Non-destructive wood species classification and multiple characteristics evaluation of wood by near-infrared spatially resolved spectroscopy

2020 Fiscal Year Final Research Report

• Project/Area Number: 19K15886
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 40020:Wood science-related
• Research Institution: Nagoya University
• Principal Investigator: Ma Te 名古屋大学, 生命農学研究科, 特任助教 (70824316)
• Project Period (FY): 2019-04-01 – 2021-03-31
• Keywords: 近赤外空間分解分光法 / 木材の樹種判別 / 多波長の吸収・散乱特性 / 非破壊評価

Outline of Final Research Achievements

In this research, an optical system was developed to acquire multi-wavelength absorption and scattering characteristics from the spatial distribution of a point light source that used to illuminate wood sample. As a result, it was possible to create a calibration model that can be used to classify multiple wood species, and it was possible to improve the prediction accuracy compared to conventional spectroscopy. We also found the possibility of measuring wood tensile strain by the same spatially resolved spectroscopy. Furthermore, in connection with international joint research, we found the possibility of measuring the hardness of fruits by this method in a non-destructive and highly accurate manner. These results were submitted to two international journals (Holzforschung, Postharvest Biology and Technology) and once at a domestic conference (71st Japan Wood Society).

Free Research Field

農業工学

Academic Significance and Societal Importance of the Research Achievements

従来の測定法で得られる拡散反射スペクトルには木材の化学成分(吸収係数)と物理構造(散乱係数)の情報が混在しており、複雑な多変量解析によって構築した材質予測モデルの分光学的および物理学的な解釈が曖昧になることがある(実用化に向けての大きな課題)。本研究はこれまでのスペクトル解析では障害となっていた散乱情報を有効に活用して、分光学的な情報のみから独自のアルゴリズムによって木材含水率、引張ひずみなどの高精度予測に成功した。さらに細胞構造の違いによる散乱情報に着目することで樹種識別の可能性を見出した。また、本手法に必要となる分光装置は低価格化かつ小型化もでき、現場での実用化を期待できる。