Semisupervised heterogeneous ensemble for ship target discrimination in synthetic aperture radar images

Ship detection using synthetic aperture radar (SAR) plays an important role in marine applications. The existing methods are capable of quickly obtaining many candidate targets, but numerous non-ship objects may be wrongly detected in complex backgrounds. These non-ship false alarms can be excluded by training discriminators, and the desired accuracy is obtained with enough verified samples. However, the reliable verification of targets in large-scene SAR images still inevitably requires manual interpretation, which is difficult and time consuming. To address this issue, a semisupervised heterogeneous ensemble ship target discrimination method based on a tri-training scheme is proposed to take advantage of the plentiful candidate targets. Specifically, various features commonly used in SAR image target discrimination are extracted, and several acknowledged classification models and their classic variants are investigated. Multiple discriminators are constructed by dividing these features into different groups and pairing them with each model. Then, the performance of all the discriminators is tested, and better discriminators are selected for implementing the semisupervised training process. These strategies enhance the diversity and reliability of the discriminators, and their heterogeneous ensemble makes more correct judgments on candidate targets, which facilitates further positive training. Experimental results demonstrate that the proposed method outperforms traditional tri-training.     

Enhanced Kalman Filter Algorithm Using the Invariance Principle

Target tracking in multistatic active sonar systems is often limited in shallow-water environments due to the high level of bottom reverberation that produces false detections. Past research has shown that these false alarms may be mitigated when complete knowledge of the environment is available for discrimination, but these methods are not robust to environmental uncertainty. Recent work has demonstrated the existence of a waveguide invariant for active sonar geometries. Since this parameter is independent of specifics of the environment, it may be used when the environment is poorly known. In this paper, the invariance extended Kalman filter (IEKF) is proposed as a new tracking algorithm that incorporates dynamic frequency information in the state vector and uses the invariance relation to improve tracker discrimination. IEKF performance is quantified with both simulated and experimental sonar data and results show that the IEKF tracks the target better than the conventional extended Kalman filter (CEKF) in the presence of false detections.      

基于合成孔径雷达交叉极化通道数据的海上目标探测

Azimuth ambiguities(ghost targets) discrimination is of great interest with the development of a synthetic aperture radar(SAR). And the azimuth ambiguities are often mistaken as actual targets and cause false alarms. For actual targets, HV channel signals acquired by a fully polarimetric SAR are approximately equal to a VH channel in magnitude and phase, i.e., the reciprocity theorem applies, but shifted in phase about ± for the first-order azimuth ambiguities. Exploiting this physical behavior, the real part of the product of the two cross-polarized channels, i.e.(S S)HV VH, hereafter called A12 r, is employed as a new parameter for a target detection at sea. Compared with other parameters, the contrast of A12 r image between a target and the surrounding sea surface will be obviously increased when A12 r image is processed by mean filtering algorithm. Here, in order to detect target with constant false-alarm rates(CFARs), an analytical expression for the probability density function(pdf) of A12 r is derived based on the complex Wishart-distribution. Because a value of A12 r is greater/less than 0 for real target/its azimuth ambiguities, the first-order azimuth ambiguities can be completely removed by this A12r-based CFAR technology. Experiments accomplished over C-band RADARSAT-2 fully polarimetric imageries confirm the validity.     

Morphological and statistical approaches to improve detection in the presence of reverberation

The detection of a target echo in a sonar image is usually a difficult task since the reverberation, consisting of a large number of spurious echoes, generates a lot of false alarms. In this paper, we propose two new detectors derived from image processing algorithms. These detectors are respectively based on a morphological and a statistical contrast. Each detector only requires setting a few parameters. This setting is done using some prior knowledge about the data (shape of the emitted signal and the used antenna, characteristics of the reverberation). Nevertheless, an extensive statistical study of the detection performances proves that the proposed methods are robust and that even an imprecise setting of the parameters leads to satisfactory results. Applied to the real data, these detectors and their sequential combination lead to a significant improvement on the performances: The false alarm rate is drastically reduced while the detection probability is preserved. Based on different contrasts, these detectors have complementary behaviors. Therefore, a further improvement is achieved by a fusion of the different results to classify the remaining echoes as whether spurious or true detection.     

