洁净室空气洁净度评价的模糊模式识别方法

洁净室空气洁净程度的高低属于模糊概念,将各评价样本实测指标矩阵转变为相对隶属度矩阵,同时将各标准模式指标标准值矩阵转变为指标准相对隶属度矩阵,以全体样本对于各标准模式间的加权广义距离平方和最小建立目标函数,经构造拉格朗日函数,建立了洁净室空气洁净度评价的模糊模式识别模型。斗净室洁净程度评价是一个多目标评价问题,提出了确定指标权重的主、客观结合法,即先建立求解各指标权重的迭代模型并求出指标权重的初始值,然后根据评价的经验,采用重要性语气算子调整各指标的相对隶属度,对初始值进行调整,归一化后得到合理的指标权重值,经计算实例验证,本建立的模式识别方法物理意义明确,评价结果符合实际。     

Applying deep learning to right whale photo identification

Photo identification is an important tool for estimating abundance and monitoring population trends over time. However, manually matching photographs to known individuals is time-consuming. Motivated by recent developments in image recognition, we hosted a data science challenge on the crowdsourcing platform Kaggle to automate the identification of endangered North Atlantic right whales (Eubalaena glacialis). The winning solution automatically identified individual whales with 87% accuracy with a series of convolutional neural networks to identify the region of interest on an image, rotate, crop, and create standardized photographs of uniform size and orientation and then identify the correct individual whale from these passport-like photographs. Recent advances in deep learning coupled with this fully automated workflow have yielded impressive results and have the potential to revolutionize traditional methods for the collection of data on the abundance and distribution of wild populations. Presenting these results to a broad audience should further bridge the gap between the data science and conservation science communities.     

Behavioural evidence for the involvement of Dufour’s gland secretion in nestmate recognition in the social wasp Polistes dominulus (Hymenoptera: Vespidae)

The results of recent studies by our group have suggested that in social wasps of the genus Polistes the Dufour’s gland is involved in kin recognition. In fact, the same hydrocarbons occurring on the cuticle are found in the gland secretion, and in P. dominulus the composition of the glandular secretion is more similar in foundresses belonging to the same colonies than in heterocolonial foundresses. In this study, P. dominulus colonies were experimentally presented with previously treated dead conspecific females. These lures had been deprived of their epicuticular lipids and coated with epicuticular or Dufour’s gland secretion extracts from females hetero- or homocolonial with respect to the tested colonies. The behaviour of the colonies towards these lures indicates that, like the epicuticular lipids, the Dufour’s gland secretion is involved in nestmate recognition. Received: 14 August 1995/Accepted after revision: 5 November 1995     

Analysis of floc morphology in a continuous-flow flocculation and sedimentation reactor

The floc morphology was investigated in a continuous-flow reactor, in order to understand the evolution of flocs in practical flocculation and sedimentation processes in water utilities. Kaolin-humic acid suspension was used as the test water, and polyaluminum chloride was chosen as the coagulant. An in-situ recognition system was applied to analyze the floc size, boundary fractal dimension, and eccentricity ratios. Particle numbers and turbidity were also determined in the sedimentation stage. At a coagulant dose of 1 mg/L as Al, the average floc size increased from 62 to 78 μm and the boundary fractal dimension was around 1.14, suggesting that flocs were compact and continuously grew during the entire flocculation process. However, with the dose increased to 5 mg/L, the average floc size decreased and stabilized at around 65 μm, with the fractal dimension of 1.20. It can be concluded that the excess coagulant doses resulted in the formation of chain-shaped, lower density, and more branched structure flocs, thereby restricting flocs' further growth in the subsequent flocculation. Floc morphology analysis suggested that charge neutralization dominated in the initial flocculation stage, then the bridge and sweep mechanisms were dominant in the subsequent flocculation. In addition, compared with the traditional inclined plate settler, a novel V-shaped plate settler introduced in this study had an advantage in small size floc(less than 5 μm) removal. The V-shaped region could promote aggregate restructuring and re-flocculation; therefore, the V-shaped plate settler provides an alternative method for sedimentation.     

