基于图像分析的杂草分形维数计算

介绍了分形维数的理论计算公式和试验公式，分析了计盒维数的计算方法，运用Matlab软件设计了图像的处理和基于图像的计盒维数的计算程序。随机地选择了4种双子叶杂草和两种单子叶杂草共6种杂草作为研究对象，运用自行设计的程序分别计算了杂草叶的RGB三色图像的分形维数并以其平均值作为杂草叶的分形维数。研究结果表明该文给出的分形维数的计算方法可靠；研究结果也验证了杂草叶具有分形特征；研究结果还说明不同种类的杂草 ，其叶的分形维数明显不同，可以用分形维数为特征参数识别杂草。     

大理河流域地貌分形特征空间变异研究

以1：5万地形图为数据源,通过建立流域数字高程模型,采用基于变分法的分形维数计算模型和ArcGIS的窗口分析功能,对大理河流域中的10条子流域地貌分形维数进行了计算,分析了其空间变异特点。结果表明,大理河流域地貌分形维数从上游至下游呈明显增加趋势,右岸流域地貌分形维数平均值大于左岸,地貌分形维数较好地揭示了大理河流域地貌破碎程度和复杂程度的空间变异特点。本研究对于促进地貌分形维数在地貌学及水土流失预报等领域的应用研究具有重要意义。     

大同市南郊区治沙工程对区域景观生态影响的评价

景观生态影响评价是非污染生态环境影响评价的重要组成部分，对于区域生态治理工程尤其关键。该文以大同市南郊区2002～2003年的生态治沙工程为例，以2001年的1∶50000土地利用现状图和2002年、2003年的1∶10000治沙工程设计图为基础数据，对工程实施前后的区域景观格局与景观生态影响进行对比分析。结果显示，大同市南郊区2002～2003年治沙工程总体使该区景观向着有利于沙化逆转的方向发展，表现为研究区景观水平的分形维值降低，未利用地以及耕地等占优势且高度聚集的状态得以改善，使景观的多样性提高；使耕地和林地景观类型水平分形维下降，未利用地景观类型水平分形维值升高；山区半山区乡镇的景观水平分形维值降低，耕地和未利用地的聚集度和优势度下降，林地和牧草地的聚集度和优势度上升。     

基于分形维数和HHT的蚜虫刺吸电位波形机器识别

昆虫刺吸电位(electricalpenetrationgraph,EPG)技术在研究刺吸式昆虫取食行为、昆虫与植物的关系、昆虫传毒机制、作物抗虫机制等方面得到了广泛的应用,然而EPG信号的识别和分析一直是靠人工进行,亟需波形自动识别系统以提高分析效率。刺吸式昆虫取食植物时,产生的EPG波形跟昆虫和植物的种类有关,不同类别的刺吸式昆虫EPG波形差别很大,即使是同种类型的EPG波形其幅值和频率间也会有差异,这给EPG波形的机器识别带来了困难。该文以蚜虫的EPG信号为研究对象,对np波、pd波、E1波、E2波、G波、C波和F波的特征提取和分类识别进行了研究,提出了一种基于分形维数、希尔伯特-黄变换(hilbert-huang transform,HHT)和决策树的EPG波形识别方法。首先对EPG仪器采集得到的信号进行去噪预处理,分别提取分形维数和HHT共10维特征,组成不同维数的特征向量进入决策树分类器进行对比试验。试验结果表明,可采用分形盒维数、Hurst指数、前2层的谱质心和加权频率融合的6维特征向量获得较高的识别率。在EPG波形的机器识别中采用6维特征向量输入的决策树进行分类,通过对4组不同样本进行测试,得到了92.14%、89.29%、95%和89.29%的识别率,平均识别率为91.43%。研究结果表明该文提出的基于分形维数和HHT的特征提取方法以及构建的决策树分类器具有一定的可行性,可为研发EPG信号自动识别分析系统提供理论参考。     

