Active learning for detection of mine-like objects in side-scan sonar imagery

A data-adaptive algorithm is presented for the selection of the basis functions and training data used in classifier design with application to sensing mine-like targets with a side-scan sonar. Automatic detection of mine-like targets using side-scan sonar imagery is complicated by the variability of the target, clutter, and background signatures. Specifically, the strong dependence of the data on environmental conditions vitiates the assumption that one may perform a priori algorithm training using separate side-scan sonar data collected previously. In this paper, a novel active-learning algorithm is developed based on kernel classifiers with the goal of enhancing detection/classification of mines without requiring an a priori training set. It is assumed that divers and/or unmanned underwater vehicles (UUVs) may be used to determine the binary labels (target/clutter) of a small number of signatures from a given side-scan collection. These sets of signatures and associated labels are then used to train a kernel-based algorithm with which the remaining side-scan signatures are classified. Information-theoretic concepts are used to adaptively construct the form of the kernel classifier and to determine which signatures and associated labels would be most informative in the context of algorithm training. Using measured side-looking sonar data, the authors demonstrate that the number of signatures for which labels are required (via diver/UUV) is often small relative to the total number of potential targets in a given image. This procedure designs the detection/classification algorithm on the observed data itself without requiring a priori training data and also allows adaptation as environmental conditions change.     

Model-based covariance mean variance classification techniques:algorithm development and application to the acoustic classification ofzooplankton

For inversion problems in which the theoretical relationship between observed data and model parameters is well characterized, a promising approach to the classification problem is the application of techniques that capitalize on the predictive power of class-specific models. Theoretical models have been developed for three zooplankton scattering classes (hard elastic-shelled, e.g., pteropods; fluid-like, e.g., euphausiids; and gas-bearing, e.g., siphonophores), providing a sound basis for model-based classification approaches. The covariance mean variance classification (CMVC) techniques classify broad-band echoes from individual zooplankton based on comparisons of observed echo spectra to model space realizations. Three different CMVC algorithms were developed: the integrated score classifier, the pairwise score classifier, and the Bayesian probability classifier; these classifiers assign observations to a class based on similarities in covariance, mean, and variance while accounting for model spare ambiguity and validity. The CMVC techniques were applied to broad-band (~350-750 kHz) echoes acquired from 24 different zooplankton to invert for scatterer class and properties. All three classification algorithms had a high rate of success with high-quality high SNR data. Accurate acoustic classification of zooplankton species has the potential to significantly improve estimates of zooplankton biomass made from ocean acoustic backscatter measurements     

Simultaneous acoustic observations of turbulence and zooplankton in the ocean

Models and laboratory experiments show that zooplankton may locate food more easily in turbulent waters, but whether plankton seek or avoid turbulence in the ocean is an open question. It is difficult to measure turbulence and plankton simultaneously and with the necessary spatial resolution using traditional methods (nets and airfoil shear sensors). Acoustics is commonly used to survey zooplankton abundance and recent studies have shown that stratified turbulence can also be a significant source of sound scatter. This may seem like more of a complication than a boon for those aiming to use acoustics to observe plankton in turbulence. We present acoustic data, however, that show that zooplankton and turbulence can be observed simultaneously with a single 307 kHz sounder. The different natures of the two targets (discrete targets versus a volume effect) allow them to be distinguished. The key is sampling the same targets at multiple ranges. The volume scattering strength of a discrete target will increase as the target nears the sounder, because the volume sampled decreases. Turbulence, as a volume scattering effect, has little range dependence to its volume scattering strength.     

地波雷达目标检测跟踪联合处理粒子滤波方法

针对固定粒子数PF-TBD算法计算量大、复杂环境下地波雷达海上船只目标检测与跟踪性能不佳的问题,本文将粒子滤波方法应用于地波雷达船只目标检测与跟踪中,提出了基于自适应粒子滤波的地波雷达目标检测与跟踪联合处理方法。该方法结合地波雷达回波谱中目标展宽特性,充分利用了地波雷达回波谱中面目标的粒子权重信息来设置粒子自适应采样策略,提高了目标检测和跟踪联合处理的效果。通过地波雷达实测数据的目标跟踪结果及与同步AIS信息的比对分析,结果表明:提出的检测跟踪联合处理方法在对低信噪比、快速机动等复杂环境下的多目标跟踪时,可提高目标整体跟踪性能。     

