基于高斯-马尔可夫随机场的板材表面纹理分析

阐述了高斯-马尔可夫随机场模型的基本原理,并针对板材表面纹理建立了高斯-马尔可夫随机场模型,以利用其对板材表面纹理进行分析;详细的研究了高斯-马尔可夫随机场模型的参数估计过程,在试验数据的基础上,验证了高斯-马尔可夫随机场模型在纹理分析中应用的可行性。     

基于高斯-马尔可夫随机场的木材表面缺陷类型识别

为了识别死节、活节和虫眼三种木材表面缺陷类型,本文采用高斯-马尔可夫随机场模型提取木材表面缺陷图像的纹理参数,结合缺陷区域的矩形度和伸长度两个几何特征,形成14维特征向量.设计三层BP神经网络来识别缺陷的类型.试验表明,三种缺陷的整体识别正确率达到96.67％,验证了该方法的有效性.     

基于高斯-马尔可夫随机场木材纹理特征的研究

将高斯—马尔可夫随机场(GMRF)引入木材纹理的研究,提取了二阶与五阶特征参数,并对二阶特征参数做了详细分析,得出通过θ2可以判断纹理的主方向,而结合θ1、θ2、θ3、θ4能够区分开木材的弦切和径切纹理。将五阶特征参数组成的特征向量输入给BP神经网络分类器,其分类识别率约为85%,表明了高阶GMRF参数对木材纹理描述的有效性。     

基于Markov随机场的木材表面缺陷图像分割方法

针对木材表面缺陷的彩色图像,建立Markov随机场分割模型,采用EM法估计了模型参数,采用模拟退火算法(SA)和条件选代算法(ICM)求解分割结果.试验表明,Markov随机场能对木材表面缺陷图像进行有效分割,但对死结图像存在欠分割现象.SA分割效果要优于ICM,但SA算法计算过程复杂,用时较长.     

基于空间灰度共生矩阵木材纹理分类识别的研究

以10种木材纹理样本为对象,研究了木材纹理参数体系的建立方法,并进行了分类识别的仿真实验。首先,针对木材纹理特点并结合类别可分性判据,构造了适于描述木材的空间灰度共生矩阵,并在此基础上提取了木材的11个纹理特征参数。其次,借助相关性分析对参数进行了特征选择,进而建立了能直接与人的感官对应的木材纹理参数体系。最后,利用 BP 神经网络分类器对木材样本进行了分类识别研究,识别率为87.50%,验证了参数体系的有效性,表明用本文提出的纹理参数体系对木材进行分类识别是可行的。     

小波变换的木材纹理在线分选

以180幅木材样本图片为对象,研究以小波变换方法提取特征参数,分析几种小波基的特点和性质,最终以对称性为依据,选择使用sym4小波对图像进行二级小波分解,可以得到一级水平细节HL1、垂直细节LH1、对角细节HH1,二级的近似LL2、水平细节HL2、垂直细节LH2、对角细节HH2共7个子图,提取整幅图像的熵和每个子图小波系数的均值及标准差作为特征参数。将木材纹理按照直纹、抛物线和乱纹3种纹理的分类标准,以BP神经网络作为分类器进行了木材纹理分类的验证,并与灰度共生矩阵的方法进行了对比。试验表明:采用小波变换的方法对木材纹理特征进行描述,不但提高了分类的准确率,重要的是缩短了运算时间,可以达到在线监测的要求。     

基于BP神经网络木材纹理分类的研究

应用 BP 神经网络对常见的10种木材纹理进行了分类研究,获得了比较满意的效果。首先,应用灰度共生矩阵提取了木材的纹理特征参数;其次,在此特征参数体系下,应用 BP 神经网络对木材纹理进行了分类研究,识别率达89%。     