不同风沙土壤颗粒的分形特征

应用土壤颗粒的重量与粒径分布原理来描述了古尔班通古特沙漠地风沙土壤颗粒的分形特征。通过对10种样品颗粒的机械组成进行分析,分别计算出了它们的分形维数(D=2.3237～2.9347),并分析了其与流动风沙土、半固定风沙土和固定风沙土之间的关系。分析结果表明,风沙土壤结构具有明显的分形特征,其粒径分布分形维数为2～3。土粒表面分形维数与2～0.2mm间的土粒含量存在显著的负相关;而与0.02～0.002mm和<0.002mm的颗粒含量存在显著的正相关,表现为随着土壤质地从流动风沙土、半固定风沙土到固定风沙土的变化,其土粒表面的平均分形维数呈依次增高。土粒表面分形维数与三种典型风沙土壤有机质含量为极显著正相关,而与其硬度为显著负相关。     

土壤颗粒的分形特征及其应用

本文应用土壤颗粒的质量分布原理来描述土壤颗粒的分形特征。通过对10种土壤颗粒的机械组成进行分析,分别计算出它们的分形维数（D=2.489～2.896）,并分析了其与土壤质地之间的关系。同时对土壤颗粒分形维数与所对应土壤的幂函数型水分特征曲线的拟合分形维数进行比较分析,建立了二者之间的相关关系。结果表明:分形维数的大小反映了土壤质地中粘粒、粉粒和砂粒含量的变化,随粘粒含量的增多分形维数增大,随砂粒含量的增多分形维数减小;同时土壤颗粒分形维数与所对应的水分特征曲线的拟合分形维数呈现出良好的一致性,因而对所研究的土壤而言,可应用土壤颗粒的质量分形维数结合幂函数模型来估算土壤水分特性曲线。     

关中东部TSG全新世剖面粒度分形特征及古气候意义

利用分形理论计算可知,关中东部TSG全新世剖面的土壤粒度分布具有分形特征,剖面分形维数值在2.5443~2.7123之间,其中古土壤S0的分形维数均值最大。粒度分形维数的高值指示暖湿的气候,低值指示干冷的气候。分形特征所揭示的关中东部地区的环境演变过程为:全新世早期,粒度分形维数值相对偏低,风尘堆积占优势,为干旱寒冷的环境,成壤作用较弱;全新世中期,粒度分形维数值偏大,其平均值为2.6713,指示温暖湿润的气候,成壤作用较强;全新世晚期以来,粒度分形维数值降低,土壤颗粒粒径增大,气候变得相对干旱,成壤作用也减弱,但是全新世晚期的分形维数值高于末次冰期。     

牛肉大理石花纹计盒维和信息维的测定

对牛肉大理石花纹几何特征进行定量描述，是牛肉大理石花纹等级自动判定系统的关键技术。通过对牛肉大理石花纹图像分形特征的分析及分形维数的测定，试图以分形维对牛肉大理石花纹的分布特征进行定量描述。作者以中国及美国牛肉分级系统中的大理石花纹标准图版为研究对象，通过图像预处理，将其转化为二值数字图像，然后根据变粗视化程度的像素点覆盖法，编写出计盒维和信息维的测定程序，计算出上述每一幅图像在不同标尺下的非空盒个数及信息量，再根据标尺与非空盒个数及信息量的关系，分析图像的分形特征；通过回归分析的方法，测定出图像的计盒维数和信息维数。试验结果显示，在一定的标度范围内，牛肉大理石花纹标准图像具有分形性，并且显示出多重分形特征；图像的计盒维数和信息维数均呈现出随着牛肉大理石花纹等级的增加而增大的变化趋势。     

灌区灌溉渠系分形特征研究

灌溉渠系是一种分形体,渠系的某部分渠段在某种程度上包含了整个渠系的信息。该文将分形理论引入到灌溉渠系的研究中,对其分形特征和分维数进行了分析,并运用盒计数法计算了漳河灌区灌溉渠系的分形维数。在此基础上,论证了渠系分形维数与其灌水模数的相关性,为灌区渠系布置优化设计提供了一定的理论依据。     

