用于监测复杂系统的计算机辅助定性趋势分析

为了对传感器数据的瞬态演化用计算机进行建模,提出了定性趋势分析（Qualitative Trend Analysis）,这种定性的过程趋势描述是一种很有用的方法,已被广泛地用于过程状态监测、数据压缩、故障检测和诊断。首先概述国外QTA的产生背景和发展情况;其次提出有效的QTA趋势提取策略和趋势识别的若干重要问题。结果表明,该方法能够有效和快速地从数据中获取过程趋势。     

On Endmember Identification in Hyperspectral Images Without Pure Pixels: A Comparison of Algorithms

Hyperspectral imaging is an active area of research in Earth and planetary observation. One of the most important techniques for analyzing hyperspectral images is spectral unmixing, in which mixed pixels (resulting from insufficient spatial resolution of the imaging sensor) are decomposed into a collection of spectrally pure constituent spectra, called endmembers weighted by their correspondent fractions, or abundances. Over the last years, several algorithms have been developed for automatic endmember extraction. Many of them assume that the images contain at least one pure spectral signature for each distinct material. However, this assumption is usually not valid due to spatial resolution, mixing phenomena, and other considerations. A?recent trend in the hyperspectral imaging community is to design endmember identification algorithms which do not assume the presence of pure pixels. Despite the proliferation of this kind of algorithms, many of which are based on minimum enclosing simplex concepts, a rigorous quantitative and comparative assessment is not yet available. In this paper, we provide a comparative analysis of endmember extraction algorithms without the pure pixel assumption. In our experiments we use synthetic hyperspectral data sets (constructed using fractals) and real hyperspectral scenes collected by NASA’s Jet Propulsion Laboratory.     

基于定性趋势分析的油料灌装稳态过程管道堵塞故障诊断

针对油料灌装稳态过程中的管道堵塞故障,分析了管道堵塞对灌装稳态过程监控参数的影响,提出了基于定性趋势分析的油料灌装稳态过程故障诊断流程,给出了趋势提取、趋势识别及趋势匹配方法。以油料灌装控制实验装置为故障诊断对象,通过模拟不同类型的油料灌装稳态过程故障,采集管道输送压力、流量等过程监控参数时间序列数据,构建故障库,进行故障诊断案例验证,验证结果表明本文方法的有效性和正确性。     

对传感器网络定位技术现状的研究

随着学术界对无线传感器网络研究的不断深入,传感器网络节点的定位成为当前的热点。其中主要分为两种类型算法,分别为range-free和range-based。这两种类型的定位算法各有优势和不足之处。为了研究算法发展趋势,文中通过比较几种典型的定位算法,指出了各自的优缺点。最后指出了定位算法的发展趋势:range-based的定位算法比range-free的算法更适合传感器网络小型化、低成本化的需求,应该得到更大的发展。     

加速的QTA地形可视化算法

四叉树图集光线跟踪算法（QTA）是一种高效的平面地形可视化算法,在地形渲染相关的许多应用中（如矢量数据、谷地检测、等高线绘制等）都可以增加系统渲染速率。在QTA基础上,通过简化其中CRC算法的分支判断,引入新的屏幕误差控制策略,提出了一种新的光线投射地形可视化算法,同时对QTA中光栅化部分也进行了优化。该算法易于实现,且突破了原QTA算法只适用于平面地形的限制,算法实现了球形地形的可视化。实验结果验证了这一点,同时实验还显示,算法在不损失精度的情况下,具有更高的渲染速率。     

无线传感器网络定位技术探析

传感器网络中的节点定位问题因为与实际的很多应用直接相关而尤为受到关注,其中主要分为两种类型算法,分别为rang-free和rang-based,这两种类型的定位算法各有优势和不足之处。为了研究算法发展趋势,文章通过比较几种典型的定位算法,指出了各自的优缺点。     

Predicting aerodynamic instabilities in a blast furnace

This paper discusses the analysis of differential pressure signals in a blast furnace stack by using principal component analysis (PCA) and qualitative trend analysis (QTA) based on episodes. These methods can work jointly or separately and are applied using two toolboxes developed within the European CHEM project.¹ The objective in this paper is to predict aerodynamic instability in a blast furnace with sufficient warning to enable the blast volume to be reduced in order to minimise that instability. Both methods based on signals and the expert knowledge provide an efficient approach to slip prediction.     

