高光谱图像基于像素结构的改进PCA算法

主成分分析法(PCA)作为一种常用的降维算法,被广泛的应用到如高光谱图像处理等需要进行大量数据处理的应用中。PCA的主要目的是利用正交变换,将具有相关性的高维数据的分量转换为线性不相关的新的成分变量,但当矩阵维数超过百万时候会造成严重的计算困难问题。本文针对PCA运算中协方差矩阵计算过程中内存调度的问题,提出了一种基于像素结构的改进的协方差矩阵计算方法,可以在确保与常规PCA具有相同性能的同时有效地降低计算所需的存储器规模。实验中分别采用传统PCA算法和改进算法对高光谱图像数据进行特征提取后利用支持向量机(SVM)进行分类,对比结果验证了改进算法的有效性和可靠性。

Improving autoencoder-based unsupervised damage detection in uncontrolled structural health monitoring under noisy conditions

Structural health monitoring is widely utilized in outdoor environments,especially under harsh conditions,which can introduce noise into the monitoring system.Therefore,designing an effective denoising strategy to enhance the performance of guided wave damage detection in noisy environments is crucial.This paper introduces a local temporal principal component analysis(PCA)reconstruction approach for denoising guided waves prior to implementing unsupervised damage detection,achieved through novel autoencoder-based reconstruction.Experimental results demonstrate that the proposed denoising method significantly enhances damage detection performance when guided waves are contaminated by noise,with SNR values ranging from 10 to-5 dB.Following the implementation of the proposed denoising approach,the AUC score can elevate from 0.65 to 0.96 when dealing with guided waves corrputed by noise at a level of-5 dB.Additionally,the paper provides guidance on selecting the appropriate number of components used in the denoising PCA reconstruction,aiding in the optimization of the damage detection in noisy conditions.

马尾松两广优良家系遗传变异研究

通过调查马尾松中龄林的生长情况,对其群体遗传变异进行分析。结果表明,马尾松两广优良家系中龄林存在丰富的遗传变异,树高、胸径、材积和冠幅等性状在家系层次上有极显著或显著差异,这些差异主要由遗传因素制约,各性状受中度、中低度遗传控制。主成分分析表明,树高、胸径和材积为主要生长性状,它们的累积贡献率达89.54%。采用Structure Version 2.2软件进行群体遗传结构分析,将群体分成5大类,在第4组群中,广东的家系数量最多,且组群生长性状值最高;第2组群中广西的家系数量最多,而生长性状值仅次于第4组群。

Application of PCA and HCA to the Structure-Activity Relationship Study of Fluoroquinolones

Density functional theory （DFT） was used to calculate molecular descriptors （properties） for 12 fluoro-quinolone with anti-S.pneumoniae activity. Principal component analysis （PCA） and hierarchical cluster analysis （HCA） were employed to reduce dimensionality and investigate in which variables should be more effective for classifying fluoroquinolones according to their degree of an-S.pneumoniae activity. The PCA results showed that the variables ELUMO, Q3, Q5, QA, logP, MR, VOL and △EHL of these compounds were responsible for the anti-S.pneumoniae activity. The HCA results were similar to those obtained with PCA.The methodologies of PCA and HCA provide a reliable rule for classifying new fluoroquinolones with antiS.pneumoniae activity. By using the chemometric results, 6 synthetic compounds were analyzed through the PCA and HCA and two of them are proposed as active molecules with anti-S.pneumoniae, which is consistent with the results of clinic experiments.

中华民族D3S1358、vWA基因座的空间地理分布

目的运用地理信息系统(GIS)的理论和方法分析中国87个人群常染色体D3S1358、vWA基因座的空间群体遗传结构,为研究中华民族的起源、人口迁移、考古以及民族融合等群体遗传学和人类学问题提供分子生物学依据。方法应用主成分分析(PCA)方法及GIS中的空间差值分析(SDA)技术,从基因频率矩阵中提取综合指标,并采取可视化技术来反映中国人群2个短串联重复序列(STR)位点的空间群体遗传结构。结果中国人群D3S1358、vWA基因座的主成分空间分析结果划分出三大区域,且显示出从东南向西北递增的趋势,在每个区域内又有不同划分小区域。结论中国人群2个STR位点的空间群体遗传结构呈现明显的地理遗传梯度变异性,不同群体之间存在明显差异。

Characteristics of soil microbial community functional and structure diversity with coverage of Solidago Canadensis L

The relationship between Solidago canadensis L. invasion and soil microbial community diversity including functional and structure diversities was studied across the invasive gradients varying from 0 to 40%, 80%, and 100% coverage of Solidago canadensis L. using sole carbon source utilization profiles analyses, principle component analysis （PCA） and phospholipid fatty acids （PLFA） profiles analyses. The results show the characteristics of soil microbial community functional and structure diversity in invaded soils strongly changed by Solidago canadensis L. invasion. Solidago canadensis L. invasion tended to result in higher substrate richness, and functional diversity. As compared to the native and ecotones, average utilization of specific substrate guilds of soil microbe was the highest in Solidago canadensis L. monoculture. Soil microbial functional diversity in Solidago canadensis L. monoculture was distinctly separated from the native area and the ecotones. Aerobic bacteria, fungi and actinomycetes population significantly increased but anaerobic bacteria decreased in the soil with Solidago canadensis L. monoculture. The ratio of cyl9：0 to 18：1 co7 gradually declined but mono/sat and fung/bact PLFAs increased when Solidago canadensis L. became more dominant. The microbial community composition clearly separated the native soil from the invaded soils by PCA analysis, especially 18： lco7c, 16： lco7t, 16： lco5c and 18：2co6, 9 were present in higher concentrations for exotic soil. In conclusion, Solidago canadensis L. invasion could create better soil conditions by improving soil microbial community structure and functional diversity, which in turn was more conducive to the growth ofSolidago canadensis L.

AGGREGATE VOLUMETRIC ESTIMATION BASED ON PCA AND MOMENTUM-ENHANCED BP NEURAL NETWORK

This paper proposes a Back Propagation (BP) neural network with momentum enhancement aiming to achieving the smooth convergence for aggregate volumetric estimation purpose. Network inputs are first selected by optically measuring the eight geometry-related parameters from the given particle image. To simplify the network structure, principal component analysis technique is applied to reduce the input dimension. The specific network structure is finalized based on both empirical expertise and analysis on selecting the appropriate number of neurons in hidden layer. The network is trained using the finite number of randomly-picked particles. The training and test results suggest that, compared to the generic BP network, the training duration of the proposed neural network is greatly attenuated, the complexity of the network structure is largely reduced, and the estimation precision is within 2%, being sufficiently up to technical satisfaction.