ON THE UTILITY OF CANONICAL CORRELATION ANALYSIS FOR DOMAIN ADAPTATION IN MULTI-VIEW HEADPOSE ESTIMATION

The utility of canonical correlation analysis (CCA) for domain adaptation (DA) in the context of multi-view head pose estimation is examined in this work. We consider the three problems studied in 1], where different DA approaches are explored to transfer head pose-related knowledge from an extensively labeled source dataset to a sparsely labeled target set, whose attributes are vastly different from the source. CCA is found to benefit DA for all the three problems, and the use of a covariance profile-based diagonality score (DS) also improves classification performance with respect to a nearest neighbor (NN) classifier.

PCR-DGGE Fingerprinting Analysis of Plankton Communities and Its Relationship to Lake Trophic Status

Plankton communities in eight lakes of different trophic status near Yangtze, China were characterized by using denatured gradient gel electrophoresis (DGGE). Various water quality parameters were also measured at each collection site. Following extraction of DNA from plankton communities, 16S rRNA and 18S rRNA genes were amplified with specific primers for prokaryotes and eukaryotes, respectively; DNA profiles were developed by DGGE. The plankton community of each lake had its own distinct DNA profile. The total number of bands identified at 34 sampling stations ranged from 37 to 111. Both prokaryotes and eukaryotes displayed complex fingerprints composed of a large number of bands: 16 to 59 bands were obtained with the prokaryotic primer set; 21 to 52 bands for the eukaryotic primer set. The DGGE-patterns were analyzed in relation to water quality parameters by canonical correspondence analysis (CCA). Temperature, pH, alkalinity, and the concentration of COD, TP and TN were strongly correlated with the DGGE patterns. The parameters that demonstrated a strong correlation to the DGGE fingerprints of the plankton community differed among lakes, suggesting that differences in the DGGE fingerprints were due mainly to lake trophic status. Results of the present study suggest that PCR-DGGE fingerprinting is an effective and precise method of identifying changes to plankton community composition, and therefore could be a useful ecological tool for monitoring the response of aquatic ecosystems to environmental perturbations.

DCA、CCA和DCCA三种排序方法在中国草地植被群落中的应用现状

排序是用来分析植被与环境之间生态关系的重要手段。该文主要是对除趋势对应分析(DCA)、典范对应分析(CCA)和除趋势典范对应分析(DCCA)这三种排序方法进行总结,讨论它们在中国草地植被群落研究中的应用现状,并得出除趋势对应分析与聚类分析结合使用效果比较好,主要是用来揭示群落之间的关系,并且在实际的应用中也比较多,而典范对应分析和除趋势典范对应分析在揭示种与环境关系方面具有明显的优势,但是由于某些条件的限制,在实际研究中应用的比较少。可见这几种排序方法在实际应用中具有一定的优势,所以应该加强这方面的应用,更加深入的研究草地植被群落,以期对草地生态学及草地植被群落生态关系的研究工作发挥重要的借鉴作用。

The use of ensemble empirical mode decomposition with canonical correlation analysis as a novel artifact removal technique

Biosignal measurement and processing is increasingly being deployed in ambulatory situations particularly in connected health applications. Such an environment dramatically increases the likelihood of artifacts which can occlude features of interest and reduce the quality of information available in the signal. If multichannel recordings are available for a given signal source, then there are currently a considerable range of methods which can suppress or in some cases remove the distorting effect of such artifacts. There are, however, considerably fewer techniques available if only a single-channel measurement is available and yet single-channel measurements are important where minimal instrumentation complexity is required. This paper describes a novel artifact removal technique for use in such a context. The technique known as ensemble empirical mode decomposition with canonical correlation analysis (EEMD-CCA) is capable of operating on single-channel measurements. The EEMD technique is first used to decompose the single-channel signal into a multidimensional signal. The CCA technique is then employed to isolate the artifact components from the underlying signal using second-order statistics. The new technique is tested against the currently available wavelet denoising and EEMD-ICA techniques using both electroencephalography and functional near-infrared spectroscopy data and is shown to produce significantly improved results. © 1964-2012 IEEE.

