Application of SPRT to an image data sequence for a remote monitoring system

Recently, remote monitoring camera systems have been widely used for security. In such systems, one important function is that the system automatically detects any change in the scenes from the monitoring cameras. In wireless remote monitoring camera systems, the images of the scenes are generally transmitted as compressed data (e.g., JPEG file), because of the capacity of the wireless channel. This article shows the automated detection of the change point in time-series data of compressed JPEG file quantity (Kbytes) from the monitoring camera by applying the sequential probabilistic ratio test (SPRT) and the Chow test, which is well known as a standard method for detecting structural change in time-series data.     

Temporal difference learning applied to sequential detection

This paper proposes a novel neural-network method for sequential detection, We first examine the optimal parametric sequential probability ratio test (SPRT) and make a simple equivalent transformation of the SPRT that makes it suitable for neural-network architectures. We then discuss how neural networks can learn the SPRT decision functions from observation data and labels. Conventional supervised learning algorithms have difficulties handling the variable length observation sequences, but a reinforcement learning algorithm, the temporal difference (TD) learning algorithm works ideally in training the neural network. The entire neural network is composed of context units followed by a feedforward neural network. The context units are necessary to store dynamic information that is needed to make good decisions. For an appropriate neural-network architecture, trained with independent and identically distributed (iid) observations by the TD learning algorithm, we show that the neural-network sequential detector can closely approximate the optimal SPRT with similar performance. The neural-network sequential detector has the additional advantage that it is a nonparametric detector that does not require probability density functions. Simulations demonstrated on iid Gaussian data show that the neural network and the SPRT have similar performance.     

Detecting structural anomalies of quadcopter UAVs based on LSTM autoencoder

Detecting a structural anomaly, such as a damaged propeller or motor, is crucial for mission-critical operation of unmanned aerial vehicles (UAVs). The existing solutions often fail to detect structural anomalies because the pre-defined parameters required for the solution are limited in reflecting the flight pattern or the external environment, such as wind conditions. In this paper, we propose a method for detecting structural anomalies in quadcopter UAVs, using only regular data and specifically considering flight patterns and runtime flight conditions. To this end, we employ a long short-term memory (LSTM) autoencoder model to learn complex features of regular flight data. While flying the UAV, the trained model estimates the degree of outlierness of the incoming data and assesses abnormal behavior of UAV by adaptively considering its movement. This way, the proposed method accurately detects structural anomalies in UAVs regardless of the runtime environment or flight mission. Our experiment results with an off-the-shelf UAV show that the proposed approach detects diverse structural anomalies by an average of 98.6% specificity and 90.3% sensitivity.     

Fault detection based on one-class deep learning for manufacturing applications limited to an imbalanced database

Despite extensive studies for the industrial applications of deep learning, its actual usage in manufacturing sites has been extremely restrained by the difficulty in obtaining sufficient industrial data, especially for production failure cases. In this study, we introduced a fault-detection module based on one-class deep learning for imbalanced industrial time-series data, which consists of three submodules, namely, time-series prediction based on deep learning, residual calculation, and one-class classification using one-class support vector machine and isolation forest. Four different networks were used for the time-series prediction: multilayer perception (MLP), residual network (ResNet), long–short-term memory (LSTM), and ResNet–LSTM, each trained with the one-class data having only the production success cases. We adopted the residuals of the deep-learning prediction as an elaborated feature for the construction of the one-class classification. We also tested the fault-detection module with the actual mass production data of a die-casting process. By adopting the features elaborated by the deep-learning time-series prediction, we showed that the total accuracy of the one-class classification significantly improved from 90.0% to 96.0%. Especially for its capability to detect production failures, the accuracy improved from 84.0% to 96.0%. The area under the receiver operating characteristics (AUROC) also improved from 87.56% to 98.96%. ResNet showed the best performance for detecting production failures, whereas ResNet–LSTM produced better results for ensuring the production success. Our results suggest that the one-class deep learning is a promising approach for extracting important features from time-series data to realize a one-class fault-detection module.     

