基于模糊C均值聚类-支持向量机的海底沉积物分类识别

在总结了目前海底底质分类研究的基础之上,率先提出利用计算机数值模拟技术对海底底质进行分类识别研究。相较于目前海底底质分类研究中所使用的水槽实验法,提出采用计算机数值正演技术模拟实际地震勘探中数据采集过程。在分类识别算法上,分别采用支持向量机(SVM)和模糊C均值聚类(FCM)算法对采集的数据进行分类,为使支持向量机分类识别率达到最大,引入差分进化算法对支持向量机中关键参数进行最优化搜索,并研究了向原始地震记录中加入10%,30%,50%的高斯白噪音时算法的稳定性。在分析了这两种算法分类识别的正确率及其各自的优缺点后,提出了海底底质分类识别的两步法,即(1)先利用模糊C均值聚类进行一粗糙的预测分类,在每一类中挑选聚类性较好的数据作为支持向量机的训练样本;(2)将上一步中筛选的样本作为支持向量机的训练样本,并用差分进化算法优化支持向量机分类参数,再利用训练好的支持向量机对其余数据做预测分类。鉴于计算机数值模拟的可重复性、高效快速性及本文提出的模糊C均值聚类-支持向量机方法的鲁棒性,为便于开展进一步研究,归纳总结了一套行之有效的采用计算机数值模拟技术开展海底底质分类识别研究的一般化流程。     

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     

基于最优特征集的HY-1C卫星海岸带成像仪影像海冰分类方法研究

基于海洋一号C（HY-1C）卫星海岸带成像仪（CZI）遥感影像,提出了一种基于最优特征集的支持向量机海冰分类方法。分别提取CZI影像的光谱特征和纹理特征,采用基于距离可分性的判据进行特征选择,得到最优特征集,以最优特征集作为支持向量机分类器输入,分别对3期辽东湾海域CZI影像开展海冰分类实验和结果分析。结果表明:本文方法得到的海冰分类结果精度优于仅利用光谱特征或纹理特征的海冰分类精度;基于本文方法的3期影像的海冰分类精度均较高,2020年12月19日、2021年1月10日与2021年1月16日的海冰分类总体精度分别为93.67%、91.75%、84.89%,均在80%以上;利用海冰分类结果图估算海冰面积,发现3期辽东湾海冰面积依次增大,最大约为11 998.98 km2。     

基于国产高分遥感的人工种植红树林种间分类方法研究——以广西茅尾海红树林为例

红树林的种间结构组成对红树林生态系统的健康和发展至关重要,而红树林种间分类问题一直以来都是基于遥感手段的红树林监测中的难点。针对该问题,以人工种植为特点的广西茅尾海红树林遥感种间分类为例,基于面向对象的分类思想,提出了一种现场样本与分割对象相结合的红树林种间分类方法。利用GF-2 PMS1高分辨率卫星遥感影像数据,开展了广西茅尾海红树林湿地典型植被精细分类和空间分布研究,并将分类结果与基于像素和传统面向对象SVM分类方法进行了对比。结果显示:总体上,面向对象分类方法更适合用于茅尾海红树林湿地典型植被分类;对于局部混生明显的区域使用基于像素SVM分类方法效果会更好;传统面向对象分类方法中将整个影像分割对象单元作为训练样本可能会在某种程度上造成负面影响。因此,使用文中提出的样本选择新方法进行面向对象分类精度最高,总体精度达到了93.13%,Kappa为0.89。     

海底声学底质分类的ELM-AdaBoost方法

基于自适应增强算法（AdaBoost）结合极限学习机（ELM）,通过迭代、调整、优化ELM分类器之间的权值,从而构建了具有强鲁棒性、高精度的ELM-AdaBoost强分类器,增强了现有的ELM分类器的稳定性。以珠江口海区侧扫声呐图像为实验数据,对礁石、砂、泥3类典型底质进行分类识别,该方法的平均分类精度超过90%,优于单一ELM分类器的平均分类精度85.95%,也优于LVQ、BP等传统分类器,且在分类所耗时间上也远少于传统分类器。实验结果表明,本文构建的ELM-AdaBoost方法可有效应用于海底声学底质分类,可满足实时底质分类的需求。     