磁法延拓在地磁背景场匹配中的应用研究

在浅海地区,由于复杂的地磁背景场会对航空磁异常探测(MAD)造成影响,导致目标识别的虚警率过高。为了降低虚警率,需要将地磁背景场从飞机探测的磁场中剥离出来。文中使用的地磁背景场通过船磁测量获得。但是MAD和船磁测量不在同一个高度/平面内,需要通过磁法向上延拓的方法,将船测磁场向上延拓到飞机飞行高度。基于海上磁测数据,通过磁法向上延拓的方法,获得飞机飞行平面内的地磁背景场,然后将实际飞行的磁测数据减去地磁背景场来完成地磁匹配,最终达到降低MAD信号识别输出虚警的目的。     

数学形态学方法在海面船目标检测中的应用

为进一步提高检测精度，提出1种将数学形态学方法应用到船目标检测数据后处理中的算法，根据该算法设计的数学形态学滤波器可以有效地消除SAR图像斑点噪声在船目标检测中造成的影响，减少船目标检测的误判，提高参数估计准确性。     

Signal excess in K-distributed reverberation

Active sonar systems have recently been developed using larger arrays and broad-band sources to counter the detrimental effects of reverberation in shallow-water operational areas. Increasing array size and transmit waveform bandwidth improve the signal-to-noise ratio-and-reverberation power ratio (SNR) after matched filtering and beamforming by reducing the size of the range-bearing resolution cell and, thus, decreasing reverberation power levels. This can also have the adverse effect of increasing the tails of the probability density function (pdf) of the reverberation envelope, resulting in an increase in the probability of a false alarm. Using a recently developed model relating the number of scatterers in a resolution cell to a K-distributed reverberation envelope, the effect of increasing bandwidth (i.e., reducing the resolution cell size) on detection performance is examined for additive nonfluctuating and fluctuating target models. The probability of detection for the two target models is seen to be well approximated by that for a shifted gamma variate with matching moments. The approximations are then used to obtain the SNR required to meet a probability of detection and false-alarm performance specification (i.e., the detection threshold). The required SNR is then used to determine that, as long as the target and scatterers are not over-resolved, decreasing the size of the resolution cell always results in an improvement in performance. Thus, the increase in SNR obtained by increasing bandwidth outweighs the accompanying increase in false alarms resulting from heavier reverberation distribution tails for K-distributed reverberation. The amount of improvement is then quantified by the signal excess, which is seen to be as low as one decibel per doubling of bandwidth when the reverberation is severely non-Rayleigh, as opposed to the expected 3-dB gain when the reverberation is Rayleigh distributed.     

Economic justification of tsunami research: A specific example based on reduction of false alarms in Hawaii

The most significant potential improvement to the Tsunami Warning System, at least as it affects Hawaii, and one of the more important practical justifications of tsunami research, is the reduction in false alarms. There are both immediate and deferred costs of tsunami false alarms. The immediate costs are the costs of responding to tsunami warnings, false or not. The deferred costs are the tsunami casualties resulting from failures to respond to subsequent warnings, insofar as these are attributable to the loss of public confidence in the warning system due to the false alarms. It is estimated that the Hawaiian response to a tsunami false alarm costs about $777,000, and that with present warning system policy the average annual costs of such responses is $264,000. Assigning values to human life and injury, it is estimated that the deferred costs of false alarms average about $42,000 per year, bringing the total annual costs to $306,000. An 80% reduction of these costs would justify an annual research expenditure of $307,000 per year for the next ten years.     

