基于Bandelet变换的图像去噪算法研究

Bandelet变换是一种信号稀疏表示方法。与二维小波变换无法有效利用图像几何特征不同的是,Bandelet变换通过计算图像的几何流并在每个局部区域内选择最佳方向实施小波变换。本文分析了Bandelet变换的基本原理并将Bandelet变换应用于图像去噪,提出一种基于Bandelet变换的图像去噪算法并在MATLAB中实现,结果验证了本文算法比小波变换具有更好的性能。本文算法的去噪图像更加平滑,视觉效果更佳,峰值信噪比(PSNR)、等效视数(ENL)等评价指标也更好。     

玉米光谱小波奇异熵的铜污染监测研究

不同浓度铜离子(Cu2+)胁迫梯度下的玉米叶片光谱存在细微差异,但传统的光谱相似性测度方法难以区分健康的与受污染的玉米叶片光谱间的变异信息.为此,本文结合小波变换和奇异熵理论,构建小波奇异熵(Wavelet Singular Entropy,WSE)方法用于Cu2+胁迫下的玉米污染程度分析与污染监测.通过设置不同浓度Cu2+胁迫梯度下的玉米盆栽实验,依据玉米叶片的光谱测量数据和所测定的叶片中Cu2+含量,并与绿峰高度(Green-peak Height,GH)和红谷吸收深度(Red-valley Depth,RD)光谱特征参数法污染监测应用结果进行对比分析,研究WSE方法的Cu2+污染监测效果.结果表明,WSE值与Cu2+胁迫梯度、玉米叶片中Cu2+含量呈显著正相关,WSE值大小能够较好地表征叶片光谱的污染信息奇异性及其污染程度.     

基于LabVIEW和MATLAB的矿山微震信号小波分析与研究

为满足矿山微震监测信号实时在线处理要求,提出了基于可视化软件LabVIEW和数学计算软件MATLAB联合编程的矿山微震信号小波去噪方法。该方法利用LabVIEW中的MATLAB Script节点调用MATLAB实现信号小波分析,将LabVIEW、MATLAB和小波分析三者功能有机结合,从而实现了信号在友好界面的前提下快速准确地进行小波变换处理,大大提高了微震信号在线处理效率。通过对实际矿山微震信号采用4种不同小波阈值方法进行去噪分析,显示了小波分析具有良好的去噪能力,其中无偏似然估计阈值方法的去噪效果更优。     

Removal of polycyclic aromatic hydrocarbons from aqueous solution by raw and modified plant residue materials as biosorbents

Removal of polycyclic aromatic hydrocarbons （PAHs）, e.g., naphthalene, acenaphthene, phenanthrene and pyrene, from aqueous solution by raw and modified plant residues was investigated to develop low cost biosorbents for organic pollutant abatement. Bamboo wood, pine wood, pine needles and pine bark were selected as plant residues, and acid hydrolysis was used as an easily modification method. The raw and modified biosorbents were characterized by elemental analysis, Fourier transform infrared spectroscopy and scanning electron microscopy. The sorption isotherms of PAHs to raw biosorbents were apparently linear, and were dominated by a partitioning process. In comparison, the isotherms of the hydrolyzed biosorbents displayed nonlinearity, which was controlled by partitioning and the specific interaction mechanism. The sorpfion kinetic curves of PAHs to the raw and modified plant residues fit well with the pseudo second-order kinetics model. The sorption rates were faster for the raw biosorbents than the corresponding hydrolyzed biosorbents, which was attributed to the latter having more condensed domains （i.e., exposed aromatic core）. By the consumption of the amorphous cellulose component under acid hydrolysis, the sorption capability of the hydrolyzed biosorbents was notably enhanced, i.e., 6-18 fold for phenanthrene, 6-8 fold for naphthalene and pyrene and 5-8 fold for acenaphthene. The sorpfion coefficients （Kd） were negatively correlated with the polarity index [（O＋N）/C], and positively correlated with the aromaticity of the biosorbents. For a given biosorbent, a positive linear correlation between logKoc and logKow for different PAHs was observed. Interestingly, the linear plots of logKoc-logKow were parallel for different biosorbents. These observations suggest that the raw and modified plant residues have great potential as biosorbents to remove PAHs from wastewater.     