Stability Analysis on Speed Control System of Autonomous Underwater Vehicle

The stability of the motion control system is one of the decisive factors of the control quality for Autonomous Underwater Vehicle (AUV).The divergence of control,which the unstable system may be brought about,is fatal to the operation of AUV.The stability analysis of the PD and S-surface speed controllers based on the Lyapunov' s direct method is proposed in this paper.After decoupling the six degree-of-freedom (DOF) motions of the AUV,the axial dynamic behavior is discussed and the condition is deduced,in which the parameters selection within stability domain can guarantee the system asymptotically stable.The experimental results in a tank and on the sea have successfully verified the algorithm reliability,which can be served as a good reference for analyzing other AUV nonlinear control systems.     

An application of fuzzy fault tree analysis for spread mooring systems

In this paper, a fuzzy fault tree analysis methodology for spread mooring systems is presented. The methodology combines the effects of operational failures and human errors under fuzzy environment for the spread mooring configurations. In conventional fault tree analysis (FTA), which is an established technique in hazard identification, the ambiguous and imprecise events such as human errors cannot be handled efficiently. In addition to this, the tolerances of the probability values of hazards are not taken into account. Moreover, it is difficult to have an exact estimation of the failure rates of the system components or the probability of the occurrence of undesired events due to the lack of sufficient data. To overcome these disadvantages, a fault tree analysis based on the fuzzy set theory is proposed and applied to the spread mooring system alternatives. Furthermore, sensitivity analysis is carried out based on the fuzzy weighted index (FWI) in order to measure the impact of each basic event on the top event. The results show that the fuzzy fault tree risk analysis method (FFTA) is more flexible and adaptive than conventional fault tree analysis for fault diagnosis and hazard estimation of spread mooring systems.     

Closed-loop randomized kinodynamic path planning for an autonomous underwater vehicle

A randomized kinodynamic path planning algorithm based on the incremental sampling-based method is proposed here as the state-of-the-art in this field applicable in an autonomous underwater vehicle. Designing a feasible path for this vehicle from an initial position and velocity to a target position and velocity in three-dimensional spaces by considering the kinematic constraints such as obstacles avoidance and dynamic constraints such as hard bounds and non-holonomic characteristic of AUV are the main motivation of this research. For this purpose, a closed-loop rapidly-exploring random tree (CL-RRT) algorithm is presented. This CL-RRT consists of three tightly coupled components: a RRT algorithm, three fuzzy proportional-derivative controllers for heading and diving control and a six degree-of-freedom nonlinear AUV model. The branches of CL-RRT are expanded in the configuration space by considering the kinodynamic constraints of AUV. The feasibility of each branch and random offspring vertex in the CL-RRT is checked against the mentioned constraints of AUV. Next, if the planned branch is feasible by the AUV, then the control signals and related vertex are recorded through the path planner to design the final path. This proposed algorithm is implemented on a single board computer (SBC) through the xPC Target and then four test-cases are designed in 3D space. The results of the processor-in-the-loop tests are compared by the conventional RRT and indicate that the proposed CL-RRT not only in a rapid manner plans an initial path, but also the planned path is feasible by the AUV.     

