Identification and velocity computation of multiple moving objectsin images

Analysis for identifying and computing the velocity of multiple moving objects in sequences of images is presented. While previously published analyses addressed sequences with one or two moving objects, this analysis is extended to sequences with multiple moving objects. simulation is performed to demonstrate the applicability of this technique. Multiple moving objects appear in the spectral domain as groups of individual peaks which identify the objects and are used to compute their velocities. Algorithms which evaluate the spectral peaks and calculate their velocity are also presented     

Spatio-temporal pattern recognition using hidden Markov models

A spatio-temporal method for identifying objects contained in an image sequence is presented. The Hidden Markov Model (HMM) technique is used as the classification algorithm, making classification decisions based on a spatio-temporal sequence of observed object features. A five class problem is considered. Classification accuracies of 100% and 99.7%, are obtained for sequences of images generated over two separate regions of viewing positions. HMMs trained on image sequences of the objects moving in opposite directions showed a 98.1% successful classification rate by class and direction of movement. The HMM technique proved robust to image corruption with additive correlated noise and had a higher accuracy than a single-look nearest neighbor method. A real image sequence of one of the objects used was successfully recognized with the HMMs trained on synthetic data. This study shows the temporal changes that observed feature vectors undergo due to object motion hold information that can yield superior classification accuracy when compared with single-frame techniques     

Autofocusing of ISAR images based on entropy minimization

A novel autofocusing technique is developed for random translational motion compensation in inverse synthetic aperture radar (ISAR) imaging of objects. This technique is based on an entropy minimization principle and validated via a nonparametric estimation method. Images of a simulation and a real flying aircraft are used for illustration. Images of encouraging quality confirm the feasibility of autofocusing the radar images by just the requirement of minimizing the image entropy     

边缘检测技术在火箭发动机实验诊断中的应用

对高速运动分析仪拍摄的Ｕ型燃气发生器固体装药燃烧和双组元液体喷嘴雾化图像进行了边缘检测图像分析,根据被分析实物边缘层次性较强的特点,在简单边缘检测的基础上采用自动门限、二次门限和最佳滤波的方法,获得了固体燃气发生器在燃烧过程中的内部结构、推进剂瞬态燃速、绝热隔板烧蚀率等参数,获得了双组元液体喷嘴的雾化角、粒子分布等特性。     

Model-based orientation-independent 3-D machine vision techniques

Orientation-dependent techniques for the identification of a three-dimensional (3-D) object by a machine vision system are represented in parts. In the first part, the data consist of intensity images of polyhedral objects obtained by a single camera, while in the second part, the data consist of range images of curved objects obtained by a laser scanner. In both cases, the attributed graph representation of the 3-D object surface is used to drive the respective algorithm. In this representation, a graph node represents a surface path and a link, the adjacency between two patches. The attributes assigned to nodes are moment invariants of the corresponding face for polyhedral objects. For range images, the Gaussian curvature is used as a segmentation criterion for providing symbolic shape attributes. Identification is achieved by an efficient graph-matching algorithm used to match the graph obtained from the data to a subgraph of one of the model graphs stored in the computer memory     

Ultra-wideband ground-penetrating radar for the detection of buriedmetallic mines

The suitability of ultra-wideband ground-penetrating radar as a tool for the detection of buried metallic mines is explored in this paper. The analysis centers around a 200-800 MHz, dual-polarized ground penetrating radar (GPR) designed and built by SRI International. The analysis consisted of fusing the images from the dual polarizations into a single image to enhance the target objects and suppress clutter. Results are shown for several variations of a Mahalanobis-based fusion technique, and “soft decision” minefield detection results based upon Monte Carlo statistical techniques are also presented. Although relatively few scenes were analyzed, these results show that the dual-polarized GPR is potentially very effective at finding buried mines and minefields     

基于改进HOG特征的空间非合作目标检测

传统的非合作目标检测方法大都基于一定的匹配模板,这不仅需要预先指定先验信息,进而设计合适的检测模板,而且同一模板只能对具有相似形状的目标进行检测,不易直接用于检测形状未知的非合作目标。为降低检测过程中对目标形状等先验信息的要求,借鉴基于规范化梯度的物体区域估计方法,提出一种基于改进方向梯度直方图特征的目标检测方法,首先构建包含有自然图像和目标图像的训练数据集;然后提取标记区域的改进方向梯度直方图特征,以更好地保持局部特征的结构性,并根据级联支持向量机训练模型,从数据集中自动学习目标物体的判别特征;最后,将训练后的模型用于检测测试集图像中的目标。实验结果表明,算法在由4953幅和100幅图像构成的测试集中分别取得94.5%和94.2%的检测率,平均每幅图像的检测时间约为0.031 s,具有较低的时间开销,且对目标的旋转及光照变化具有一定的鲁棒性。     