基于EOF分析对南海西北海域水体光谱特性的研究

This study presents an analysis of the spectral characteristics of remote sensing reflectance(R_(rs)) in northwestern South China Sea based on the in situ optical and water quality data for August 2018.R_(rs)was initially divided into four classes,classes A to D,using the max-classification algorithm,and the spectral properties of whole R_(rs) were characterized using the empirical orthogonal function(EOF) analysis.Subsequently,the dominant factors in each EOF mode were determined.The results indicated that more than 95% of the variances of R_(rs) are partly driven by the back-scattering characteristics of the suspended matter.The initial two EOF modes were well correlated with the total suspended matter and back-scattering coefficient.Furthermore,the first EOF modes of the four classes of R_(rs)(A-D R_(rs)-EOF_1) significantly contributed to the total variances of each R_(rs) class.In addition,the correlation coefficients between the amplitude factors of class A-D R_(rs)-EOF_1 and the variances of the relevant water quality and optical parameters were better than those of the unclassified ones.The spectral shape of class AR_(rs)-EOF_1 was governed by the absorption characteristic of chlorophyll a and colored dissolved organic matter(CDOM).The spectral shape of class B R_(rs)-EOF_1 was governed by the absorption characteristic of CDOM since it exhibited a high correlation with the absorption coefficient of CDOM(a_g(λ)),whereas the spectral shape of class C R_(rs)-EOF_1 was governed by the back-scattering characteristics but not affected by the suspended matter.The spectral shape of class D R_(rs)-EOF_1 exhibited a relatively good correlation with all the water quality parameters,which played a significant role in deciding its spectral shape.     

Sonar tracking of multiple targets using joint probabilistic data association

The problem of associating data with targets in a cluttered multi-target environment is discussed and applied to passive sonar tracking. The probabilistic data association (PDA) method, which is based on computing the posterior probability of each candidate measurement found in a validation gate, assumes that only one real target is present and all other measurements are Poisson-distributed clutter. In this paper, a new theoretical result is presented: the joint probabilistic data association (JPDA) algorithm, in which joint posterior association probabilities are computed for multiple targets (or multiple discrete interfering sources) in Poisson clutter. The algorithm is applied to a passive sonar tracking problem with multiple sensors and targets, in which a target is not fully observable from a single sensor. Targets are modeled with four geographic states, two or more acoustic states, and realistic (i.e., low) probabilities of detection at each sample time. A simulation result is presented for two heavily interfering targets illustrating the dramatic tracking improvements obtained by estimating the targets' states using joint association probabilities.     

基于频域典型相关分析的水下目标回波识别

水下目标回波的特征提取与分类识别是当前主动声纳关键技术之一。采用基于回波频域特性的典型相关分析算法(CCA:Canonical Correlation Analysis)提取回波的特征,这些特征集中体现了不同目标回波的综合相关特性。设计合适的支持向量机分类器,并获得识别结果。利用这一方法对湖试中的不同目标回波进行分类识别,分析了不同接收信噪比条件下的性能,获得了理想的结果。     

基于分形纹理特征的侧扫声呐图像沉船识别方法研究

为提高侧扫声呐图像中沉船等目标信息的识别精度和识别效率,根据盒维数、毯维数与多重分形谱的侧扫声呐图像纹理特征提取算法,构建了基于分形纹理特征的Adaboost级联分类器沉船目标识别流程。结合实测侧扫声呐图像数据进行水下沉船识别实验,并与灰度共生矩阵和Tamura纹理特征的识别结果进行对比。研究表明,基于分形纹理特征的识别方法综合考虑了图像全局与局部纹理特征,且不依赖人工选取阈值参数与特征向量,可有效提高目标识别精度和识别效率。     

基于水声环境空间中多模态深度融合模型的目标识别方法研究

随着对水下目标特性研究的深入和声学探测技术的发展,基于单模态的阵列式信息融合或基于空间信息的分布式信息融合的水下目标识别方法研究已有一定成果,但针对复杂海况导致单一物理场或单一融合层次的系统识别性能提高有限等方面影响的水下目标识别方法研究还有所不足,因此,开展基于多模态深度融合模型的水下目标识别方法研究可利用模态互补,共享信息而提升识别率。文中在国内外研究基础上,深入研究了基于到达时差法和多模态方法组合的检测方法,初步形成了基于水声环境空间中多模态深度融合模型的识别框架,开展了海洋中典型自然与人为事件的信号分析与特征提取,并在此基础上,设计新型基于海底基站的被动识别系统。该系统同步记录和由位置等组成的时间序列标记声、磁和压数据,可实现高精度、高分辨率的识别。本研究可满足未来海洋观测对高性能水下目标探测、定位和跟踪系统的迫切需要,为海洋安全监管、海洋突发事件应急响应等领域提供新的技术手段和科学参考。     

Study on dim target detection and discrimination from sea clutter

Dim target detection from sea clutter is one of the difficult topics in ocean remote sensing application. By aiming at the shortcoming of false alarms when using track before detect (TBD) based on dynamic programming, a new discrimination method called statistics of direction histogram (SDH) is proposed, which is based on different features of trajectories between the true target and false one. Moreover, a new series of discrimination schemes of SDH and Local Extreme Value method (LEV) are studied and applied to simulate the actually measured radar data. The results show that the given discrimination is effective to reduce false alarms during dim targets detection.     