基于混合分类规则的成像光谱数据分类研究

分类方法的选择对于成像光谱数据的分类精度有着直接影响,然而由于成像光谱数据的特点,使得分类器的选择变得十分困难。提出了一种基于混合分类规则的成像光谱数据分类方法。实验表明,按照该方法进行成像光谱数据的分类处理,可以得到很高精度的分类结果。     

K-均值聚类算法在多波束底质分类中的应用

针对海底采样点较少时,监督学习训练分类模型困难的问题,研究无监督学习的K-均值聚类分析算法在多波束海底底质分类中的应用。在探讨K-均值聚类分析算法原理的基础上,构建海底底质分类器,针对分类器需预先输入分类结果种类(K值)这一问题,提出了基于底质采样点和分类效果连续性为原则的K值确定方法。实验结果表明:基于K-均值聚类分析算法的海底底质分类器能较好的实现海底底质类型的自动划分,适用于海量多波束底质特征参数的分类。     

一种融合纹理特征与NDVI的随机森林海冰精细分类方法

海冰的精准分类对于掌握海冰生长发育状况,保障航海安全等具有重要意义。由于受数据源和分类方法等影响,使得海冰分类精度提高受限。本文面向高空间分辨率的光学遥感影像,提出了一种融合纹理特征和归一化差分植被指数（NDVI）的海冰精准分类方法,运用随机森林分类器构建海冰分类方法。以青岛胶州湾为实验区,高分二号（GF-2）为实验数据,进行了海冰类型提取,并与其他分类方法进行对比。结果显示:针对GF-2高分辨率光学遥感数据,融合纹理特征和NDVI的随机森林方法,相比于传统的随机森林、支持向量机、自动决策树和融合纹理特征的最大似然分类方法,总体分类精度分别提高13.70%、11.60%、19.22%、29.37%。Kappa系数分别提高0.16、0.13、0.22、0.44。相比于融合纹理特征和归一化水指数（NDWI）的随机森林方法,总体分类精度提高了9.67%,Kappa系数提高了0.09。这表明本文构建的海冰分类方法可有效提高海冰分类精度,为海冰的精确分类提供了一种有效的技术手段。     

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。     

模糊模式识别用于找矿预测

本文提出两类模糊模式识别用于找矿预测的具体方法．用Ｆ－ＰＦＳ法和调节特征因素及其权重以获取最佳分类，聚类中心即作为标准模式．根据单因素判对率确立了因素逆距离权重的概念．在标准模式的模糊向量与已知单元模糊向量之间关系的基础上可以建二线性不等式方程组，从而可解不同因素的距离权重，并进而用贴近度对未知单元进行识别．以上方法应用于鄂东南地区的铜及多金属的找矿预测，结果表明方法有效。成果较好。     

Maneuvering modeling and simulation of AUV dynamic systems with Euler-Rodriguez quaternion method

The purpose of this study is to develop maneuvering models and systems of a simulator to improve the motion performance of autonomous underwater vehicles (AUVs) at the preliminary design stages in advance. The AUVs simulation systems based on the standard submarine equations of motion in six-degree-of-freedom (6-DOF) integrated with the Euler-Rodriguez quaternion method for representing singularity-free AUV attitude and time-saving calculation, and with a nonlinear control model for maneuvering and depth control simulations, time-marching in the fourth-order Runge-Kutta scheme. For validation of the simulation codes, results of the ISiMI AUV open-loop tests including turning test and zigzag test as well as an AUV simulator on the basis of Euler-angle method were used to compare with the quaternion-based AUV simulator. The computational results from the proposed simulator agree well with those from both the ISiMI AUV experiments and the Euler-angle based simulations. Additionally, a new maneuvering procedure, namely "put-out" was implemented to test directional stability for a large-scale AUV in the proposed AUV simulator that can be considered for vehicles in space as well as in constrained planes.     

Classification of transient sonar sounds using perceptually motivated features

This paper describes a novel framework for classifying underwater transient signals recorded by passive sonar. The proposed approach involves two key ideas. Firstly, a feature-selection algorithm is used to identify those acoustic features that optimally model each class of transient sound. Secondly, features that are perceptually motivated are proposed, i.e., they encode information that human listeners are likely to use in transient classification tasks. Three perceptual features are proposed, which encode timbre, the physical material of the sound source, and the temporal context (pattern) in which the transient occurred. The authors show how these features, which are computed over different temporal windows, can be combined to make classification decisions. The performance of the proposed classifier is evaluated on a corpus of transient signals extracted from passive sonar recordings. Specifically, the performance of the perceptual features is compared with spectral features and with those that encode statistics of time, frequency, and power. The present results show that the perceptual features provide valuable cues to the class of a transient. However, the best performing classifier was obtained by selecting a subset of perceptual, spectral, and statistical features in a class-dependent manner.     

