microscopy combined with a liquid-crystal tunable filter. The MFA transition analysis of tree growth rings via generalized two-dimensional correlation is proposed. The unique features of two anatomically similar Cupressaceae species, Chamaecyparis obtusa S. & Z. and Thujopsis dolabrata S. & Z. and its variant, were extracted through MFA analysis. The technique efficiently visualized MFA distributions under wide field observation and uncovered seasonal differences. Classification results of supervised models exhibited 60% accuracy, despite featureless cross-sections of the conifers. Overall, MFA is a promising index that can identify specific features of wood species.
Highlights
The Japanese culture has prospered in its utilization of wood resources for many purposes
Two main peaks were observed in low (0–20 degree) and high (40–50 degree) microfibril angles (MFAs) regions
The MFA increased around pit apertures similar to the bordered pits between tracheids [22] and the cross-field pitting in the joint of tracheid and ray parenchyma [23], because a flow of cellulose microfibrils circumvented pit apertures
Summary
The Japanese culture has prospered in its utilization of wood resources for many purposes. The ability to distinguish between C. obtusa and T. dolabrata is important for understanding Japanese history, culture, and politics. Dolabrata is important for understanding Japanese history, culture, and politics Both trees are nearly identical with respect to the occurrence of axial parenchyma with resin-like substances tangentially aligned in latewood, as well as the gradual transition from earlywood to latewood. The microfibril angle (MFA) is defined as the angle between the direction of cellulose microfibrils and the longitudinal axis in an S2 wall It has the potential for revealing specific features in wood species. A completely new method for wood classification based on MFA is discussed and applied to the identification of C. obtusa and T. dolabrata. The maps were inputted into multivariate classifiers, K-Nearest-Neighbor (KNN) [8] and random forest [9]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