Deep learning analysis of mobile physiological,environmental and location sensor data for emotion detection

The detection and monitoring of emotions are important in various applications, e.g., to enable naturalistic and personalised human-robot interaction. Emotion detection often require modelling of various data inputs from multiple modalities, including physiological signals (e.g., EEG and GSR), environmental data (e.g., audio and weather), videos (e.g., for capturing facial expressions and gestures) and more recently motion and location data. Many traditional machine learning algorithms have been utilised to capture the diversity of multimodal data at the sensors and features levels for human emotion classification. While the feature engineering processes often embedded in these algorithms are beneficial for emotion modelling, they inherit some critical limitations which may hinder the development of reliable and accurate models. In this work, we adopt a deep learning approach for emotion classification through an iterative process by adding and removing large number of sensor signals from different modalities. Our dataset was collected in a real-world study from smart-phones and wearable devices. It merges local interaction of three sensor modalities: on-body, environmental and location into global model that represents signal dynamics along with the temporal relationships of each modality. Our approach employs a series of learning algorithms including a hybrid approach using Convolutional Neural Network and Long Short-term Memory Recurrent Neural Network (CNN-LSTM) on the raw sensor data, eliminating the needs for manual feature extraction and engineering. The results show that the adoption of deep-learning approaches is effective in human emotion classification when large number of sensors input is utilised (average accuracy 95% and F-Measure=%95) and the hybrid models outperform traditional fully connected deep neural network (average accuracy 73% and F-Measure=73%). Furthermore, the hybrid models outperform previously developed Ensemble algorithms that utilise feature engineering to train the model average accuracy 83% and F-Measure=82%)     

一种基于各向异性高斯核核惩罚的PCA特征提取算法

提出了一种基于各向异性高斯核核惩罚的主成分分析的特征提取算法.该算法不同于传统的核主成分分析算法.在非线性数据降维中,传统的核主成分分析算法忽略了原始数据的无量纲化.此外,传统的核函数在各维度上主要由一个相同的核宽参数控制,该方法无法准确反映各维度不同特征的重要性,从而导致降维过程中准确率低下.为了解决上述问题,首先针对现原始数据的无量纲化问题,提出了一种均值化算法,使得原始数据的总方差贡献率有明显的提高.其次,引入了各向异性高斯核函数,该核函数每个维度拥有不同的核宽参数,各核宽参数能够准确地反映所在维度数据特征的重要性.再次,基于各向异性高斯核函数建立了核主成分分析的特征惩罚目标函数,以便用较少的特征表示原始数据,并反映每个主成分信息的重要性.最后,为了寻求最佳特征,引入梯度下降算法来更新特征惩罚目标函数中的核宽度和控制特征提取算法的迭代过程.为了验证所提出算法的有效性,各算法在UCI公开数据集上和KDDCUP99数据集上进行了比较.实验结果表明,所提基于各向异性高斯核核惩罚的主成分分析的特征提取算法比传统的主成分分析算法在9种公开的UCI公开数据集上准确率平均提高了4.49%.在KDDCUP99数据集上,所提基于各向异性高斯核核惩罚的主成分分析的特征提取算法比传统的主成分分析算法准确率提高了8%.     

A Novel Algorithm for Mining Association Rules in Wireless Ad Hoc Sensor Networks

In this paper, we propose a comprehensive framework for mining Wireless Ad-hoc Sensor Networks (WASNs) that is able to extract patterns regarding the sensors' behaviors. The main goal of determining behavioral patterns is to use these to generate rules that will improve the WASN's Quality of Service by participating in the resource management process or compensating for the undesired side effects of wireless communication. The proposed framework consists of a formal definition of sensor behavioral patterns and sensor association rules, a novel representation structure named the Positional Lexicographic Tree (PLT) that is able to compress the data gathered for the mining process and thus allows the fast and efficient mining of sensor behavioral patterns, as well as a distributed data extraction mechanism to prepare the data required for mining sensor behavioral patterns. To report on the performance of the mining framework, several experiments have been conducted to evaluate the PLT structure and the proposed data extraction mechanism.     

Adaptive tuning of a Kalman filter via fuzzy logic for an intelligent AUV navigation system

This paper describes the implementation of an intelligent navigation system, based on the integrated use of the global positioning system (GPS) and several inertial navigation system (INS) sensors, for autonomous underwater vehicle (AUV) applications. A simple Kalman filter (SKF) and an extended Kalman filter (EKF) are proposed to be used subsequently to fuse the data from the INS sensors and to integrate them with the GPS data. The paper highlights the use of fuzzy logic techniques to the adaptation of the initial statistical assumption of both the SKF and EKF caused by possible changes in sensor noise characteristics. This adaptive mechanism is considered to be necessary as the SKF and EKF can only maintain their stability and performance when the algorithms contain the true sensor noise characteristics. In addition, fault detection and signal recovery algorithms during the fusion process to enhance the reliability of the navigation systems are also discussed herein. The proposed algorithms are implemented to real experimental data obtained from a series of AUV trials conducted by running the low-cost Hammerhead AUV, developed by the University of Plymouth and Cranfield University.     

