数字图像相关技术表征木竹材干缩湿胀的适用条件

Applicability of digital image correlation on characterizing shrinkage and swelling of wood and bamboo

数字图像相关技术(DIC)是表征木竹材干缩湿胀的有效手段。传统DIC分析需对木竹材表面施加人工散斑点,近年来也有报道利用管胞、木纤维等天然孔隙作为自然散斑点进行DIC分析,但自然散斑与人工散斑表征木竹材干缩湿胀是否具有差异尚无定论,采用自然散斑对不同树种的干缩湿胀进行表征时的适用条件也尚不明确。因此,以人工林杨木、杉木和毛竹为研究对象,对比了2种散斑方式(自然散斑、人工散斑)DIC与实测法对干缩湿胀变形的表征差异,讨论了采用自然散斑进行干缩湿胀表征时的适用条件。研究结果表明,在水分解吸、吸着过程中,2种散斑方式对木竹材干缩、湿胀变形的表征存在差异,且该差异与树种、纹理方向、相对位置等因素有关。相较于人工散斑,自然散斑与实测结果无显著差异,即自然散斑更适合表征木竹材的干缩湿胀变形。管胞、木纤维、竹薄壁细胞以及维管束均可作为自然散斑点表征干缩湿胀变形。根据作为自然散斑点的细胞类型,调整DIC相应的像素点尺寸和步长:像素点尺寸需调整至该类型细胞(腔)尺寸的1/10~1/5;步长需调整至确保子集移动距离与该类型细胞(腔)尺寸相近。研究结果为准确利用DIC表征木竹材干缩湿胀变形提供了理论依据与实际参考。

Digital image correlation(DIC) has emerged as a robust and versatile tool for assessing the dimensional changes,i.e.,shrinkage and swelling,in wood and bamboo.Conventionally,DIC requires the application of artificial speckle patterns on the material surface to track deformation under varying environmental or loading conditions.However,recent studies have explored the feasibility of utilizing inherent features like vessels and wood fibers as natural speckles for DIC analysis.This approach aims to determine whether natural speckles can provide comparable or superior accuracy compared to artificial patterns in characterizing the dynamic changes in wood and bamboo.Despite these advancements,conclusive evidence regarding the comparative efficacy of natural versus artificial speckle patterns in DIC analysis remains inconclusive.Moreover,the optimal conditions for employing natural speckles across different tree species are poorly defined.This study addressed these gaps by focusing on poplar,Chinese fir,and moso bamboo.Moisture desorption tests from 97% relative humidity(RH) to 0% RH were conducted and subsequent adsorption tests from 0% RH to 97% RH were carried out.Shrinkage and swelling were evaluated during moisture desorption and adsorption tests.Specifically,this study compared the performances of natural and artificial speckle methods in DIC analysis and validated their accuracy by the juxtaposing hygro-deformation analysis with experimental measurements.Significant differences between natural and artificial speckle methods were observed during moisture desorption and absorption,which was influenced by factors such as tree species,grain orientation,and relative radial position.Remarkably,results obtained from natural speckle patterns closely mirrored experimental outcomes,suggesting their superiority in capturing the hygro-deformation in wood and bamboo.Natural features like vessels,wood fibers,bamboo parenchyma cells,and vascular bundles proved effective as reliable markers for DIC analysis,underscoring their potential as preferred alternatives to artificial patterns.To optimize DIC effectiveness,this study recommended adjusting pixel size relative to the size of these natural features,typically ranging between 1/10 to 1/5 of their dimensions,with the step size approximating the feature size to ensure precise subset movement tracking.These findings not only provide a robust theoretical framework but also offer practical guidelines for accurately characterizing shrinkage and swelling at the tissue scale in wood and bamboo using DIC.Such insights are pivotal for enhancing measurement precision and reliability across various applications,thereby facilitating informed decision-making in industries reliant on dimensional stability and performance of wood,bamboo,and other plant materials.