基于LBP-DEELM的木材纹理分类算法

为解决传统木材纹理分类的准确率低且难度大的问题,提出了一种基于LBP-DEELM(局部二值-差分演化优化极限学习机)模型的木材纹理分类算法。在阐述局部二值算子(LBP)和差分演化优化极限学习机(DEELM)算法的基础上,使用均匀旋转不变的LBP模式提取纹理的特征值,结合差分演化算法进行极限学习机优化,通过训练得到每类纹理所对应的分类器模型参数,构造分类器,实现了对木材纹理准确高效的分类。实验结果表明,相比于BP神经网络,SVM支持向量机等分类算法,该模型的实验误差率为2%左右,准确率高,实用性强。     

应用空间灰度共生矩阵定量分析木材表面纹理特征

引入空间灰度共生矩阵对木材表面纹理进行定量分析 ,在对国内 5 0个树种径弦向纹理计算、分析的基础上得出结论 :像素点对间距d取 3,像素点对角度θ在径向纹理时取 0°,在弦向纹理时取 0°、4 5°和 135°的平均值对于反映木材纹理的特点较适宜。在 11种纹理特征参数的基础上归纳出 4个纹理主成分因子 ;讨论了主成分上木材纹理的分布规律和特点 ,并具体对木材的径、弦向纹理分别进行了分析 ,得出了各自的变化特点 ;提出了纹理综合值的计算方法 ,以及通过纹理综合值判定 2种纹理间相似性的方法。     

基于小波的木材纹理分频信息提取与分析

通过引入小波方法 ,对木材纹理进行了多尺度的频谱分解 ,并利用所得到的特征向量分析了水平、垂直和对角方向上的木材纹理频率分布特点 ,比较了针叶树材与阔叶树材、径向切面与弦向切面木材纹理的统计差异。并在试验基础上 ,提出了以小波分解子图像能量值的标准差进行木材纹理最佳分解尺度的筛选 ,探索出滤波长度取 8、分解尺度取 2对充分表现木材纹理特征最为适宜。同时还发现可将垂直中高频分量HL和低频分量LL的能量值作为木材纹理区别与归类的重要参数 ,将EHL ELH值作为木材纹理的方向性量度     

Assessment of wood surface roughness: comparison of tactile roughness and three-dimensional parameters derived using a robust Gaussian regression filter

Japanese oak and Japanese beech were sanded by hand with abrasive papers of varying grit number. Two three-dimensional parameters selected to characterize their surface roughness – one parameter for the distribution of roughness-profile peaks and the other for the relative area of the roughness-profile peaks above the threshold height – were compared against tactile roughness. The parameters were obtained from roughness profiles as determined by a robust Gaussian regression filter (RGRF) using seven cutoffs. The RGRF filtering process was adjusted specifically for the evaluation of wood surface roughness. Except for a cutoff wavelength of 0.25mm, the RGRF lent itself well to the determination of roughness profiles. No distortion of roughness profiles occurred around deep valleys, and there was a good correlation between the parameters and tactile roughness.Part of this study was presented at the 2002 Kansai Branch Office lecture meeting of the Japan Society for Precision Engineering, Kyotanabe, August 2002     

Evaluation of surface smoothness by laser displacement sensor 1: effect of wood species

The best methods for determining surface roughness in an industrial environment are of the noncontact variety, with reproduction of the profile. The objective of this work was to compare the roughness profile obtained by a contact stylus with a commercial laser displacement sensor (LDS). Measurements were done using 15 wood species with different densities and colors, based on which special triangle profiles were prepared. The accuracy of the laser sensor was examined by statistical analysis of roughness parameters measured from the profiles. Experimental results show that LDS profiles were imitated correctly. However, LDS accuracy depends on the scanned wood properties (density and color), installation position of the sensor, and profile shape. It was found that evaluation of dark and high-density wooden surfaces was imperfect.Part of this work was presented at the 50th Annual Meeting of the Japan Wood Research Society, Kyoto, April 2000; and at the 51st Annual Meeting of the Japan Wood Research Society, Tokyo, April 2001     