鄂尔多斯高原景观要素空间格局及植物与环境关系的定量解析

鄂尔多斯高原是一个多层次、复杂的生态过渡带，具有复杂多样的环境条件、生态特点，因此也就具有复杂多样的植物与环境关系。本文从群落和景观两个尺度水平上研究鄂尔多斯高原植物或植被与环境关系及景观空间格局。利用鄂尔多斯高原野外植物群落样方调坦数据、微生境环境数据、气候数据，以典范对应分析（Canonical Correspondence Analysis, CCA）的方法分析了鄂尔多斯高原植物分布空间格局与环境要素的关系并对秋类环境要素对鄂尔多斯高原植物空间分布格局的贡献进行了定量分解;利用1:500 000 鄂尔多斯高原植被、土壤、土地利用、土地沙漠化类型等专题地图在GIS支持下分析了鄂尔多斯高原景观空间格局、并利用上述专题图数据加上鄂尔多斯高原气候数据库分析了土壤、土地利用、土地利用、土地沙漠化和气候等对鄂尔多斯高原植被空间分布格局的作用。通过分析，得到了以下主要结论： 1 在分析方法上、利用典范对应分析的方法，把植物分布的空间因素与环境因素分离的方法发展为植物分布空间格局不同类型影响因素作用的定量分离，提出了相应的概念模型和实现方法。 2 影响鄂尔多斯高原植物分布空间格局的主要微生境环境要素是基质类型、地下水位、覆沙厚度等，而影响鄂尔多斯高原分布空间格局的主要气候要素中，降水和干湿指标的作用大于温度和热量指标的作用。 3 通过对鄂尔多斯高原植物分布空间格局与环境关系的研究，以植物对微生境环境要素的反应为根据，把鄂尔多斯高原主要植物划分为4个大类群：梁地植物、沙地植物、草甸植物和耐盐植物。根据它们对气候要素的反应，把鄂尔多斯高原主要植物划分为典型草原植物、荒漠草原植物和草原化荒漠植物3大类。进一步，根据鄂尔多劳动保护高原植物与环境关系的研究，进行了鄂尔多斯高原植被功能型划分的尝试，得到了鄂尔多斯高原的12种主要植被功能型。 4 对鄂尔多斯高原植物分布的空间格局的影响环境因素的贡献作了定量地分解。分析结果显示：鄂尔多斯高原植物分布空间格局中有27.02％可由已知环境变量得到解释，其中21.56％与微生境环境要素相关，7.51％与气候要素的作用有关，而气候与微生境环境要素的耦合作用的份额为2.05％。根据植物生长是否直接受到地下水的影响，鄂尔多斯高原存在两大类型生态特点差异明显的生境类型：中性立地和隐域生境，对两大类型生境上影响植物空间分布格局的环境要素的作用也进行了定量分解。分析结果表明：对于植物生长不直接受地下水影响的中性立地，已知环境要素的作用可以解释植物空间分布格局总信息的29.36％，稍大于对总体上鄂尔多斯高原植物分布空间格局的解释，其中9.23％与气候要素相关，22.08％与微生境环境要素相关，而两种类型环境要素的耦合作用则占1.95％。对于植物生长直接受到地下水影响的隐域生境，所有已知环境要素对植物分布空间格局的贡献率为72.28％，其中气候要素的作用为30.31％，微生境环境要素的作用为49.08％，两类环境要素的耦合作用为7.11％。 5 描述景观空间格局的指数多种多样，这些能数在描述特定区域的景观空间格局时是有信息冗余的。本文对利用FRAGSTATS所获得的鄂尔多斯高原植被、土壤、土地利用、土地沙漠化等景观分量的20个景观指数实施了因子分析。通过因子分析，我们可以把描述鄂尔多斯高原景观空间格局的景观指数归并为以下8类：多样性指数、斑块多度指数、斑块类型丰富度指数、斑块面积指数、斑块形状指数、分形维数、空间配置指数和斑块面积变异指数。通过因子分析，还得到了这些景观指数对描述鄂尔多斯景观格局的共性特征：在描述鄂尔多斯高原景观空间格局时，作用最大的是多样性指数、斑块多度指数、面积加权平均斑块形状指数和面积加权平均分形维数，其次是斑块类型丰富度指数、平均分维指数、平均形状指数和斑块面积指数，而空间配置指数（扩散与毗连指数）和斑块面积变异指数的作用则比较微弱。 6 对鄂尔多斯高原景观指数的因子分析是非常有效和成功的。因子分析对鄂尔多斯高原植被、土壤、土地利用、土地沙漠化景观指数的分析分别得到了5-6个主要因子，可以表达原有20个景观指数所表达信息的91.1-96.0％，即可以反应鄂尔多斯高原景观空间格局的大部分信息。本文所进行的因子分析对因子进行了方差最大化(Varimax)正交旋转的处理，因子分析得到的每一个主要因子都有一个或几个与之相关性非常高的景观指数与之对应，因此，就可以用与因子分析所得主要因子相关性最高的景观指数代替该主要因子来表达鄂尔多斯高原的景观空间格局。另外还因为有些景观指数之间具有极高的相关系数，所以对因子分析所得到的景观指数可以进一步精减，最后利用因子分析成功地把原有20个景观指数减少到了11个。最后被选来描述鄂尔多斯高原景观格局的景观指数有下列11个：MSIEI（修正的Simpson均匀度指数）、AWMPFD（面积加权平均斑块分形维数）、AWMSI（面积加权平均形状指数）、NP（斑块数目）、PR（斑块类型丰富度）、MSI（平均形状指数）、MPFD（平均斑块分形维数）、MPS（平均斑块面积）、PSCV（斑块面积变异系数）、DLFD（双对数分形维数）和IJI（扩散与毗连指数）。 7 鄂尔多斯高原植被、土壤、土地利用在景观组成结构上具有一个共同特点，就是各种类型的面积差异极大，少数类型占有极大比重，而其余面积则很小。产生这一情形的原因主要与人为活动的强烈影响有关，表现在地带性的植被与土壤面积所占的比重不高，沙地、沙生植被与风沙土则占有很大比重。 8 以地带性植被和滩地隐域性植被表示的鄂尔多斯高原的原生植被仅占高原面积的不足30％，而以地带性土壤和滩地隐域性土壤表示的原生性土壤占鄂尔多斯高原总面积的近40％，说明土壤退化不如植被退化严重，或滞后于植被退化。 9 鄂尔多斯高原各景观指数的空间变化曲线，植被与土壤很相近，具有非常相似的格局;土地利用景观格局空间变化特征与植被、土壤等明显不同;土地沙漠化的景观格局空间变化曲线介植被曲线、土壤曲线与土地利用曲线之间，说明土地沙漠化不仅是一个受人为活动影响的过程，而且与自然过程密切相关。 10 鄂尔多斯高原景观格局的空间梯度变化表现出了东西向和南北向的梯度，但总体上以东西向的变化比较明显。 11 通过鄂尔多斯高原土壤类型、土地利用、土地沙漠化等景观要素、气候、空间要素与鄂尔多斯高原植被空间分布格局的CCA分析，探讨了它们之间的相互关系。以对鄂尔多斯高原植被组成数据的总方差解释的百分率为标准，土壤对鄂尔多斯植被分布的空间格局的作用最大，其方差贡献率可达44.28％，其次是土地利用与鄂尔多斯高原植被的关系也很密切，土地利用对鄂尔多斯高原植被空间分布格局的方差贡献率为22.45％，空间因素对鄂尔多斯高原植被空间分布格局的贡献率为17.51％，土地沙漠化对鄂尔多斯高原植被空间分布格局的贡献为15.65％，排在第四位;气候因素对鄂尔多斯高原植被空间格局的贡献率为11.95％，居第五位。 12 在气候要素对鄂尔多斯高原植被空间分布格局的作用中，降水与干湿指标的作用大于温度与热量指标的作用。这一点与利用野外调查样方的群落数据植物与气候关系的分析是完全一致的。CCA分析还表明鄂尔多斯高原植被空间格局的东西向变化大于南北向分异。 13 在群落和景观水平上，鄂尔多斯高原植物空间分布或植被格局的影响因素的作用具有相似的格局，即气候因子的作用明显地小于地质、土壤、水文等微生境环境要素（群落水平）或土壤（景观水平）的作用，并在这两个尺度上气候要素对植物空间分布或植被格局的定量解释份额上也是非常相近的，都仅有10％左右。气候因子对鄂尔多斯高原植物空间分布格局的这种弱的解释能力，从侧面说明了人为活动等非自然因素对鄂尔多斯高原植物空间分布格局的强烈作用。 14 在鄂尔多斯高原生态系统管理上，应协调人与自然的关系;加强鄂尔多斯高原的生物多样性保育，对于本区生态和经济对非常重要的滩地，应协调好对其开发利用与保护的关系;在鄂尔多斯高原土地沙漠化防治方面，应把调整人地关系与自然生态背景与条件相结合，如使用“三圈”模式等生态系统管理模式等。