A change detection method using spatial-temporal-spectral information from Landsat images

ABSTRACT

Remote sensing data and techniques are reliable tools for monitoring land cover and land-use change. For time-series change detection algorithms, detecting the breakpoints accurately is the key element. However, the current state-of-art algorithms are vulnerable to cloud/cloud shadow or noises in the time-series imagery. The objective of this study is to develop a new method to detect land cover change using Landsat imagery by integrating temporal, spectral and spatial information to increase the accuracy of breakpoints detection. In the temporal dimension, the time-series model is decomposed into seasonality and trend. Due to different land cover types corresponding to different seasonal characteristics, breakpoints exist only in the seasonal component. In the spectral dimension, two-step judgement is applied. The first judgement detects a change when the seasonal breakpoint positions are the same in different spectral bands. The second judgement involves detecting a changed pixel when the classification result indicates different types on either side of the breakpoint. In the spatial dimension, neighbour information is utilized to control the false-positive rate. Experimental results using all available Landsat images acquired between 2001 and 2006 in Kansas City, US, illustrate the effectiveness and stability of the proposed approach. All pixels were used for assessing the classification and change detection accuracy compared with National Land Cover Database products. The overall accuracy of classification into eight categories was about 81% and the accuracy of change detection was 88%. Maps of timing of breaks and change times are also provided in this article.     

Automatic detection of forest fire disturbance based on dynamic modelling from MODIS time-series observations

Forest disturbances provide an important reference and a basis for studying the carbon cycle, biodiversity, and eco-social development. Remote sensing is a promising data source for monitoring forest ecosystem dynamics and detecting disturbance areas. This research used a seasonal trend method to model Moderate Resolution Imaging Spectroradiometer (MODIS) normalized difference vegetation index (NDVI) time series from 2007 to 2011 recursively with a fixed-size temporal sliding window and a step length of 1 (i.e. 16 days). Model parameter variations were monitored to detect changes in the structure of the time-series data. Significant changes in the time-series structure were captured as disturbance signals. The method was applied to the 2009 Minto Flats fire in Alaska, USA, and the forest-disturbance detection results obtained using the proposed method essentially agreed with the Monitoring Trends of Burned Severity data set. This result indicates that the proposed method can reliably reveal the occurrence of forest fire disturbances. Moreover, because the model parameter variations reflect the disturbance signal, and the modelling and detection process requires only MODIS NDVI time-series data without any other ancillary ground information, the disturbance area can be detected effectively and automatically.     

基于Chow检验的最优分段建模

基于已知变点Ｃｈｏｗ检验问题研究，提出了最优二分段建模方法，该方法将变结构参数稳定性分析、变结构点数的检测、变步位置的诊断与估计及变结构检验统一处理，计算方便、实用。经济实例的分析及仿真研究结果都验证了所提方法的有效性。     

Generating test data for both paths coverage and faults detection using genetic algorithms: multi-path case

Generating test data that can expose the faults of the program is an important issue in software testing. Although previous methods of covering path can generate test data to traverse target path, the test data generated by these methods are difficult in detecting some low-probabilistic faults that lie on the covered paths. We present a method of generating test data for covering multiple paths to detect faults in this study. First, we transform the problem of covering multiple paths and detecting faults into a multi-objective optimization problem with constraint, and construct a mathematical model for it. Then, we give a strategy of solving the model based on a weighted genetic algorithm. Finally, we apply our method to several real-world programs, and compare it with several methods. The experimental results confirm that the proposed method can more efficiently generate test data that not only traverse the target paths but also detect faults lying in them than other methods.     

Optimal randomized RANSAC

A randomized model verification strategy for RANSAC is presented. The proposed method finds, like RANSAC, a solution that is optimal with user-specified probability. The solution is found in time that is (i) close to the shortest possible and (ii) superior to any deterministic verification strategy. A provably fastest model verification strategy is designed for the (theoretical) situation when the contamination of data by outliers is known. In this case, the algorithm is the fastest possible (on average) of all randomized \\RANSAC algorithms guaranteeing a confidence in the solution. The derivation of the optimality property is based on Wald's theory of sequential decision making, in particular a modified sequential probability ratio test (SPRT). Next, the R-RANSAC with SPRT algorithm is introduced. The algorithm removes the requirement for a priori knowledge of the fraction of outliers and estimates the quantity online. We show experimentally that on standard test data the method has performance close to the theoretically optimal and is 2 to 10 times faster than standard RANSAC and is up to 4 times faster than previously published methods.     