Satellite Image Classification Method Using the Dempster–Shafer Approach

In this paper the problem of the supervised classification of satellite images is considered. A new image classification method focused on application under conditions when the training sample is (in particular, considerably) contaminated is proposed. The method is based on using the Dempster–Shafer evidence theory and is applicable both for hyperspectral and multispectral satellite images. Problems of organizing the supervised classification process and content of its constitutive procedures are presented. The developed method has been implemented algorithmically and in software. Results obtained in the classification of hyperspectral images by the proposed method testify to its efficiency.

     

基于资源三号的雷州湾红树林分类研究

基于面向对象的分类方法,不同参数组合会对红树林分类精度产生影响。以雷州半岛东岸附城镇沿海一带为研究区域,探索最优的参数组合以实现红树林的精确分类。利用资源三号(ZY-3)高分影像,基于图像光谱、形状和空间关系特征,对红树林进行分层次提取。结合红树林种类的光谱、空间特征差异,对比分析面向对象方法下不同因子、分割尺度及分类器对应下的分类精度,得出该研究区红树林树种在面向对象分类方法中的最优参数组合。结果表明:基于形状因子0.6+紧致度因子0.6、分割尺度为46的条件下,随机树分类器能有效区分无瓣海桑、白骨壤和秋茄三种红树林,总体精度为87.55%,Kappa系数为0.81。     

Robust control based on feedback linearization for roll stabilizing of autonomous underwater vehicle under wave disturbances

In the case of Autonomous Underwater Vehicle(AUV) navigating with low speed near water surface,a new method for design of roll motion controller is proposed in order to restrain wave disturbance effectively and improve roll stabilizing performance.Robust control is applied,which is based on uncertain nonlinear horizontal motion model of AUV and the principle of zero speed fin stabilizer.Feedback linearization approach is used to transform the complex nonlinear system into a comparatively simple linear system.For parameter uncertainty of motion model,the controller is designed with mixed-sensitivity method based on H-infinity robust control theory.Simulation results show better robustness improved by this control method for roll stabilizing of AUV navigating near water surface.     

ELASTIC CLASSIFICATION OF MODIFIED WATER MASS IN SHALLOW SEA

In this paper, the principle and steps for differentiating water masses by fuzzy cluster method are introduced, and a scalar formula based on Euclidean distance and a method for determining objectively the number of water masses by F-test are proposed. Consequently, a method and specific steps for differentiating modified water masses in shallow sea according to fuzzy elastic classification are given. Computation of the membership degree in which each sample belongs to every water mass determines conveniently and quantitatively the cores, boundaries of water masses and mixed zones. An example for the Huanghai Sea and East China Sea is shown and compared with previous results.     

Artificial intelligence-based hull structural plate corrosion damage detection and recognition using convolutional neural network

With the support of big data and GPU acceleration training, the artificial intelligence technology with deep learning as its core is developing rapidly and has been widely used in many fields. At the same time, feature extraction operations are required by the current image-based corrosion damage detection method in the field of ships, with little effect but consuming the large amount of manpower and financial resources. Therefore, a new method for hull structural plate corrosion damage detection and recognition based on artificial intelligence using convolutional neural network is proposed. The convolutional neural network (CNN) model is trained through a large number of classified corrosion damage images to obtain a classifier model. Then the classifier model is used with overlap-scanning sliding window algorithm to recognize and position the location of corrosion damage. Finally, the damage detection pattern for hull structural plate corrosion damage as well as other types of superficial structural damage using convolutional neural network is proposed, which can accelerate the application of artificial intelligence technology into the field of naval architecture & ocean engineering.