Active sonar detection in shallow water using the Page test

The use of active sonar in shallow water results in received echoes that may be considerably spread in time compared to the resolution of the transmitted waveform. The duration and structure of the spreading and the time of occurrence of the received echo are unknown without accurate knowledge of the environment and a priori information on the location and reflection properties of the target. A sequential detector based on the Page test is proposed for the detection of time-spread active sonar echoes. The detector also provides estimates of the starting and stopping times of the received echo. This signal segmentation is crucial to allow further processing such as more accurate range and bearing localization, depth localization, or classification. The detector is designed to exploit the time spreading of the received echo and is tuned as a function of range to the expected signal-to-noise ratio (SNR) as determined by the transmitted signal power, transmission loss, approximate target strength, and the estimated noise background level. The theoretical false alarm and detection performance of the proposed detector, the standard Page test, and the conventional thresholded matched filter detector are compared as a function of range, echo duration, SNR, and the mismatch between the actual and assumed SNR. The proposed detector and the standard Page test are seen to perform better than the conventional thresholded matched filter detector as soon as the received echo is minimally spread in time. The use of the proposed detector and the standard Page test in active sonar is illustrated with reverberation data containing target-like echoes from geological features, where it was seen that the proposed detector was able to suppress reverberation generated false alarms that were detected by the standard Page test     

基于虚警函数的侧扫声纳水下目标实时检测方法

研究了侧扫声纳系统进行水下目标探测过程中目标信号的检测问题。通过分析海底回波信号的统计模型及其参数的估计,讨论了目标信号对统计模型拟合的影响规律,提出了侧扫声纳回波信号虚警函数和虚警率的概念,及其对Ping信号中目标信号的检测方法。算例结果表明,回波信号的三种分布模型中K分布拟合程度最优,在相同虚警率的条件下,基于K分布的虚警函数目标检测率最高。该法可为侧扫声纳回波信号中目标的实时报警提供技术支撑。     

Detection of mines in acoustic images using higher order spectral features

A new pattern-recognition algorithm detects approximately 90% of the mines hidden in the Coastal Systems Station Sonar0, 1, and 3 databases of cluttered acoustic images, with about 10% false alarms. Similar to other approaches, the algorithm presented here includes processing the images with an adaptive Wiener filter (the degree of smoothing depends on the signal strength in a local neighborhood) to remove noise without destroying the structural information in the mine shapes, followed by a two-dimensional FIR filter designed to suppress noise and clutter, while enhancing the target signature. A double peak pattern is produced as the FIR filter passes over mine highlight and shadow regions. Although the location, size, and orientation of this pattern within a region of the image can vary, features derived from higher order spectra (HOS) are invariant to translation, rotation, and scaling, while capturing the spatial correlations of mine-like objects. Classification accuracy is improved by combining features based on geometrical properties of the filter output with features based on HOS. The highest accuracy is obtained by fusing classification based on bispectral features with classification based on trispectral features.     

基于AIS距离-多普勒投影的地波雷达CFAR检测验证方法

恒虚警(CFAR)检测是地波雷达船只目标探测的主要方法。目前基于船舶自动识别系统(AIS)信息的CFAR检测验证方法是一种间接验证方式,容易受地波雷达系统测向误差的影响,且不具备对错检与漏检目标的分析能力。本文提出了一种基于AIS信息的评价地波雷达CFAR检测结果的直接验证方法。该方法将有效的AIS信息转换到地波雷达的距离-多普勒谱中,通过在该谱中AIS信息与CFAR检测结果的关联分析,实现CFAR检测结果的直接评价。论文首先给出了方法的原理和处理流程,然后利用实测数据给出了该方法在CFAR检测结果评价中的实际应用,验证了方法有效性,而且该方法也为低可观测目标的CFAR检测提供了参数优化调整的依据。     

