Use of regularizing methods in modeling and prediction of wood weathering as a response to UV radiation

摘要

Samples of norway spruce (Picea abies) were exposed to various amounts of natural orarti cial UV radiation and recorded with a NIR hyperspectral camera. Multivariate statis-tical analysis using Tikhonov regularization, Principal Component Regression and PartialLeast squares were applied to the resulting data. Results were mixed, but with consis-tently better models for the samples exposed to arti cial radiation, able to predict exposureamounts with a reasonable degree of accuracy and generalizability. Tikhonov regularizationperformed better or on par with PCR or PLS in most cases, and showed a tendency tobetter predict new data. Samples exposed to natural weathering yielded signi cantly worseresults, presumably related to the myriad of interfering, unquanti ed phenomena present inthe environment. It was thus concluded that the weathering e ect on wooden surfaces iswell reflected in the NIR spectrum. NIR hyperspectral imaging holds promise as a usefultool in the further investigations into the subject, with the caveat that advancements needto be made in the statistical modeling of the data, to make it more robust and applicableoutside of a purely research oriented environment.