THE EFFECT OF WOOD SURFACE ROUGHNESS OF DIFFERENT SPECIES OF TREES ON WOOD COLOR IN WORKING OPERATION

Woodsurfaceroughnessisthemicrogeometrypropertycomposedbyalittlespacingandpeak-tovalley.Generallythesurfaceroughnesswascomposedbytheworkingmethodandotherfactorst123.Woodsurfaceroughnesswasnotonlyanimportantindexwhichmeasuredthesur-facesmoothextentofwoodproductandtheworkingqualitybutalsothewoodsur-fitceroughnessaffectedwoodsurfacefeel--lngandpsychologicalfeeling[3'4,lo,ll:.Th,'quantitativesurftlceproPertyparametersweretheimportantpartwhichmarkedthewoodsurfacevisualpsycho1ogicalmagni-tude,whichha…     

Feature Extraction of defects on wood surface

IntroductionAtpresent,usingthecomputerimageprocessil1gsys-temtofinishthefeatureextractiol1ofw0oddefectsisi11anexpIorat0rystage.ThismetI1odmaydecreasetI1ellmitati0noftl1eexperilnel1talcol1ditions,suChasIights,instrumentsandsomespccialmateriaIs.Itiscol1venic11ttousethiswayandthegoodresuItswilIbeacI1ieved.TherearethreetypicaIwo0dsurfacedefectst0bestudiedintheexperiment,whicI1arekn0t,boreI1oleal1ddecayofLarixandBeIlIlu.Tl1eworkist0separatetl1edefectsfromthewoodsurfaceimagesandprovideagoodba…     

CLASSIFICATION AND COMPARISON OF WOOD DRYING MODELS

The Various models established by many scholar in the near years areclassified and compared between them.     

Water sorption,surface structure and surface energy characteristics of wood composite fibres refined at different pressures

During fibre processing, wood fibres are subjected to a range of physical and chemical conditions sufficient to slightly alter their chemical composition and hence their ultimate performance when used in the manufacture of wood fibre-based composites. In order to better understand the effects of refiner conditions on material performance, wood fibres were subjected to processing at different refiner pressures (4, 6, 8 and 10 bar) and subsequently dried in a flash drier. The fibres were analysed for changes in surface area, surface energy, surface structure and water vapour sorption characteristics. The methods applied were nitrogen adsorption utilising the Brunauer–Emmett–Teller theory, inverse gas chromatography, scanning electron microscopy and dynamic vapour sorption. It was found that increasing refiner pressure resulted in fibres of lower surface area, accompanied by increasing dispersive surface energies up to operating refiner pressures of 8 bar. It was found with fibres refined at different pressures that as the refiner pressure increased the equilibrium moisture content of the fibre decreased at the set relative humidities. However, it was also noted that the hysteresis was not significantly different between each of the refiner pressures. The results suggest that different refiner pressures can be used to tune the surface characteristics which may be beneficial to product development and the improvement of the environmental profile of the wood fibre composites.     

STUDY ON MEASUREMENT OF WOOD SURFACEROUGHNESS BY COMPUTER VISION

The working processes,machining devices and tools,cutting amount,consumptionof materials,productivity and quality of products are directly affected wood surface roughness.This paper gives an extensive review of methods used previously to measure wood surfaceroughness,and concludes that computer vision is the most suitable.The preliminarystudy shows that computer vinon method has the advantages of a noncontact,three--dimensionalmeasurement,high speed and well correlates with stylus tracing method.The method can be usedin classification and in-time measurement of wood surface roughness after being improved.     

基于数字图像处理技术实现木材横纹压缩过程中构造学形态特征的研究

根据计算机视觉技术,结合横纹压缩木材构造形态特征,提出了适合于木材横纹压缩过程中构造形态特征的图像处理方法。对图像质量的完善较全面地进行了量化,充分表露出计算机视觉技术在实现木材横纹压缩过程中构造学形态特征动态变化分析中的优越性,实现了木材细胞边缘轮廓的跟踪,形态特征的提取,为木材科学的再发展奠定了理论基础。