土壤粒径分布单重分形与孔隙单重分形

土壤颗粒大小的分布是重要的土壤物理性质,对土壤水、气、热传导特性有着显著的影响。鉴于土壤颗粒分形在分析和描述岩土介质多孔结构的优势,本文通过对5种不同质地的土样进行颗粒分析与水分特征曲线的测试。结果表明:土壤结构定量化表征的方法就是确定土壤结构的分形维数,针对不同地区的土壤,可以通过土壤粒径分布分形维数体现土壤粒径分布的情况,分形维数越大,颗粒粒径越小,细粒含量越高,质地越发呈现不均匀性;且可利用土壤粒径分形维数估算土壤孔隙分形维数。为建立土壤粒径分布模型提供数据支持,同时也为推进土壤结构的研究进展提供了新的方向。     

Characterization of pine nuts in the U.S. market, including those associated with "pine mouth", by GC-FID

Taste disturbances following consumption of pine nuts, referred to as "pine mouth", have been reported by consumers in the United States and Europe. Nuts of Pinus armandii have been associated with pine mouth, and a diagnostic index (DI) measuring the content of Δ5-unsaturated fatty acids relative to that of their fatty acid precursors has been proposed for identifying nuts from this species. A 100 m SLB-IL 111 GC column was used to improve fatty acid separations, and 45 pine nut samples were analyzed, including pine mouth-associated samples. This study examined the use of a DI for the identification of mixtures of pine nut species and showed the limitation of morphological characteristics for species identification. DI values for many commercial samples did not match those of known reference species, indicating that the majority of pine nuts collected in the U.S. market, including those associated with pine mouth, are mixtures of nuts from different Pinus species.     

Purification and characterization of the 7S vicilin from Korean pine (Pinus koraiensis)

Pine nuts are economically important as a source of human food. They are also of medical importance because numerous pine nut allergy cases have been recently reported. However, little is known about the proteins in pine nuts. The purpose of this study was to purify and characterize pine nut storage proteins. Reported here is the first detailed purification protocol of the 7S vicilin-type globulin from Korean pine (Pinus koraiensis) by gel filtration, anion exchange, and hydrophobic interaction chromatography. Reducing SDS-PAGE analysis indicated that purified vicilin consists of four major bands, reminiscent of post-translational protease cleavage of storage proteins during protein body packing in other species. The N-terminal ends of vicilin peptides were sequenced by Edman degradation. Circular dichroism (CD) and differential scanning calorimetry (DSC) analyses revealed that pine nut vicilin is stable up to 80 degrees C and its folding-unfolding equilibrium monitored by intrinsic fluorescence can be interpreted in terms of a two-state model.     

Use of the chloroplast gene ycf1 for the genetic differentiation of pine nuts obtained from consumers experiencing dysgeusia

Pine nuts are a part of traditional cooking in many parts of the world and have seen a significant increase in availability/use in the United States over the past 10 years. The U.S. Food and Drug Administration (US FDA) field offices received 411 complaints from U.S. consumers over the past three years regarding taste disturbances following the consumption of pine nuts. Using analysis of fatty acids by gas chromatography with flame ionization detection, previous reports have implicated nuts from Pinus armandii (Armand Pine) as the causative species for similar taste disturbances. This method was found to provide insufficient species resolution to link FDA consumer complaint samples to a single species of pine, particularly when samples contained species mixtures of pine nuts. Here we describe a DNA based method for differentiating pine nut samples using the ycf1 chloroplast gene. Although the exact cause of pine nut associated dysgeusia is still not known, we found that 15 of 15 samples from consumer complaints contained at least some Pinus armandii, confirming the apparent association of this species with taste disturbances.     