An efficient 3-D near-field ISAR algorithm

This paper presents a new three-dimensional (3-D) near-field inverse synthetic aperture radar (ISAR) imaging technique. A 3-D ISAR image can be obtained by processing coherently the backscattered fields as a function of the frequency and two rotation angles about axes which are mutually orthogonal. Most of the existing ISAR algorithms are based on the Fourier transform and as such can tolerate only small amounts of wavefront curvature. Wavefront curvature must be taken into account when imaging an object in the near-field. Near-field ISAR imaging of large objects using a direct Fourier inversion may result in images which are increasingly unfocused at points which are more distant from the center of rotation. An algorithm based on an azimuth convolution between a near-field focusing function and the frequency domain backscattered fields is discussed. This convolution is efficiently implemented by using fast Fourier transform (FFT) techniques. Furthermore, in order to further alleviate the computational load of the algorithm, the discrete Fourier transform (DFT) of the focusing function is evaluated by means of the stationary phase method. Experimental results show that this technique is precise and virtually impulse invariant     

一种三维物体的视觉识别方法研究

研究了一种利用二维图像识别三维物体的方法。通过颜色特征提取物体的有效区域．使用链码描述了区域中三维物体的形状特征，并形成物体的形状特征向量。将该特征向量作为所设计的BP神经网络的输入进行物体匹配识别，网络输出为属于某种物体的隶属度。通过在各种光照环境下不同三维物体的二维彩色图像进行实验，证明了所提出方法的有效性。     

Temperature-Driven Segmentation for Autonomous Anti-Tank Weapons

In this paper we discuss a simple temperature-driven segmentation technique that can be used for autonomous terminal guidance of anti-tank munitions. A ratio between two images from different IR bands is taken to form an emissivity independent image that is a function of temperature only. Size and temperature discrimination can then be used to find hot spots of interest. Candidate targets are segmented from the image about the points of interest using an.algorithm that finds all regions in multi-region objects and does not require a priori intensity information. Size and gross shape features are used to determine if the segmented objects are to be classified as targets. The key to this approach is the initial hot spot extraction. The temperature dependent ratio image allows quick and confident screening of the entire field of view, so that only the regions of interest require further processing. The described techniques effectively found tanks in the presence of common battlefield clutter when used with synthetic IR imagery provided by Georgia Tech Research Institute. The simplicity, effectiveness and potential speed of this technique make it ideal for autonomous guidance of expendable anti-tank weapons, especially where only low resolution IR imagery is available.     

FOPEN SAR imaging using UWB step-frequency and random noisewaveforms

The detection and identification of targets obscured by foliage have been topics of great interest. Several synthetic aperture radar (SAR) experiments have demonstrated promising images of terrain and man-made objects obscured by dense foliage, by using either linear frequency modulation (LFM) or step-frequency waveforms. We present here the methodology and results of a comparative study on foliage penetration (FOPEN) SAR imaging using ultrawideband (UWB) step-frequency and random noise waveforms. A statistical-physical foliage transmission model is developed for simulation applications. The foliage obscuring pattern is analyzed by means of the technique of paired echoes. The results of the comparative study demonstrates the ability of a UHF band UWB random noise radar to be used as a FOPEN SAR. Advantages of the random noise radar system include covert detection and immunity to radio frequency interference (RFI)     

Object-based SAR image compression using vector quantization

A simple and elegant algorithm is presented to encode images with rich content, which allows easy access to various objects. An object-plane-based encoding method for compression of synthetic aperture radar (SAR) imagery is developed, with different object planes for target classes and background. A variable-rate residual vector quantization (VQ) algorithm is developed to encode the background information. This algorithm is very powerful as indicated by the experimental results. The proposed coding scheme allows compression matched to the final application of the images, which in this case is target recognition and classification.     

