用基于人工神经网络的数据融合法测量水流泥沙含量

在采用电容传感器测量泥沙含量的过程中 ,电容传感器的输出值受环境温度的影响较大 ,为消除温度对测量数据的影响 ,提出了采用人工神经网络法对传感器进行数据融合处理的方法 ,该方法以传感器的泥沙含量值与温度值作为网络的输入 ,通过对网络的训练达到消除非目标参量——温度的影响。试验结果表明该方法收敛速度快 ,输出稳定性可显著提高 。

数据融合法结合决策树优化东方草莓全草的地理溯源性研究

目的:通过融合HPLC与ICP-MS的分析结果建立数据集,结合模式识别法对来自5个产地96批次的藏药东方草莓进行地理溯源性研究,以期为未知来源的东方草莓提供便捷有效的产地判别方法。方法:采用HPLC建立东方草莓指纹图谱,并以LC-MS对各色谱峰进行成分归属,色谱峰信息作为数据集1;采用ICP-MS对东方草莓中21种无机元素进行含量测定,并以对数法建立无机元素指纹图谱,无机元素信息作为数据集2;二者数据融合作为数据集3。结合CA、PCA、PCA-LDA及C5.0决策树算法,对比3个数据集的地理追溯结果。结果:数据融合法相较于单一技术所得数据集,CA、PCA和PCA-LDA均可使96批东方草莓成功归类。对数据融合法的PCA-LDA模型进行内部验证和外部验证,其正确分类率均大于88.3%,表明该模型可作为东方草莓的正确分类依据。C5.0决策树筛选出4个主要贡献变量,并获得树深度为3的分类规则,其十折交叉验证结果显示平均准确度为98.9%。结论:数据融合法可提高不同产地东方草莓地理溯源的准确性,且决策树算法得出的分类规则可降低实际操作难度。以数据融合法结合决策树规则所得出的东方草莓地理分类机制为未知样品的产地来源提供新的鉴别依据,且操作简便,准确率高,有利于东方草莓的质量控制,也为其他药材品种溯源提供参考。

基于PHM的现代有轨电车车辆状态维修

针对现代有轨电车维修策略存在的不足,该文论述了一种基于PHM技术的现代有轨电车车辆状态维修策略,提出了基于车地无线通信的车辆安全检测设施与行车安全辅助设施的整合方案,通过数据融合法和层次分析法等数据处理方法,解决车辆状态监测数据相对孤立的问题,为运维数据的融合和关联分析提供参考,为现代有轨电车车辆维修策略的制定提供技术路线。

数字水印泛化盲检测模型的仿真研究

为降低水印检测的虚警率和漏警率,提出一个泛化盲检测模型。通过在Matlab环境下的实验,遴选7个数字水印检测算法,采用顺序检测法对待检测图像进行检测,然后利用数据融合法综合分析检测结果,判断其中的水印信息。相对单个算法来说,泛化检测模式的检测效率更高,检测耗费时间大为缩短,对仿真实验所采用的17个数字水印嵌入系统进行检测,达到虚警率为5.495%,漏警率为1.648%,正确检测率为92.857%的结果。经过检测分析,泛化检测模型还对适度的剪切、拉伸、反色、旋转以及几何失真等攻击具有更高的抵抗能力。因此,该泛化检测模型具有较高的可行性。

Multisensor Fuzzy Stochastic Fusion Based on Genetic Algorithms

To establish a parallel fusion approach of processing high dimensional information, the model and criterion of multisensor fuzzy stochastic data fusion were presented. In order to design genetic algorithm fusion, the fusion parameter coding, initial population and fitness function establishing, and fuzzy logic controller designing for genetic operations and probability choosing were completed. The discussion on the highly dimensional fusion was given. For a moving target with the division of 1 64 (velocity) and 1 75 (acceleration), the precision of fusion is 0 94 and 0 98 respectively. The fusion approach can improve the reliability and decision precision effectively.

Multi-sources information fusion algorithm in airborne detection systems

To aim at the multimode character of the data from the airplane detecting system, the paper combines Dempster- Shafer evidence theory and subjective Bayesian algorithm and makes to propose a mixed structure multimode data fusion algorithm. The algorithm adopts a prorated algorithm relate to the incertitude evaluation to convert the probability evaluation into the precognition probability in an identity frame, and ensures the adaptability of different data from different source to the mixed system. To guarantee real time fusion, a combination of time domain fusion and space domain fusion is established, this not only assure the fusion of data chain in different time of the same sensor, but also the data fusion from different sensors distributed in different platforms and the data fusion among different modes. The feasibility and practicability are approved through computer simulation.