Spatial distribution of benthic algae in the Gangqu River, Shangrila, China

This study consisted of sampling benthic algae at 32 sites in the Gangqu River, an important upstream tributary of the Yangtze River. Our aims were to characterize the benthic algae communities and relationships with environmental variables. Among the 162 taxa observed, Achnanthes linearis and Achnanthes lanceolata var. elliptica were the dominant species (17.10% and 14.30% of the total relative abundance, respectively). Major gradients and principal patterns of variation within the environmental variables were detected by principal component analysis (PCA). Then non-metric multidimensional scaling (NMS) divided all the sites into three groups, which were validated by multi-response permutation procedures (MRPP). Canonical correspondence analysis (CCA) indicated that three environmental variables (TN, TDS, and TP) significantly affected the distribution of benthic algae. Weighted averaging regression and cross-calibration produced strong models for predicting TN and TDS concentration, which enabled selection of algae taxa as potentially sensitive indicators of certain TN and TDS levels: for TN, Achnanthes lanceolata, Achnanthes lanceolata var. elliptica, and Cymbella ventricosa var. semicircularis; for TDS, Cocconeis placentula, Cymbella alpina var. minuta, and Fragilaria virescens. The present study represents an early step in establishing baseline conditions. Further monitoring is suggested to gain a better understanding of this region.

