考虑颜色特征最优组合的CART决策树火灾图像识别方法*

针对火灾图像识别过程中颜色特征数量多、特征间相关性复杂、难以在多维特征融合过程中有效融合图像颜色特征等问题,提出1种考虑颜色特征最优组合的CART决策树火灾图像识别方法。首先,在Lab、RGB、HSV 3种色彩模式下基于图像颜色特征提取火灾图像特征序列;其次,分别在3种色彩模式下基于精细决策树与特征随机排列组合方法提取颜色特征中最优组合特征;最后,将提取的火灾图像最优组合特征序列作为CART决策树输入进行模型训练,并通过测试样本以及其他机器学习方法进行模型泛化能力的分析。研究结果表明:本文方法寻找出识别火灾图像的最优颜色特征组合为“Kb1+Var1+Kg+Kb2+Var2+Kh+Ks+Kv”;CART决策树方法对于火灾图像识别的测试准确度可达84.5%,其分类效果明显优于其他决策树类与集成树类方法;9折为最佳交叉验证折数,其测试准确度可达86.47%,与5折交叉验证相比明显提升14.77%。研究结果可为火灾图像识别提供方法基础。     

Fish recruitment prediction, using robust supervised classification methods

Improving our ability to predict recruitment is a key element in fisheries management. However, the interactions between population dynamics and different environmental factors are complex and often non-linear, making it difficult to produce robust predictions. ‘Machine-learning’ techniques (in particular, supervised classification methods) have been proposed as useful tools, to overcome such difficulties. In this study, a methodology is proposed to build a robust classifier for fish recruitment prediction with sparse and noisy data. The methodology consists of 4 steps: (1) a semi-automated recruitment discretization method; (2) supervised discretization of predictors; (3) multivariate and non-redundant predictors selection; (4) learning a probabilistic classifier. In terms of fisheries management, the classifier estimated performance has important consequences and, to be useful, the manager needs to know the risk that is being taken when using this number. Probabilistic classifiers such as ‘naive Bayes’, have the advantage that, in addition to the predictions, estimate also the probability of each possible outcome. Anchovy (Engraulis encrasicolus) and hake (Merluccius merluccius) recruitments are used as application examples. ‘Two-intervals’ recruitment discretization accomplishes 70% accuracies and Brier scores of around 0.10, for both anchovy and hake recruitment. In comparison, ‘three-intervals’ recruitment discretization accomplishes 50% accuracies; and Brier scores of around 0.25 for anchovy and 0.30 for hake recruitment. These statistics are the result of validating not only the classifier, but also the previous steps, as a whole methodology.     

Automated detection of frog calls and choruses by pulse repetition rate

Anurans (frogs and toads) are among the most globally threatened taxonomic groups. Successful conservation of anurans will rely on improved data on the status and changes in local populations, particularly for rare and threatened species. Automated sensors, such as acoustic recorders, have the potential to provide such data by massively increasing the spatial and temporal scale of population sampling efforts. Analyzing such data sets will require robust and efficient tools that can automatically identify the presence of a species in audio recordings. Like bats and birds, many anuran species produce distinct vocalizations that can be captured by autonomous acoustic recorders and represent excellent candidates for automated recognition. However, in contrast to birds and bats, effective automated acoustic recognition tools for anurans are not yet widely available. An effective automated call-recognition method for anurans must be robust to the challenges of real-world field data and should not require extensive labeled data sets. We devised a vocalization identification tool that classifies anuran vocalizations in audio recordings based on their periodic structure: the repeat interval-based bioacoustic identification tool (RIBBIT). We applied RIBBIT to field recordings to study the boreal chorus frog (Pseudacris maculata) of temperate North American grasslands and the critically endangered variable harlequin frog (Atelopus varius) of tropical Central American rainforests. The tool accurately identified boreal chorus frogs, even when they vocalized in heavily overlapping choruses and identified variable harlequin frog vocalizations at a field site where it had been very rarely encountered in visual surveys. Using a few simple parameters, RIBBIT can detect any vocalization with a periodic structure, including those of many anurans, insects, birds, and mammals. We provide open-source implementations of RIBBIT in Python and R to support its use for other taxa and communities.     

基于熵权-集对分析模型的重大危险源安全等级评价研究

针对国内危险化学品事故频发,但危险源分级监管研究不足的现状,尝试将集对分析理论应用于重大危险源安全等级评价中。以评价指标集合与分级标准集合构成一组集对,利用熵权法计算出各项评价指标的客观权重,再结合同异反联系度函数求出各指标单项联系度及加权平均联系度,以此为依据判定重大危险源的安全等级。研究案例的实施情况表明,熵权-集对分析模型充分考虑了危险源安全因素中的不确定性,并且通过熵权值能够识别出影响系统安全的关键指标;与传统方法相比,该模型具有较高的可行性和易操作性,可以为重大危险源的安全监管和事故预防工作提供理论支持。     