基于多尺度小波特征的高光谱影像亚像素目标识别

提出了一种基于多分辨率小波高频特征系数的高光谱遥感影像亚像素目标识别方法。首先利用多尺度小波变换将光谱信号分解为不同尺度的高频特征信号,然后借助接收操作特性曲线(ROC)和马氏距离投影寻踪求取一维最佳识别特征,最后通过高斯最大似然决策函数求解亚像素目标的存在概率。通过38种小波函数的高光谱数据实验证明,该方法对亚像素目标的识别效果较好。     

Examination of targeting methods in a simplified setting

The effectiveness of 2 methods for targeting observations is examined using a T21 L3 QG model in a perfect model context. Target gridpoints are chosen using the pseudo‐inverse (the inverse composed of the first three singular vectors only) and the quasi‐inverse or backward integration (running the tangent equations with a negative time‐step). The effectiveness of a target is measured by setting the analysis error to zero in a region surrounding the target and noting the impact on the forecast error in the verification region. In a post‐time setting, when the targets are based on forecast errors that are known exactly, both methods provide targets that are significantly better than targets chosen at random within a broad region upstream of the verification region. When uncertainty is added to the verifying analysis such that the forecast error is known inexactly, the pseudo‐inverse targets still perform very well, while the backward integration targets are degraded. This degradation due to forecast uncertainty is especially significant when the targets are a function of height as well as horizontal position. When an ensemble‐forecast difference is used in place of the inexact forecast error, the backward integration targets may be improved considerably. However, this significant improvement depends on the characteristics of the initial‐time ensemble perturbation. Pseudo‐inverse targets based on ensemble forecast differences are comparable to pseudo‐inverse targets based on exact forecast errors. Targets based on the largest analysis error are also found to be considerably more effective than random targets. The collocation of the backward integration and pseudo‐inverse targets appears to be a good indicator of target skill.     

Improved empirical descriptions for acoustic surface backscatter inthe ocean

Using Signals, Underwater Sound (SUS) explosive charges as broad-band acoustic sources, a high-quality set of surface scattering strengths was measured throughout the Critical Sea Test (CST) experiments. These measurements were made for wind speeds ranging from ~1 to 18 m/s and covered grazing angles from ~5° to 30° and frequencies from ~60 to 1000 Hz. A new empirical algorithm was developed based on a multiparameter multidimensional nonlinear fit to all the SUS data from CST-1 through CST-7. This new algorithm returns the surface scattering strength for a given frequency, grazing angle, and wind speed. The new formulation explored the use of backaveraging the wind speeds in time (as opposed to using the instantaneous wind speed) to allow for the influence of processes driven by the wind history, In this paper, details of the development of this new algorithm will be discussed, comparisons with earlier prediction algorithms (the Ogden-Erskine and Chapman-Harris algorithms) will be made, and the important differences between the various CST SUS data sets will be highlighted and possible explanations offered. Finally, suggestions for further improvements to the algorithm are made     

Development and analysis of echo classification using time delays

The results from an investigation of an analytically based method for determining the performance of echo classifiers are presented. In particular, the problem of classifying echo waveforms reflected from objects that are composed of multiple scatterers is considered. The time delays between the multiple echo returns from the individual scattering centers that characterize an object are investigated as features. A generic stochastic point scatterer model is developed for representing the classes of reflecting objects which are of interest. The model allows for uncertainty in prior knowledge about the exact relative location of the individual component scatterers or uncertainty in the delay measurements. A classifier decision algorithm, in the form of a general optimum Bayesian binary classification decision rule suitable for a large variety of classification problems, is derived for the case when the orientation of the reflecting object is known. The case of unknown aspect angle is addressed through the numerical implementation and analysis of two classifiers. The associated performance for all three classifiers is obtained in terms of the probability of error and tied to standard sonar equation parameters. Example binary classification problems are presented and analyzed and some general observations made. A pragmatic framework is established within which complex echo classification issues can be further examined     