基于脑电信号的驾驶员情绪状态识别研究

为研究驾驶员情绪状态识别技术,利用特定题材的影音氛围诱发兴奋和悲伤2种情绪,结合汽车模拟驾驶试验采集驾驶员情绪化驾驶时的脑电信号,利用db5小波分解算法分解信号,提取受情绪状态显著的脑电信号成分,并计算相应的功率谱。结果表明:驾驶员左右额叶区脑电β波受情绪状态影响显著;平静状态时,驾驶员左右额叶脑电β波功率值Pβ大小近似,数值较小;兴奋状态时,左右额叶脑电Pβ大小近似,相比于平静时显著增加;悲伤状态时,左额叶脑电Pβ显著增加,右额叶脑电Pβ相比于平静时无显著变化。Pβ可作为区分驾驶员平静(平常)、兴奋和悲伤这3种情绪状态的依据。     

基于可变模糊组合方法的瓦斯涌出量预测

矿井工作面瓦斯涌出是一个动态不确定过程。在分析了影响瓦斯涌出因素基础上,将可变模糊聚类与可变模糊模式识别两种模型相结合,提出了一种瓦斯涌出量预测的可变模糊组合方法。该方法首先利用粗糙集理论确定权重初值及各因素的重要性,然后利用可变模糊聚类模型求解最优模糊分类中心矩阵和最优权重,最后利用模式识别模型对待预测样本进行预测。以某矿井瓦斯涌出量预测为例,给出了具体预测过程。结果表明：可变模糊组合方法可行、有效。     

生物多样性维护功能评估及其空间尺度效应分析——以陕西省为例

生物多样性维护功能评估是生态系统综合评估的重要内容。基于物种丰富度的海拔梯度格局规律来修正生物多样性维护服务能力指数模型,并探讨该模型在省级、市级和生物多样性优先区三种不同尺度水平的评估响应。结果显示：（1）三种尺度识别的生物多样性极重要区面积和空间分布差异较大,省级尺度评估的极重要区集中于秦巴山地,市级尺度评估时极重要区分布广泛且布局分散,优先区尺度评估时极重要区与自然保护区重叠面积最大。（2）三种尺度评估识别的精准度由大到小依次为：优先区尺度（19.5%）>省级尺度（11.7%）>市级尺度（11.2%）,优先区尺度评估识别的精准度明显高于行政区尺度的评估。（3）优先区尺度评估识别的极重要区面积为28032.3 km²,占陕西省国土面积的13.7%;极重要区内土地利用结构和物种重要性综合评估结果优于行政区尺度的评估结果。     

基于Multi-class SVM的车辆换道行为识别模型研究

自动安全换道是车辆实现无人驾驶的关键,为精确识别行驶车辆换道状态,保证行车安全,设计了一种基于多分类支持向量机(Multi-class Support Vector Machine,Multiclass SVM)的车辆换道识别模型。从NGSIM数据集中选取美国101公路车辆轨迹数据进行分类处理,并将车辆换道过程划分为车辆跟驰阶段、车辆换道准备阶段和车辆换道执行阶段。采用网格搜索结合粒子群优化算法(Grid Search-PSO)对SVM模型中惩罚参数C和核参数g进行寻优标定,利用多分类支持向量机换道识别模型对样本数据进行训练和测试,模型测试精度达97.68%。研究表明,模型能够很好地识别车辆在换道过程中的行为状态,为车辆换道阶段的研究提供支持。     

论金属矿山地质灾害模式识别模型

从崩塌与滑坡、泥石流、采空塌陷3个方面探讨了金属矿山地质灾害模式,运用层次分析法确定了金属矿山地质灾害模式识别指标的权重。考虑到影响金属矿山地质灾害因素的不确定性,引入SPA,并与最优化原理相结合,提出了同异反最优模式识别模型,将其与现有的最大联系测度方法应用于某金属矿山地质灾害模式识别中。实例应用表明,同异反最优模式识别模型从整体上判断金属矿山地质灾害模式,较最大联系测度方法更为科学、合理。     

硫化矿石自燃倾向性评价的属性区间识别模型

为合理判定硫化矿石的自燃倾向性大小,将属性区间识别理论与信息熵理论结合,建立硫化矿石自燃倾向性综合评价的熵权属性区间识别模型。选取矿样的吸氧速度常数、自热点、着火点3项指标作为属性区间识别模型的基本判别因子,依据信息熵理论获得各判别指标的权重。给出矿样自燃倾向性的属性空间矩阵,采用均化系数将矿样的属性测度区间转化为综合属性测度;利用置信度识别准则和分级标准判别各个矿样的自燃倾向性。对采自国内典型硫化矿山的14个代表性矿样的室内实测值进行综合判定;结果表明,基于熵权的属性区间识别模型能更好地评价硫化矿石的自燃倾向性大小,如实反映出矿样的自燃特性。