Validation of AltiKa Matching Pursuit Rain Flag

The major drawback of Ka band, operating frequency of the AltiKa altimeter on board SARAL, is its sensitivity to atmospheric liquid water. Even light rain or heavy clouds can strongly attenuate the signal and distort the signal leading to erroneous geophysical parameters estimates. A good detection of the samples affected by atmospheric liquid water is crucial. As AltiKa operates at a single frequency, a new technique based on the detection by a Matching Pursuit algorithm of short scale variations of the slope of the echo waveform plateau has been developed and implemented prelaunch in the ground segment. As the parameterization of the detection algorithm was defined using Jason-1 data, the parameters were re-estimated during the cal-val phase, during which the algorithm was also updated. The measured sensor signal-to-noise ratio is significantly better than planned, the data loss due to attenuation by rain is significantly smaller than expected (<0.1%). For cycles 2 to 9, the flag detects about 9% of 1Hz data, 5.5% as rainy and 3.5 % as backscatter bloom (or sigma0 bloom). The results of the flagging process are compared to independent rain data from microwave radiometers to evaluate its performances in term of detection and false alarms.     

基于海洋环境的AUG探测应用方法探讨

水下滑翔机(AUG)是正在开发的 UUV 品种之一,其在海洋中作为探测器运用是一个技术方法问题。 针对未来将出现的大携载能力环境探测滑翔机,在分析海洋环境和被探目标以及 AUG 技术特性基础上,预先探讨了未来应用方式及试验方法,研究了基于海洋与目标环境的区域航路导引、目标预示区、隐蔽探测、卫星通讯等应用方法,进行了典型状态试验方法仿真分析,可为相关试验研究提供参考。     

Underwater target detection using multichannel subband adaptivefiltering and high-order correlation schemes

In this paper, new pre- and post-processing schemes are developed to process shallow-water sonar data to improve the accuracy of target detection. A multichannel subband adaptive filtering is applied to preprocess the data in order to isolate the potential target returns from the acoustic backscattered signals and improve the signal-to-reverberation ratio. This is done by estimating the time delays associated with the reflections in different subbands. The preprocessed results are then beamformed to generate an image for each ping of the sonar. The testing results on both the simulated and real data revealed the efficiency of this scheme in time-delay estimation and its capability in removing most of the competing reverberations and noise. To improve detection rate while significantly minimizing the incident of false detections, a high-order correlation (HOC) method for postprocessing the beamformed images is then developed. This method determines the consistency in occurrence of the target returns in several consecutive pings. The application of the HOC process to the real beamformed sonar data showed the ability of this method for removing the clutter and at the same time boosting the target returns in several consecutive pings. The algorithm is simple, fast, and easy to implement     

融合scSE模块的改进Mask R-CNN海洋锋检测方法

海洋锋是重要的中尺度海洋现象,具有数据量小、目标小、弱边缘等特性。针对实际检测任务中弱边缘、小目标海洋锋的检测精度低、错检及漏检率高等问题,融合scSE (spatial and channel Squeeze&Excitation)空间注意力模块构建了一种改进的Mask R-CNN海洋锋检测模型。该方法首先对Mask R-CNN骨干网络结构进行改进,采用scSE模块引导的ResNet-50网络作为特征提取网络,通过加权策略对图像通道和空间位置进行特征突出,提升网络对重要特征的提取能力;其次,针对海洋锋目标边缘定位不准确的问题,引入IoU boundary loss构建新的Mask损失函数,提高边界检测精度。最后,为验证方法的有效性,从训练数据和实验模型上,分别设计多组对比实验。实验结果表明,相比传统Mask R-CNN、YOLOv3神经网络及现有Mask R-CNN改进网络,本文方法对SST梯度影像数据集上的强、弱海洋锋检测效果最好,定位准确率（IoU,Intersection-over-union））及检测精度（mAP,Mean Average Precision）均达0....     