A Digital Evidence Fusion Method in Network Forensics Systems with Dempster-Shafer Theory

Network intrusion forensics is an important extension to present security infrastructure,and is becoming the focus of forensics research field.However,comparison with sophisticated multi-stage attacks and volume of sensor data,current practices in network forensic analysis are to manually examine,an error prone,labor-intensive and time consuming process.To solve these problems,in this paper we propose a digital evidence fusion method for network forensics with Dempster-Shafer theory that can detect efficiently computer crime in networked environments,and fuse digital evidence from different sources such as hosts and sub-networks automatically.In the end,we evaluate the method on well-known KDD Cup1999 dataset.The results prove our method is very effective for real-time network forensics,and can provide comprehensible messages for a forensic investigators.

A novel fuzzy sensor fusion algorithm

A novel fusion algorithm was given based on fuzzy similarity and fuzzy integral theory. First, it calculated the fuzzy similarity among a certain sensor＇s measurement values and the multiple sensors＇ objective prediction values to determine the importance weight of each sensor and realize multi-sensor data fusion. Then according to the determined importance weight, an intelligent fusion system based on fuzzy integral theory was given, which can solve FEI-DEO and DEI-DEO fusion problems and realize the decision fusion. Simulation results were proved that fuzzy integral algorithm has enhanced the capability of handling the uncertain information and improved the intelligence degrees

Hierarchical Scene Analysis Method for Audio Sensor Networks

Abstract： A hierarchical method for scene analysis in audio sensor networks is proposed. This meth-od consists of two stages： element detection stage and audio scene analysis stage. In the former stage, the basic audio elements are modeled by the HMM models and trained by enough samples off-line, and we adaptively add or remove basic ele- ment from the targeted element pool according to the time, place and other environment parameters. In the latter stage, a data fusion algorithm is used to combine the sensory information of the same ar-ea, and then, a role-based method is employed to analyze the audio scene based on the fused data. We conduct some experiments to evaluate the per-formance of the proposed method that about 70% audio scenes can be detected correctly by this method. The experiment evaluations demonstrate that our method can achieve satisfactory results.

Microseismic Monitoring Data Fusion Algorithm and Coal and Gas Outbursts Prediction

The prediction study on coal and gas outbursts is carried out by monitoring some indices which are sensitive to the initiation of coal and gas outbursts. The values and changing roles of the indices are the foundations of coal and gas outbursts prediction. But now, only the data of ere key monitoring station is used in the coal and gas outbursts prediction practice, and the other data are ignored. In order to overcome the human factor and make full use of the monitoring information, the technique of multi-sensor target tracking is proposed to deal with the microseismic informatiion. With the results of microseismic events, the activities of geological structure, fracure-depth of roof and floor, and the location of gas channel are obtained. These studies indicate that it is considerably possible to predict the coal and gas outbursts using microseismic monitoring with its inherent ability to remotely monitor the progressive failure caused by mining.

A weighted selection combining scheme for cooperative spectrum prediction in cognitive radio networks

A weighted selection combining （WSC） scheme is proposed to improve prediction accuracy for cooperative spectrum prediction in cognitive radio networks by exploiting spatial diversity. First, a genetic algorithm-based neural network （GANN） is designed to perform spectrum prediction in consideration of both the characteristics of the primary users （PU） and the effect of fading. Then, a fusion selection method based on the iterative self-organizing data analysis （ISODATA） algorithm is designed to select the best local predictors for combination. Additionally, a reliability-based weighted combination rule is proposed to make an accurate decision based on local prediction results considering the diversity of the predictors. Finally, a Gaussian approximation approach is employed to study the performance of the proposed WSC scheme, and the expressions of the global prediction precision and throughput enhancement are derived. Simulation results reveal that the proposed WSC scheme outperforms the other cooperative spectrum prediction schemes in terms of prediction accuracy, and can achieve significant throughput gain for cognitive radio networks.

地理科学中数据空间重构最新研究进展

地理数据是反映地理特征—空间—时间三位一体的综合信息数据,随着计算机科学、地理信息系统(GIS)和空间建模等技术的发展,地理数据进行空间重构表达已经成为地理学发展一个方向。目前,基于地理数据重构研究成为地理学研究中的热点问题,重构研究主要分为数据源改进和重构方法(算法)探索。文章从地理数据重构的应用对象、空间尺度、重构方法、精度检验方法,以及存在的问题和展望等几个方面进行论述,阐明了各种方法的优缺点,并讨论该领域的未来发展方向。研究发现,地理数据重构文章中自然地理与人文地理数量比为6.3∶3.7,并且这些文献中有45.5%的研究尺度(空间分辨率)为1 km。地理数据重构方法中多源数据融合法和空间插值法较适合人文地理数据重构,波谱分析法和空间插值法较适合自然地理数据重构,"3S"数据空间分析法可适用大部分研究类型。数据重构的精度验证方法大体上有三种,但87.1%的研究中采用的是多源数据空间对比分析法。现阶段地理数据空间重构研究方向存在方法创新性不足、可适宜空间重构数据类型少和部分重构方法精度较低等问题,以数据融合空间化为代表的普适性强的方法为主要方向,未来该领域发展方向主要数据重构将会从进一步提高辅助空间数据产品的质量,以及挖掘数据融合算法这两个角度来提高数据重构的精度。