Update on the healthful lipid constituents of commercially important tree nuts

Uncharacteristic of most whole foods, the major component of tree nuts is lipid; surprisingly, information on the lipid constituents in tree nuts has been sporadic and, for the most part, not well reported. Most published papers focus on only one nut type, or those that report a cultivar lack a quality control program, thus making data comparisons difficult. The present study was designed to quantify the healthful lipid constituents of 10 different types of commercially important tree nuts (i.e., almonds, black walnuts, Brazil nuts, cashews, English walnuts, hazelnuts, macadamias, pecans, pine nuts, and pistachios) according to standardized, validated methods. The total lipid content of each nut type ranged from 44.4 ± 1.9% for cashews to 77.1 ± 1.7% for macadamias. As expected, the major fatty acids present in the tree nuts were unsaturated: oleic (18:1 ω9) and linoleic (18:2 ω6) acids. A majority of the lipid extracts contained <10% saturated fatty acids with the exceptions of Brazil nuts (24.5%), cashews (20.9%), macadamias (17.1%), and pistachios (13.3%). The total tocopherol (T) content ranged from 1.60 ± 1.27 mg/100 g nutmeat in macadamias to 32.99 ± 0.78 in black walnuts. The predominant T isomers in the nut types were α- and γ-T. Tocotrienols were also detected, but only in 6 of the 10 nut types (i.e., Brazil nut, cashews, English walnuts, macadamias, pine nuts, and pistachios). In most cases, total phytosterol contents were greater in the present study than reported in peer-reviewed journal papers and the USDA National Nutrient Database for Standard Reference, which is attributed to total lipid extraction and the inclusion of steryl glucosides in the analysis; the levels were highest for pistachios (301.8 ± 15.4 mg/100 g nutmeat) and pine nuts (271.7 ± 9.1 mg/100 g nutmeat). Minor sterols were also quantified and identified using GC-FID and GC-MS techniques.     

冀北山地典型森林土壤颗粒分形特征

采用野外调查与室内分析相结合的方法,研究冀北山地典型森林土壤颗粒的分形特征与土壤物理化学性质的关系。结果表明:冀北山地7个典型森林土壤颗粒的分形维数介于2.057 0～2.379 3之间,分形维数随土层深度增加而呈减小趋势。土壤颗粒分形维数与细黏粒、细粉粒和中粉粒含量呈极显著正相关关系,与粗粉粒含量呈负相关关系,与黏粒含量相关性不显著;土壤分形维数与土壤容重呈线性正相关关系,与最大持水量、总孔隙度呈现线性负相关关系,与田间持水量相关性不明显。与速效氮和有机质呈显著负相关,与全氮、速效钾呈负相关,与pH无相关性。     

Speciation of arsenic in different types of nuts by ion chromatography-inductively coupled plasma mass spectrometry

In this work the quantitative determination and analytical speciation of arsenic were undertaken in different types of nuts, randomly purchased from local markets. The hardness of the whole nuts and high lipid content made the preparation of this material difficult for analysis. The lack of sample homogeneity caused irreproducible results. To improve the precision of analysis, arsenic was determined separately in nut oil and in the defatted sample. The lipids were extracted from the ground sample with the two portions of a mixture of chloroform and methanol (2:1). The defatted material was dried and ground again, yielding a fine powder. The nut oil was obtained by combining the two organic extracts and by evaporating the solvents. The two nut fractions were microwave digested, and total arsenic was determined by inductively coupled plasma mass spectrometry (ICP-MS). The results obtained for oils from different types of nuts showed element concentration in the range 2.9-16.9 ng g(-)(1). Lower levels of arsenic were found in defatted material (<0.1 ng g(-)(1) with the exception of Brazil nuts purchased with and without shells, 3.0 and 2.8 ng g(-)(1) respectively). For speciation analysis of arsenic in nut oils, elemental species were extracted from 2 g of oil with 12 mL of chloroform/methanol (2:1) and 8 mL of deionized water. The aqueous layer, containing polar arsenic species, was evaporated and the residue dissolved and analyzed by ion chromatography-ICP-MS. The anion exchange chromatography enabled separation of As(III), dimethylarsinic acid (DMAs(V)), monomethylarsonic acid (MMAs(V)), and As(V) within 8 min. Several types of nuts were analyzed, including walnuts, Brazil nuts, almonds, cashews, pine nuts, peanuts, pistachio nuts, and sunflower seeds. The recovery for the speciation procedure was in the range 72.7-90.6%. The primary species found in the oil extracts were As(III) and As(V). The arsenic concentration levels in these two species were 0.7-12.7 and 0.5-4.3 ng g(-)(1), respectively. The contribution of As in DMAs(V) ranged from 0.1 +/- 0.1 ng g(-)(1) in walnuts to 1.3 +/- 0.3 ng g(-)(1) in pine nuts. MMAs(V) was not detected in almonds, peanuts, pine nuts, sunflower seeds, or walnuts, and the highest concentration was found in pistachio nuts (0.5 +/- 0.2 ng g(-)(1)).     