故障检测与诊断的SPRT方法及其修正

本文简单介绍了基于对新息序列处理而进行故障检测与诊断的 SPRT 法,然后针对这种方法有检测迟延的不足,给出了均值和方差都发生变化的修正的 SPRT 检测与诊断方法,并做了一些分析.     

Detecting local regions of change in high-dimensional criminal or terrorist point processes

A method is presented for detecting changes to the distribution of a criminal or terrorist point process between two time periods using a non-model-based approach. By treating the criminal/terrorist point process as an intelligent site selection problem, changes to the process can signify changes in the behavior or activity level of the criminals/terrorists. The locations of past events and an associated vector of geographic, environmental, and socio-economic feature values are employed in the analysis. By modeling the locations of events in each time period as a marked point process, we can then detect differences in the intensity of each component process. A modified PRIM (patient rule induction method) is implemented to partition the high-dimensional feature space, which can include mixed variables, into the most likely change regions. Monte Carlo simulations are easily and quickly generated under random relabeling to test a scan statistic for significance. By detecting local regions of change, not only can it be determined if change has occurred in the study area, but the specific spatial regions where change occurs is also identified. An example is provided of breaking and entering crimes over two-time periods to demonstrate the use of this technique for detecting local regions of change. This methodology also applies to detecting regions of differences between two types of events such as in case-control data.     

新的全球定位系统弱信号高灵敏捕获算法

通过对弱信号条件下的全球定位系统(GPS)捕获算法的分析,建立了相干累加—非相干累加结合捕获算法的信号模型及检测概率模型。为了提高强弱信号并存时GPS卫星信号的捕获性能,提出一种采用序贯概率比检测方法的GPS捕获算法。对该方法和相干累加—非相干累加算法的检测概率、时间复杂度进行了分析比较,并进行了仿真验证。通过理论分析和计算机仿真,证明该方法在保证较高检测概率性能情况下,可以有效地缩短强弱信号并存时的检测时间,提高对GPS弱信号的捕获性能。     

兼具柔顺与安全的助行机器人运动控制研究

针对助行机器人的柔顺性和安全性问题,基于多传感器系统融合技术,本文提出了一种能够兼具柔顺与安全的助行机器人运动控制方法.首先介绍了助行机器人的机械结构、控制原理以及多传感器系统,然后根据机器人多传感器系统,设计出各传感器相对应的用户意图估计方法,提出了一种基于多传感器融合的助行机器人柔顺运动控制算法.分析用户可能发生的跌倒模式,使用基于卡尔曼滤波（Kalman filter,KF）的序贯概率比检验（Sequential probability ratio test,SPRT）方法和决策函数来判断用户是否会跌倒,并判断处于哪种跌倒模式.最后,通过助行机器人柔顺运动控制实验和用户跌倒检测实验验证了算法的有效性.     

基于SMT方法的测试性验证试验方案设计

针对现有导弹测试性验证试验方案确定的故障样本量较大以及序贯类试验方案最大样本量上界无法确定的问题,提出一种基于序贯网图检验（SMT）方法的测试性验证试验方案。通过对序贯概率比检验（SPRT）的检验问题进行拆分,在确定检验点数目的基础上,提出一种检验点取值与最大样本量的优化求解方法。同时考虑最大样本量较大的情形,基于承制方风险和使用方风险设计了SMT方法的截尾策略。通过案例验证与经典验证方法以及SPRT方法进行对比分析。结果表明,该方法能控制验证试验的样本量且所确定的平均样本量也优于其他方法,导弹装备的测试性验证试验的实施具备实际指导意义。     

基于深度学习的时间序列数据异常检测方法

针对类间分布不平衡的时间序列数据的异常检测问题,提出了一种基于深度卷积神经网络的检测方法.首先采用抽样法对不平衡时间序列数据进行预处理;其次,将处理后的时间序列数据转换为尺度一致、时长一致的片段;最后将数据送入具有4层隐藏层结构的卷积神经网络模型中进行异常检测.实验结果表明,所提方法弥补了现存的检测技术由于忽略数据分布的偏斜性而造成的少数类检测精度低的缺点,并通过与现有的时间序列分类方法的比较,验证了所提方法的高效性.     