Response of macrozoobenthos communities to ecological engineering remediation in a hypertrophic urban lake

The relationship of macrozoobenthos communities with some environmental variables, and their response to the ongoing restoration measures were studied in a small hypertrophic urban lake near the Yangzte River, China. Twenty taxa including 9 oligochaetes, 7 insects, 2 mollusks and two other animals were found during March 2005 to May 2006. The reappearance of some indigenous macrozoobenthos species showed that the ecological engineering remediation carried out was helpful for the recovery of the macrozoobenthos communities. Through canonical correspondence analysis (CCA), it was detected that temperature (T), conductivity (COND), Secchi depth/deep (SD/Deep), TN, total suspended solids (SS) and chemical oxygen demand (CODcr) were significant environmental factors that influenced the pattern of macrozoobenthos. Limnodrilus hoffmeisteri, Tanypus sp. and Alocinma longicornis could be used as potential bio-indicators in monitoring the development of ecological restoration.

Plankton community composition in the Three Gorges Reservoir Region revealed by PCR-DGGE and its relationships with environmental factors

To explore the relationships between community composition and the environment in a reservoir ecosystem, plankton communities from the Three Gorges Reservoir Region were studied by PCR-denaturing gradient gel electrophoresis fingerprinting. Bacterial and eukaryotic operational taxonomic units (OTUs), generated by DGGE analysis of the PCR-amplified 16S and 18S rRNA genes, were used as surrogates for the dominant "biodiversity units". OTU composition among the sites was heterogeneous; 46.7% of the total bacteria] OTUs (45) and 64.1% of the eukaryotic OTUs (39) were identified in less than half of the sampling sites. Unweighted pair group method with arithmetic averages (UPGMA) clustering of the OTUs suggested that the plankton communities in the Xiangxi Rive sites were not always significantly different from those from the Yangtze River sites, despite clear differences in their environmental characterizations. Canonical correspondence analysis (CCA) was applied to further investigate the relationships between OTU composition and the environmental factors. The first two CCA ordination axes suggested that the bacterial community composition was primarily correlated with the variables of NO3--N, dissolved oxygen (DO), and SiO32--Si, whereas, the eukaryotic community was mainly correlated with the concentrations of DO, PO43--P, and SiO32--Si.

Controlling factors of spring-summer phytoplankton succession in Lake Taihu (Meiliang Bay, China)

The spring-summer successions of phytoplankton and crustacean zooplankton were examined weekly in Meiliang Bay of the subtropical Lake Taihu in 2004 and 2005. During the study period, the ecosystem of Meiliang Bay was characterized by (i) clearly declined nitrogen compounds (nitrate, TN, and ammonium) and slowly increased phosphorus compounds (TP and SRP), (ii) increased total phytoplankton density and rapid replacement of chlorophyta (mainly Ulothrix) by cyanobacteria (mainly Microcystis), and (iii) rapid replacement of large-sized crustaceans (Daphnia and Moina) by small-sized ones (Bosmina, Limnoithona, and Ceriodaphnia). Results from the CCA and correlation analysis indicate that the spring-summer phytoplankton succession was primarily controlled by abiotic factors. Cyanobacteria were mainly promoted by increased temperature and decreased concentrations of nitrogen compounds. The pure contribution of crustacean was low for the variation of phytoplankton suggesting a weak top-down control by crustacean zooplankton in the subtropical Lake Taihu.

Driving forces shaping phytoplankton assemblages in two subtropical plateau lakes with contrasting trophic status

1. We studied driving forces shaping phytoplankton assemblages in two subtropical plateau lakes with contrasting trophic status, the oligotrophic deep Lake Fuxian and the eutrophic shallow Lake Xingyun. 2. Phytoplankton samples were taken monthly for a year and phytoplankton species were sorted into the main taxonomic groups and associations proposed by Reynolds. A canonical correspondence analysis (CCA) was used to test the occurrence of these classification schemes and to determine their discriminatory power. 3. The results suggest that the major driving forces in Lake Fuxian were physical variables, and particularly the underwater light climate, whereas, nutrients were the important driving force in Lake Xingyun. 4. Top-down control through zooplankton grazing in Lake Fuxian was hardly ever a significant determinant itself, because of the scarcity of zooplankton and their low grazing efficiency of predation while a dominance of inedible cyanobacteria throughout the year rendered top-down controls ineffective failing in Lake Xingyun. Hence phytoplankton communities in both lakes appear to be regulated primarily by bottom-up controls.