复杂条件下的运动目标检测方法研究综述

运动目标检测是当前研究热点之一,被广泛地应用于计算机视觉、视频处理等领域。将运动目标检测的三种常用方法进行对比,总结其各自的适用性及局限性。针对光照变化、背景存在干扰运动、前景色与背景色接近等复杂条件,综述了相应的多种解决方案,并指出了方案算法的局限性,对复杂条件下运动目标检测方法的发展方向进行展望,从而为相关工作的进一步研究指出了方向。     

夜间成像光测数字图像增强技术研究

分析夜间成像的光测数字图像灰度范围、对比度和信息量等主要特点,为图像增强的策略选择和效果评价提供依据。针对夜间成像光测数字图像灰度范围较窄等特点,提出局部灰度修正的增强方法适当增强图像对比度,又避免细节损失,同时,利用拉普拉斯锐化突出目标轮廓和边缘,并利用人眼对彩色敏感的特点,采用伪彩色增强提高对图像中模糊目标的分辨能力。通过上述3种增强技术,提高了对夜间成像暗图像的目标识别能力,有利于后续判读处理。     

电大尺寸散射体的RCS计算方法研究

为了更加有效的求解电大目标散射计算问题,在避免谐振区效应情况下获得复杂目标的隐身雷达散射截面(RCS)特性,引用混合场积分方程(CFIE),在快速算法的求解迭代过程中采用共轭梯度算法(CG)的收敛技术,能够稳定的求解电大尺寸的RCS。计算和分析了金属球双站RCS和金属立方体双站RCS,并和精确计算、相关参考文献进行了比较,证明了方法的正确性,计算结果稳定,在工程实际上有较大的应用价值。     

A Technique of Object Detection Using Information about Changes in the Terrain Image

In this paper, detection of objects in airborne systems of unmanned aerial vehicles using information about changes in the image of the terrain is considered. A proposed technique includes initial detection of objects by the level of local contrast over the field of the image misalignment function, calculation of the amplitude characteristics of objects, and their further filtering by a set of calculated values of characteristics.     

Multiassignment for tracking a large number of overlapping objects[and application to fibroblast cells]

In this paper we present a new technique for data association using multiassignment for tracking a large number of closely spaced (and overlapping) objects. The algorithm is illustrated on a biomedical problem, namely the tracking of a group of fibroblast (tissue) cells from an image sequence, which motivated this work. Because of their proximity to one another and due to the difficulties in segmenting the images accurately from a poor-quality image sequence, the cells are effectively closely spaced objects (CSOs). The algorithm presents a novel dichotomous, iterated approach to multiassignment using successive one-to-one assignments of decreasing size with modified costs. The cost functions, which are adjusted depending on the “depth” of the current assignment level and on the tracking results, are derived. The resulting assignments are used to form, maintain and terminate tracks with a modified version of the probabilistic data association (PDA) filter, which can handle the contention for a single measurement among multiple tracks in addition to the association of multiple measurements to a single track. Estimation results are given and compared with those of the standard 2D one-to-one assignment algorithm. It is shown that iterated multiassignment results in superior measurement-to-track association. The algorithms presented can be used for other general tracking problems, including dense air traffic surveillance and control     

MTI data clustering and formation recognition

Image exploitation technology approaches have generally focused on the detection and spatial analysis of stationary groups of objects on the ground using various sensors. While spatial arrangement is clearly necessary in analyzing military formations, it is usually not sufficient. Typically the arrangement must be examined within some context in order to interpret a pattern of deployment. For moving objects the spatial arrangement of the group relative to the direction of motion is key to recognizing the formation. By examining ground moving target indicator (MTI) radar data over time, motion can be inferred and used to establish a context for interpreting the spatial arrangement of the data. New techniques that exploit the multitemporal nature of MTI data are described. The first is a space-time clustering technique that locates compact groups of objects that persist in time. The technique Is an application of Marr and Hildreth's edge detection methodology to the dual problem of region segmentation, or more accurately, volumetric segmentation of space-time. The second technique is based on the use of the Hough transform for recognizing moving formations such as columns, wedges, and lines abreast by analyzing the shape of clustered MTI detections (specifically the orientation of linear arrangements within the group) with respect to their direction of motion. Preliminary results from simulated MTI data sets are presented     

单天基平台测向的空间非合作目标跟踪

针对仅含角度测量信息的单个天基平台可观测性较弱的问题,提出了一种含脉冲机动检测的空间非合作目标跟踪算法,并设计了非合作目标实时跟踪数据处理流程.该算法利用抗差估计技术和UKF（Unscented Kalman Filter,无迹卡尔曼滤波）算法构造目标跟踪滤波器,并综合残差多项式拟合和新息分布特征等方法实现目标机动检测,在天基平台观测信息类型有限和观测几何较差的情况下,可以同时排除孤立野值和成片测量野值的影响,实现非合作机动目标的连续稳定跟踪.数值实验验证了算法的可行性和有效性,也表明了跟踪精度和可靠性与测量精度密切相关.