Statistical normalization of spherically invariant non-Gaussian clutter

Conventional detection in active sonar involves comparing the normalized matched filter output power to a fixed preset threshold. Threshold crossings from contacts of interest are labeled as detections and those from undesired clutter echoes as false alarms. To maintain a constant false-alarm rate (CFAR) in the presence of strong transient clutter, the system can either increase the threshold or apply some function that suppresses this background down to an acceptable level. The latter approach leads to a more consistent background on the display, which enables operator-assisted detection. Background clutter suppression should not come at the expense of contact detection; to maximize the probability of detection (PD) for a given probability of false alarm (PFA), the likelihood ratio test (LRT) is used. However, the LRT does not address display issues, since the threshold that achieves a desired PFA varies with the input distribution. Ideally, the LRT output is monotonically transformed using a "statistical normalizer" (SN) that returns a consistent CFAR background without degrading the optimized PD. Within the radar community, clutter suppression is proposed using a LRT tuned to a K-distributed spherically invariant random vector (SIRV) model. However, this model does not lend itself to SN, as a closed-form expression for the LRT output density does not exist. In contrast, the proposed SIRV clutter model, with Pareto distributed power, leads to a closed-form density from which the SN function is readily derived. This combined Pareto-LRT/SN detector nearly matches the optimized PD performance of the K-distributed LRT and maintains a consistent CFAR background for display purposes.     

Fast Double Selectivity Index-CFAR Detection Method for the Multi-Beam Echo Sounder

Abstract

Seafloor terrain and water column target detection is of great significance in marine surveys. However, the complex clutter environment can dramatically affect the detection performance of a multi-beam echo sounder. To simultaneously and robustly detect the seafloor terrains and water column targets, this paper proposes a fast two-dimensional double selectivity index-constant false alarm rate (DSI-CFAR) detection method. A two-dimensional cross sliding window, which includes reference cells and guard cells, is used in this paper. The proposed method improves the detection performance by using the appropriate clutter power level estimation strategies in different directions. An exponential model is accurately built by fitting various distributions, so the DSI-CFAR detection performance is greatly ameliorated. To reduce the computational load of the method, this paper uses a fast algorithm based on the global threshold. The region of interest (ROI) is selected by the preset global threshold, and the DSI-CFAR detector is only performed in the region of interest, which significantly reduces the number of pixels that must be calculated. In this paper, the measured results of different sliding windows and CFAR detection methods validate the basic performance of the proposed method.     

基于YOLOv3深度学习的海雾气象条件下海上船只实时检测

海雾气象条件下船只高精度检测识别面临较大困难,传统的目标识别、定位方法效果差强人意。作者围绕海雾气象条件下不同类型船只的实时检测问题,提出一种基于YOLOv3深度学习的实时海上船只检测新思路。首先构建清晰图片和模糊图片（海雾、雨）的判别方法,实现图片清晰度分类处理;其次为提高海雾气象条件下海上船只的实时检测精度,消除海雾遮挡对目标识别的影响,运用暗通道先验去雾方法对含有海雾的图像实行去雾;最后基于YOLOv3深度学习算法对精细处理后的图像进行船只实时检测。实验结果表明该方法能够在海雾气象条件下高效、准确地检测到船只,对海上复杂环境条件下的船只实时检测研究具有一定的理论指导意义。     

Signal detection for communications in the underwater acousticenvironment

Signal detection is a critical stage in the implementation of any effective communications system. The underwater acoustic environment, particularly in the presence of underwater vehicles, presents significant challenges to reliable detection without excessive false alarms. While there is often sufficient signal-to-noise ratio with respect to stationary broad-band background noise to permit reliable operation, the presence of strong event-like interference signals such as narrow-band signals and impulsive broad-band signals complicates the detection problem significantly. Frequency-hopped signals interleaved with quiescent bands are proposed as the basis of a robust detection system. These signals also make robust detection possible in a multi-access communications system. Two new detection algorithms that exploit the particular structure of these frequency-hopped signals are developed and their performance is analyzed. This analysis uses a modification of the doubly noncentral F-distribution that has not been used previously for such analysis. This distribution makes possible the direct calculation of probabilities of detection and false-alarm under interference and signal scenarios that cannot be analyzed using the better known noncentral F-distribution. With this analysis, the two developed detectors are shown to offer superior performance to that of either the CFAR detector or the binary data sequence detector. Experimental data confirms the theoretically derived results