Dyes adsorption using a synthetic carboxymethyl cellulose-acrylic acid adsorbent

Removal of noxious dyes is gaining public and technological attention. Herein grafting polymerization was employed to produce a novel adsorbent using acrylic acid and carboxymethyl cellulose for dye removal. Scanning electron microscopy and Fourier-transform infrared spectroscopy verified the adsorbent formed under optimized reaction conditions. The removal ratio of adsorbent to Methyl Orange, Disperse Blue 2BLN and malachite green chloride reached to 84.2%, 79.6% and 99.9%, respectively. The greater agreement between the calculated and experimental results suggested that pseudo second-order kinetic model better represents the kinetic adsorption data. Equilibrium adsorptions of dyes were better explained by the Temkin isotherm. The results implied that this new cellulose-based absorbent had the universaiity for removal of dyes through the chemical adsorption mechanism.     

2016~2020年山东省空气质量时空分布特征及影响因素分析

基于2016～2020年山东省城市空气质量监测结果,结合人口密度和城镇化率等社会经济数据以及风速、气温和相对湿度等气象要素,综合应用地理加权回归模型（GWR）、多尺度地理加权回归模型（MGWR）和小波分析等方法,探究山东省空气污染物的时空分布特征及其与社会经济和气象要素的关系.结果表明：（1）近5年来山东省空气质量整体呈好转趋势,除O₃外,SO₂、 NO₂、 PM_2.5和PM₁₀等污染物浓度逐年降低.空气污染物分布存在明显的空间差异,沿海地区污染物浓度更低.（2）山东省PM_2.5与人口密度和第二产业占比具有极显著的正相关关系（P<0.01）,而与城镇化率呈极显著负相关关系（P<0.01）.同时,这种相关关系在空间上存在尺度差异,人口密度、民用汽车量和工业用电量与PM_2.5空间关系较平稳,而城镇化率和第二产业占比对PM_2.5的影响的空间异质性较高.（3）气象要素对菏泽和威海两市PM_2.5     

石羊河流域1959年至2008年风速变化特征分析

利用线性拟合、Mann—Kendall方法、小波分析对石羊河流域五个测站空间平均的近50年来的逐日风速资料进行了倾向、周期、突变等特性分析。研究结果表明：1959年至2008年间,整个石羊河流域风速呈减小趋势,变化率约为-0．004m／（s．a）．而上游风速呈现明显的上升趋势。四个季节中冬季风速降低速度最大,而夏季风速呈现上升趋势。风速的长期变化还具有一定的突变性,年平均风速在1988年出现了由高到低的突变,表明风速开始下降。复值Morlet小波分析的结果显示,风速变化存在6a、13a、19a、25a四个峰值。其中19a的时间尺度为第一主周期。     

黄河上中游天然径流多时间尺度变化及动因分析

运用小波分析等方法对黄河上中游天然径流序列进行了分析,揭示了河川径流多时间尺度变化规律,并对径流在不同时间尺度上变化的影响因素进行了探讨。研究结果对于认识黄河径流的变化规律、开发利用管理黄河水资源有着重要的意义。     

浅谈烟尘采样的两种分析验证方案

本文应用两种污染物传递数学模型分析烟尘采样过程,第一个模型假设烟尘经过采样后又经过一段物理运动重新达到平衡,因此得到采样后烟尘浓度和采样器内浓度的数学等式,第二种模型则是在第一种的模型基础之上给出烟尘浓度随,的变化而变化的数学表示式子.     

铝合金的实验室盐雾试验腐蚀行为图像特征提取

目的提取铝合金材料在盐雾箱试验中可以表征腐蚀程度的表面形貌图像特征量。方法首先,对采集于表面化学清洗过的试样原始图像进行图像增强等预处理,突出腐蚀部位;其次,基于数字图像处理的分形和小波分解方法提取出分形维数和小波能量特征值。结果与以质量损失量为基础的腐蚀深度特征值进行相关性对比,基于图像分析的特征提取法的准确度和精度比较高。结论该方法可以应用于对试样的腐蚀程度进行定性和定量分析,判断并预测试样的腐蚀速度。