基于IOCS的内蒙古潜在植被NPP空间分布特征研究

以综合顺序分类系统(IOCS)为基础,利用1985-2009年内蒙古及周边常规气象站的观测资料,采用NPP分类指数模型模拟内蒙古各潜在植被类型NPP,分析其空间分布特征及NPP与气候因子之间的关系。研究表明:①内蒙古地区植被NPP在空间分布上表现出一定的经向分布特征,多年均值介于0~366.5 gC·m^-2·a^-1之间,高值区中心分布在大兴安岭东北部;②各潜在植被类型中,碳汇潜力贡献率最大的潜在植被类型为微温微干温带典型草原类(ⅢC),其值为23.6%;最小的为寒温微干山地草原类(ⅡC),接近于0;③内蒙古潜在植被类型NPP表现出明显的干湿地带性和纬度地带性,且随寒温→微温→暖温、微干→微润→湿润→潮湿的变化其NPP值逐渐增大;但暖温极干暖温带荒漠类(ⅣA)和暖温干旱暖温带半荒漠类(ⅣB)并不符合这一规律,主要原因是较高的温度和较少的降水,达不到植被物质生长的需求。     

An Index of Risk as a Measure of Biodiversity Conservation Achieved through Land Reform

Abstract:  We measured the net progress of land reform in achieving a national policy goal for biodiversity conservation in the context of ongoing clearing of native vegetation and additions of land to a highly nonrepresentative (residual) reserve network, interior South Island, New Zealand. We used systematic conservation-planning approaches to develop a spatially explicit index of risk of biodiversity loss (RBL). The index incorporated information from national data sets that describe New Zealand's remaining indigenous land cover, legal protection, and land environments and modeled risk to biodiversity on the basis of stated assumptions about the effects of past habitat loss and legal protection. The index identified irreplaceable and vulnerable native habitats in lowland environments as the most at risk of biodiversity loss, and risk was correlated with the density of threatened plant records. To measure achievement, we used changes in the index that reflected gains made and opportunity costs incurred by legal protection and privatization. Application of the index to measure the difference made by land reform showed it had caused a net increase in the risk of biodiversity loss because most land vulnerable to habitat modification and rich in threatened plant species was privatized and land at least risk of biodiversity loss was protected. The application revealed that new high-elevation reserves did little to mitigate biodiversity decline, that privatization of low-elevation land further jeopardized the most vulnerable biodiversity in lowland native habitats, and that outcomes of land reform for biodiversity deteriorated over time. Further development of robust achievement measures is needed to encourage more accountable biodiversity conservation decisions.     

A mesoscale vortex in a small stratified lake

A mesoscale vortex structure in the small stratified Lake Stechlin has been revealed by field experiments with satellite-tracked quasi-lagrangian drifters. The vortex with a radius of about 200 m drifted at 300 m/day along the western bight of the lake with nearly constant rotation speed of 3 cpd. Analysis of kinematical properties of the vortex motion demonstrates solid body character of rotation. Extrapolation of the vortex drift trajectory over the period preceding the observations combined with data on local winds and seiche dynamics has allowed tracing the vortex fate from its generation point. The normal modes analysis of the internal seiching in the lake reveals the vortex generation mechanism to be the interaction of certain seiche modes with local bottom topography and suggests generation of the mesoscale vortices to be the a regular feature of the lake circulation. Analysis of vorticity suggests additional energy supply to rotational flow, possibly from inverse cascading of small-scale turbulent motions—a feature typical for quasi-2D turbulence. The vortices can play an important role in the energy transport from basin-scale motions to small-scale boundary mixing. They can also contribute significantly to the horizontal heterogeneity of phyto- and zooplankton distribution as well as to the transport of dissolved matter such as nutrients between littoral and profundal areas. The topographically generated traveling vortices represent an analog of the synoptic eddies in the Ocean and in the Atmosphere, whereas their role in the lake hydrodynamics is practically unknown.     

基于模糊聚类法研究绥宁黄桑自然保护区种子植物多样性与邻近地区之间的关系

通过对绥宁黄桑自然保护区植物多样性与其相邻的其余的9个地区植物多样性进行模糊聚类分析,结果表明,按与其区系关系的密切程度的大小进行排列,与绥宁县黄桑自然保护区的关系的密切从大至小的顺序为：台湾岛〉井冈山〉武夷山〉大罗山〉白云山〉北山〉北仑山〉黄山〉神农架,旨在对植物分类方法的研究作一个有益的尝试,探索一条植物分类的新方法,同时其结果对黄桑自然保护区的经营管理和保护具有一定的实践指导意义．图1,表2,参14．     

城市轨道交通危险源分类优化方案及应用

如何准确查找城市轨道交通运营过程中存在危险源,避免或减少运营事故的发生,是城市轨道交通运营企业风险管理工作的重要内容。通过对国内运营成熟的城市轨道交通运营企业及其他行业在风险源管理工作的对比分析,提出了城市轨道交通危险源登记手册及危险源分类的优化设计方案,并将其应用于国内某市城市轨道交通的车辆中心、综合机电中心、自动化中心等关键部门的危险源统计分析工作中,得到了较好的效果,为提升国内城市轨道交通运营企业的风险管理提供了有益的支撑。     

Q型类聚法在海域监测站位优化中的应用——以辽东湾为例

站位布设是海域环境监测工作的一个重要技术部分,是获取有代表性监测数据,为环境管理服务的重要环节。为了最高效地布设采样点,结合辽东湾的实例,利用Q型类聚法优化了研究区域的监测点位,由原来的29个水质采样点优化到22个代表站位,优化后的站位所表征的环境质量,与近岸海域环境质量等效,并比原常规监测站位更具有代表性。在提高采样效率的同时节约了监测成本。因此,研究近岸海域环境监测站位优化具有十分重要的现实意义。