不同耕作年限水稻土土壤颗粒的体积分形特征研究

应用基于土壤颗粒体积分布确定土壤颗粒大小分布分形维数的方法,对浙江慈溪耕作年限分别为50、100、500、700、1000和2000年的水稻土土壤颗粒体积分形特征进行了研究.结果表明:同基于土壤颗粒质量分布的分形维数相比,基于土壤颗粒体积分布的分形维数计算不需要密度假设,更具有合理性.不同年限水稻土的土壤颗粒大小分布能被体积分形模型显著拟合(p＜0.01).土壤颗粒体积分形维数与黏粒的体积百分含量之间呈显著正相关(p＜0.01),与粉粒和砂粒的体积百分含量之间呈显著负相关(p＜0.01),其中与黏粒含量关系最为密切.耕作年限对水稻土土壤颗粒的体积分形维数和分形模型的拟合度均有一定影响.与50 ～ 500年水稻土相比,700 ～ 2000年水稻土剖面中土壤颗粒体积分形维数的变异性和均值较高,但其拟合决定系数则较低;耕作年限会对水稻土剖面中黏粒的含量和分布产生影响,进而影响到剖面土壤的颗粒体积分形特征;对于同一质地类型的土壤,耕作年限越长,分形维数越高.     

基于图像法和粒度分形模型的打叶过程烟片结构在线检测

为实现烟叶采收后初加工过程打叶质量的实时监测与调控,该文采用图像分析法在线检测烟叶打后叶片结构,即叶片面积分布,应用分形理论建立烟叶打后叶片面积分布的分形模型,用分维数表征叶片面积分布特征。结果表明:分形模式计算的打后叶片面积分布结果与实测值相关系数较高,粒度分形模式可较准确描述烟叶打后叶片面积分布;分维数可表征打后叶片面积分布均匀程度,可反映出不同等级烟叶打叶特性差异;分维数与目前打叶过程的主要工艺控制指标大中片率成线性关系,相关系数大于0.90。基于图像法和粒度分形模型建立的打叶过程烟叶结构在线检测方法,可用于打叶质量的实时监测与调控。     

成都平原土壤颗粒分形特征及应用

采用分形模型研究了成都平原土壤颗粒分形维数的变化特征以及分形维数与土壤性状间的关系。结果表明成都平原土壤颗粒分形维数(0￣40cm)介于2.903￣3.005之间。在水平方向上东北方的中江县最大,西南方向的温江区和双流县次之,北方的绵竹市最小。土壤颗粒分形维数是反映土壤质地的一个较好的指标,重点反映粘粒含量,其次是粉粒含量。分形维数的变化能很好的表征平原区土壤化学、物理性状的变化趋势,可以作为评价土壤结构、土壤肥力诊断的一个综合指标。     

石羊河尾闾水面形成区土壤颗粒的分形特征

青土湖是石羊河的尾闾湖,观测分析了距青土湖水面边缘0-500m范围内的土壤颗粒粒径分布、分形维数及其与土壤颗粒含量的关系。结果表明,0-20cm和20-40cm土层土壤颗粒的分形维数范围分别在2.3708~2.8796和2.3590~2.7509。在0~150m范围内,0-20cm土层土壤粒径分形维数与水面间距呈缓慢增加趋势,在150m处达到最大(2.8796),随后逐渐降低而20-40cm土层在0~100m范围内呈缓慢增加趋势,在100m处分形维数达到最大(2.7509)土壤颗粒分形维数大小与土壤质地的细粒化有一致的变化趋势,且与大于0.05mm粒径含量呈极显著负相关关系,而与黏粒含量、粉粒含量呈极显著正相关关系。小于0.05mm粒径物质含量的增加和>0.05mm粒径物质含量的降低共同导致了土壤颗粒分形维数在水面形成150m范围内的增大,说明水面形成对土壤粒径的影响范围在0~150m。