基于MSET和SPRT的导弹振动故障诊断

提出了一种基于多元状态估计技术(MSET,multivariate state estimation techniques)和序贯概率比检验(SPRT,sequential probability ratio test)的导弹机构振动故障诊断方法.首先建立常规情况下导弹3处振动传感器所收集的振动信号的关联模型;然后根据导弹3处异常振动信号的当前观测测特征向量与各建模样本特征向量之间的相似性程度,使用MSET对当前异常信号特征向量进行估计,得到与异常信号特征向量相对应的估计残差;最后使用SPRT对异常信号的估计残差进行均值和方差检验,确定系统的工作状态.仿真结果表明,MSET可有效地增强故障状态下的信号特征呈现,而SPRT可在较少的周期内实现对弹体机构异常工作的识别,MSET和SPRT的结合有效地实现了对导弹机构异常工作的早期诊断.     

Bayes序贯决策法在舰炮武器系统试验中的应用

传统的试验分析与评定方法需要建立在大子样的基础上进行,而舰炮武器系统进行大量的试验是难以承受的,针对这一问题,提出了Bayes序贯决策法,并对Bayes截尾序贯决策法在舰炮武器系统试验中的应用进行了研究,该方法的应用可以减少试验样本数量,节省试验经费,降低试验风险。     

Structural Periodic Measures for Time-Series Data

This work motivates the need for more flexible structural similarity measures between time-series sequences, which are based on the extraction of important periodic features. Specifically, we present non-parametric methods for accurate periodicity detection and we introduce new periodic distance measures for time-series sequences. We combine these new measures with an effective metric tree index structure for efficiently answering k-Nearest-Neighbor queries. The goal of these tools and techniques are to assist in detecting, monitoring and visualizing structural periodic changes. It is our belief that these methods can be directly applicable in the manufacturing industry for preventive maintenance and in the medical sciences for accurate classification and anomaly detection.

---

     

Multi-scale detection and interpretation of spatio-temporal anomalies of human activities represented by time-series

Spatio-temporal patterns of human activities can be affected by events, such as extreme weather. Events cause anomalies that could be expressed by abnormal activity patterns deviating from the inherent ones. The detection of spatio-temporal anomalies thus helps to understand the implicit influencing mechanism with which the external factors affect human activities. Existing methods of spatio-temporal anomaly detection usually treat the temporal information as attributes of spatial units, which is an over-simplification as it ignores complex temporal patterns (e.g., periodic components of time-series). Moreover, as the spatio-temporal resolutions affect expressed characteristics of anomalies, the sensitivity of anomalies to scale is also worth investigating. This study intends to detect and interpret the spatio-temporal anomalies of human activities from a multi-scale perspective. Being different from the single-scale consideration and independent consideration of multiple scales, this research investigates how the anomalies' characteristics change at multiple scales by anomaly matching. The criteria of anomaly matching are the overlapping degree of spatio-temporal influence ranges of anomalies. It helps to specify the events that caused the expressed anomalies. Besides, we introduce the time-series decomposition methods to decompose complex temporal patterns, highlighting the abnormal changes in activity patterns. The study is validated using a multi-temporal-scale simulation experiment, and a multi-spatial-scale experiment based on taxi data in Beijing. Results show that the multi-scale method can detect various anomalies. Moreover, obtained multi-scale characteristics of anomalies are easy to compare with external data, and thus benefit anomaly interpretation (validated by two sample anomalies). This study highlights the significance of scales in anomaly detection of human activities and provides references for related works.     

Model specification in panel data unit root tests with an unknown break

Although the impact of structural breaks on testing for unit root has been studied extensively for univariate time-series, such impact on panel data unit root tests is still relatively unknown. A major issue is the choice of model in accommodating different types of break prior to testing for unit root. Model misspecification has been known to affect unit root tests performance in the univariate case but the effect of misspecification on panel tests is still unknown. This paper has two objectives: (i) it proposes a new test for unit root in the presence of structural break for panel data. The test allows the intercepts, the trend coefficients or both to change at different date for different individuals. Moreover, the test allows for the possibility that only some, but not all, of the individuals experienced structural breaks. Under some mild assumptions, the test statistics is shown to be asymptotically normal which greatly facilitates valid inferences. (ii) This paper provides a systematic study on the impact of structural instability on testing for unit root using Monte Carlo Simulation. The results show that correct specification is crucial for unit root testing in the presence of structural instability. In addition, the proportion of individuals experienced structural instability can also affect the performance of the test substantially.