Effects of seasonal succession and water pollution on the protozoan community structure in an eutrophic lake

The purpose of the research is to study the seasonal succession of protozoa community and the effect of water quality on the protozoa community to characterize biochemical processes occurring at a eutrophic Lake Donghu, a large shallow lake in Wuhan City, China. Samples of protozoa communities were obtained monthly at three stations by PFU (polyurethane foam unit) method over a year. Synchronously, water samples also were taken from the stations for the water chemical quality analysis. Six major variables were examined in a principal component analysis (PCA), which indicate the fast changes of water quality in this station I and less within-year variation and a comparatively stable water quality in stations II and III. The community data were analyzed using multivariate techniques, and we show that clusters are rather mixed and poorly separated, suggesting that the community structure is changing gradually, giving a slight merging of clusters form the summer to the autumn and the autumn to the winter. Canonical correspondence analysis (CCA) was used to infer the relationship between water quality variables and phytoplankton community structure, which changed substantially over the survey period. From the analysis of cluster and CCA, coupled by community pollution value (CPV), it is concluded that the key factors driving the change in protozoa community composition in Lake Donghu was water qualities rather than seasons. (c) 2006 Elsevier Ltd. All rights reserved.

Structure of macrozoobenthos of the East Dongting nature reserve, with emphasis on relationships with environmental variables

Ecological survey of macrozoobenthos assemblages was carried out at 32 sites in the East Dongting Nature Reserve, located in the northern region of the East Dongting Lake in the middle basin of the Yangtze River, China. All total 51 taxa including 18 oligochaetes, 15 mollusks, 14 insects and four other animals were recorded. Mollusks composed the dominant group and accounted for more than 70% of the total abundance. Assemblages were composed mainly of scrapers (66.7%) and collector-gatherers (nearly 20%), and to a lesser extent collector-filterers (roughly 12%), predators (ca. 7%), and shredders (ca. 6%). Two-way indicator species analysis, detrended correspondence, and canonical correspondence analysis (CCA) were employed to identify the relationships between macrozoobenthos assemblages and environmental variables. Thirty-two sites were separated into four site groups based on composition and relative abundance of benthic macroinvertebrates. CCA detected that water depth, pH, conductivity, SiO2, total nitrogen, total phosphorus, alkalinity, hardness, and Ca2+, were significant environmental factors influencing the pattern of macozoobenthos. In this minimal subset, water depth, pH, alkalinity and hardness were the most influential variables.

Distribution of epilithic algae in the Xiangxi River system and their relationships with environmental factors

The spatial pattern of epilithic algae in the Xiangxi River system was studied in relation to several environmental factors by two-way indictor species analysis (TWINSPAN), detrended correspondence analysis (DCA), and canonical correspondence analysis (CCA). Eighty-nine taxa including diatoms, green algae, and blue-green algae were observed. Diatoms were dominant, and Cocconeis placentula, Cymbella minuta, Diatoma vulgare, and Gomphonema angustatum appeared in most of sampling sites. By TWINSPAN and DCA, thirty-one sites were divided into three groups based on composition and relative richness of benthic algae. CCA indicated that SiO2, pH, total phosphorus, Ca2+, velocity, elevation, and Cl- were significant environmental factors affecting the distribution of algae communities. In this minimal subset, SiO2 and pH were the most influential variables.

长白山地区曲尾藓属植物生态分化的排序研究

The relationships of eight moss species of Dicranum in 31 sites in main ecological systems in the Changbai Mountain with environmental factors were studied by canonical correspondence analysis (CCA). The results showed that altitude, soil sand percentage, water percentage, acidity and canopy density were important environmental factors influencing the distribution of the species of Dicranum . The relationships between Dicranum elongatum Schleich. ex Schwaegr ., D.groenlandicum Brid. and altitude,between D.japonicum Mitt., D.scoparium Hedw. and canopy density,between D.polysetum Sw., D. undulatum Schrad. ex Brid. and soil acidity and water percentage,were positively correlative. The niche overlaps among the eight species of Dicranum were calculated. The minimal spanning tree of the eight species on the two-dimensional scatter plot were also drawn based on their niche overlaps, which clearly revealed the ecological similarities of eight species.