基于改进的独立分量分析的人脸识别方法

将独立分量分析（Independent Component Analysis，ICA）作为人脸特征提取方法。ICA所提取的特征分类能力强、相互独立，对像素间高阶统计特性敏感，并且不易受光照变化的影响。实验结果表明，基于IcA的人脸特征提取方法的识别性能优于特征脸法。针对传统的ICA算法（Informax算法）存在迭代次数多，难收敛，并且需要人工设定步长来调整学习速度的不足，本文采用FastICA作为ICA的快速算法，并将其关键迭代步骤加以改进，减少了耗时的雅可比矩阵求逆的运算次数。所提出的改进的FastICA具有无需人工参与，收敛速度快，迭代次数少的优点。在特征选择方面，本文将遗传算法（Genetie Algorithm，GA）应用到独立分量的选择与优化中，从而在保证较高识别性能的前提下，获得最优的人脸特征子集。     

An automated method for navigation assessment for Earth survey sensors using island targets

An automated method has been developed for performing navigation assessment on satellite-based Earth sensor data. The method utilizes islands as targets which can be readily located in the sensor data and identified with reference locations. The essential elements are an algorithm for classifying the sensor data according to source, a reference catalogue of island locations, and a robust pattern-matching algorithm for island identification. The algorithms were developed and tested for the Sea-viewing Wide Field-of-view Sensor (SeaWiFS), an ocean colour sensor. This method will allow navigation error statistics to be automatically generated for large numbers of points, supporting analysis over large spatial and temporal ranges.     

CGHA for principal component extraction in the complex domain

Principal component extraction is an efficient statistical tool which is applied to data compression, feature extraction, signal processing, etc. Representative algorithms in the literature can only handle real data. However, in many scenarios such as sensor array signal processing, complex data are encountered. In this paper, the complex domain generalized Hebbian algorithm (CGHA) is presented for complex principal component extraction. It extends the real domain generalized Hebbian algorithm (GHA) proposed by Sanger (1992). Convergence of CGHA is analyzed. Like GHA, CGHA can be implemented by a single-layer linear neural network with simple computation. An example is given where CGHA is utilized in direction-of-arrival estimation of multiple narrowband plane waves received by a sensor array.     

Feature extraction and dimensionality reduction algorithms and their applications in vowel recognition

Feature extraction is an important component of a pattern recognition system. It performs two tasks: transforming input parameter vector into a feature vector and/or reducing its dimensionality. A well-defined feature extraction algorithm makes the classification process more effective and efficient. Two popular methods for feature extraction are linear discriminant analysis (LDA) and principal component analysis (PCA). In this paper, the minimum classification error (MCE) training algorithm (which was originally proposed for optimizing classifiers) is investigated for feature extraction. A generalized MCE (GMCE) training algorithm is proposed to mend the shortcomings of the MCE training algorithm. LDA, PCA, and MCE and GMCE algorithms extract features through linear transformation. Support vector machine (SVM) is a recently developed pattern classification algorithm, which uses non-linear kernel functions to achieve non-linear decision boundaries in the parametric space. In this paper, SVM is also investigated and compared to linear feature extraction algorithms.     

Evaluation of a grey scale tactile array sensor pad for robotic applications

Since sensory feedback is an important part of robot control and the acquisition, manipulation, and recognition of objects, incorporating a sense of touch into a robotic system can greatly enhance the performance of that system. This article describes the evaluation of a recently developed low-resolution tactile array sensor pad system for use in robotic applications. Computer algorithms are developed which acquire data from the sensor pad and display the data on a CRT screen. Vision algorithms are implemented in order to extract the necessary information from the tactile data which will aid in the acquisition, manipulation, and recognition of objects. An object's pose is estimated by calculating its center of gravity (position) and principal axis (orientation). Recognizing an object and distinguishing between different objects is accomplished by implementing algorithms which estimate an object's perimeter (shape) and area (size). This work demonstrates that a low-resolution tactile array sensor is capable of providing the information that is required for many robotic applications in which objects must be acquired, manipulated, and recognized. Such a system provides a low-cost alternative to more conventional vision-based systems.     

Segmenting sensor data for activity monitoring in smart environments

Within a smart environment, sensors have the ability to perceive changes of the environment itself and can therefore be used to infer high-level information such as activity behaviours. Sensor events collected over a period of time may contain several activities. The fundamental process of any automatic activity monitoring system is therefore to process streams of sensor events and detect occurrences of activities. In this study, we propose three segmentation algorithms to separate time series sensor data into segments to be further processed by an activity recognition algorithm. A preliminary evaluation of the approaches developed has been conducted on a data set collected from a single person living in an apartment over a period of 28 days. The results show that the proposed approaches can segment sensor data to detect activities and infer sensor segments to recognise activities with accuracies of 81.6, 81.6 and 82.9%, respectively.     

Computational cost of image registration with a parallel binary array processor

The application of a simulated binary array processor (BAP) to the rapid analysis of a sequence of images has been studied. Several algorithms have been developed which may be implemented on many existing parallel processing machines. The characteristic operations of a BAP are discussed and analyzed. A set of preprocessing algorithms are described which are designed to register two images of TV-type video data in real time. These algorithms illustrate the potential uses of a BAP and their cost is analyzed in detail. The results of applying these algorithms to FLIR data and to noisy optical data are given. An analysis of these algorithms illustrates the importance of an efficient global feature extraction hardware for image understanding applications.