基于LLE低维嵌入的高光谱影像分类

针对高光谱数据中内在的非线性流行结构,分析了LLE低维嵌入算法的基本原理,给出了该算法的计算步骤。介绍了模糊ISODATA分类算法的基本思想,在计算目标函数中,利用测地距离代替欧氏距离,对模糊ISODATA分类算法进行改进。利用两套PHI高光谱影像数据,在LLE低维嵌入结果上实现了ISODATA分类实验。结果表明:LLE低维嵌入后的数据能够降低ISODATA影像分类的迭代次数与计算时间,提高分类的效率;与原始ISODATA分类算法相比,改进的ISODATA分类算法能够更好地挖掘类别之间的自组织关系,提高分类的可靠性。     

利用极化SAR数据进行土地覆盖分类研究

基于相关矩阵特征向量的目标分解将地物回波复杂的散射过程分解成相互独立的三种单一散射分量:单向散射、双向散射和交叉散射,分别对应各自的目标相关矩阵。目标分解技术降低了散射回波之间的相关性,有利于分析地物散射机理,有助于提高分类精度。对荷兰F levoland地区全极化数据进行分解,经过试验和相关性分析,选用7种数据形成多参数数据组合,对其进行最大似然监督分类,同时进行常规三种极化加相位差的分类和基于复W ishart分布的最大似然分类,逐像元计算混淆矩阵,分析对比三种分类结果的精度,试验表明:相对于常规数据组合分类,基于复W ishart分布的监督分类可以小幅度提高分类精度,而利用目标分解得到多参数组合数据进行分类则有大幅度的提高。     

Fluid-Filled Passive Sonar Calibration Spheres: Design, Modeling, and Measurement

Hollow spheres have long been used as simple underwater targets for testing acoustic projector systems. While spheres offer a mathematically simple shape with a resolvable scattering strength, their usage as a passive target has been less successful due to the complicated manner in which a hollow sphere scatters energy from its exterior and interior as a function of frequency and temperature. Furthermore, a sphere's aspect independent scattering requires a surface area that is physically much greater than a wavelength which in turn requires mechanical support systems that are also large, often with target strengths that rival that of the test target itself. This paper discusses the development of several thin-walled spheres, ranging in diameters from 0.1524 to 0.4953 m, filled with a high-density fluid, to be used collectively as calibrated underwater sonar targets in the 5–50-kHz frequency range and an additional 0.4953-m diameter sphere tested over the range of 5–120 kHz. The combination of the spherical shape and focusing effects of the fluid enhances the acoustic scattering strength of the sphere and produces a significantly greater backscattered response than a rigid sphere. A simple theoretical model is presented to compare several fill fluid possibilities and is then used to compare the chosen fluid, fluorolube, against measured data for each sphere.      

基于极化参数的SVM海上溢油识别

溢油对海洋环境造成的危害越来越大,及早发现对于减灾防灾具有重要意义。目前,运用极化SAR进行溢油探测已成为遥感监测的一个重要方面,本文基于SIR-C数据,开展极化SAR的溢油监测,提取极化参数熵H,散射角α和反熵A,运用SVM监督分类方法,进行溢油信息提取。结果表明,基于SVM的分类精度要强于基于H-α分类的分类结果。     

Transient sonar signal classification using hidden Markov modelsand neural nets

In ocean surveillance, a number of different types of transient signals are observed. These sonar signals are waveforms in one dimension (1-D). The hidden Markov model (HMM) is well suited to classification of 1-D signals such as speech. In HMM methodology, the signal is divided into a sequence of frames, and each frame is represented by a feature vector. This sequence of feature vectors is then modeled by one HMM. Thus, the HMM methodology is highly suitable for classifying the patterns that are made of concatenated sequences of micro patterns. The sonar transient signals often display an evolutionary pattern over the time scale. Following this intuition, the application of HMM's to sonar transient classification is proposed and discussed in this paper. Toward this goal, three different feature vectors based on an autoregressive (AR) model, Fourier power spectra, and wavelet transforms are considered in our work. In our implementation, one HMM is developed for each class of signals. During testing, the signal to be recognized is matched against all models. The best matched model identifies the signal class. The neural net (NN) classifier has been successfully used previously for sonar transient classification. The same set of features as mentioned above is then used with a multilayer perceptron NN classifier. Some experimental results using “DARPA standard data set I” with HMM and MLP-NN classification schemes are presented. A combined NN/HMM classifier is proposed, and its performance is evaluated with respect to individual classifiers