Optimum Filters for Image Registration

Two-dimensional cross-correlation techniques are applied to the problem of image registration under the assumption of small geometric distortion. Optimum filter functions for use with arbitrary window functions are derived for two performance measures of interest: peak-to-sidelobe ratio, and mean- square registration error; the latter is examined in terms of the contribution due to distortion and the contribution due to noise. A generalized Lagrange multiplier approach is used to derive at least approximate ate solutions to the case where the images are modeled as random processes and to the case where the images are assumed to be deterministic.     

A Coarse Search Correlation Tracker for Image Registration

A high-speed compact digital correlation tracker which determines the frame-to-frame translational motion between two images is presented. By combining a similarity detection approach with the correlation coefficient in a parallel pipelined architecture, effective tracking rates comparable to conventional fast Fourier transform (FFT) techniques can be achieved. Examples of the tracker's performance using simulated data and image data acquired by an active infrared radar are given.     

Correlation Techniques of Image Registration

An extension to the basic concept of correlation detection as a means of image registration is developed. The technique involves linear spatial preprocessing of the inages to be registered prior to the application of a correlation measure. This preprocessing operation utilizes the spatial correlation within each image and greatly improves the detectability of image misregistration. An analysis of the computational aspects of the algorithm is given. Also, results of a computer simulation to evaluate the technique are given.     

遥感图像的亚像素配准技术研究

高分辨遥感技术在军民领域应用广泛,图像配准技术是高分辨遥感图像的预处理技术。实际的机载和星载遥感成像仪,由于自身成像原理和成像仪平台抖动原因,出现了遥感图像亚像素偏移误差。对亚像素图像配准技术进行系统研究,对高精度亚像素配准算法进行研究分析,并对其算法进行仿真、处理,对仿真的结果进行系统分析,通过误差分析提出配准误差修正算法,修正算法对亚像素配准进行了很好的改进,达到遥感图像预处理的目的。     

Optimum Detection of an Optical Image on a Photoelectric Surface

The detection of an optical image in the presence of uniform background light is based on a likelihood ratio formed of the numbers of photoelectrons emitted from small elements of a photoelectric surface onto which the image is focused. When diffraction is negligible and the surface has unit quantum efficiency, this detector is equipollent with the optimum detector of the image-forming light. Its performance is compared with that of the threshold detector and that of a detector basing its decisions on the total number of photoelectrons from a finite area of the image. The illuminance of the image is postulated to have a Gaussian spatial distribution. All three detectors exhibit nearly the same reliability.     

基于改进微粒群优化算法的互信息医学图像配准

医学图像配准是寻找使两幅图像对应点达到空间位置和解剖结构上一致的过程,对于医学临床科研、诊断等方面具有非常重要的意义。基于互信息的配准方法是目前医学图像配准中无创、自动化程度高且配准精度很高的一种方法,已被广泛应用。但是由于插值赝像导致在其目标函数中存在幅值振荡现象,使用局部最优化搜索有时会终止于局部极值,得到错误的配准参数。提出针对互信息配准方法的特点,使用改进微粒群全局搜索算法,调节该算法的参数以适应不同搜索阶段,保证了最优化搜索的准确性,提高了基于互信息方法配准的成功率。     

景象匹配中的几何透视校正和归一化互相关模板匹配

几何失真是使景象匹配性能下降的一个很重要的因素,基于参考图与失真实时图中四对对应点,本文采用透视变换方法,对失真实时图进行校正。对校正后的实时图,本文采用归一化互相关方法,进行模板匹配。仿真实验表明,这两种方法简单易行,校正与匹配结果精确。     

基于控制线方法的机载SAR和可见光图像匹配应用研究

根据无人机(UAV)景象匹配导航的现实需求,对具有典型人造场景的机载合成孔径雷达(SAR)图像与可见光图像,提出一种基于直线特征的SAR图像与可见光图像配准方法.首先,利用改进的直线段检测(LSD)方法提取图像直线特征;其次,构造控制线并设计了一种基于控制线的图像配准方法;最后,依据仿射变换模型实现了待配准图像的精确自动配准.实验表明,在SAR和可见光图像存在较大灰度差异、旋转和平移的情况下,该算法仍能精确配准图像,且运算时间大幅减少,能够满足一些实时性较强的应用.     

On-Line Vehicle Motion Estimation from Visual Terrain Information Part I: Recursive Image Registration

This paper addresses estimation of motion from the optical flow observed by an airborne down looking electro-optical sensor. The paper is in two parts. Part I addresses the development and analysis of a velocity-to-height ratio estimation algorithm and its principal error characteristics. In part II, it is shown how the information provided by the motion estimator can be integrated with additional on-board sensors to provide a complete autonomous navigation system. Part I as presented here is a summary version of the full length paper.? The algorithm implements recursive registration of successive images by using the gradient of a similarity function between them to control the tracking of their relative shift. The shift estimate provides velocity/height information. Substantial saving in memory and computation as compared to conventional full frame registration is achieved by using only a single line in the TV frame. Stochastic mathematical models for the image, terrain and vehicle velocity perturbations are used in the analysis. The choice of the most appropriate similarity function in the registration algorithm is addressed. Performance analysis indicates very small error variances, as illustrated by numerical examples.     

Optimum Signals for Radar

The research reported herein deals with the general problem of the selection of radar waveforms. The investigation is specifically concerned with the synthesis of radar signals which are optimum in the sense that they are characterized by ambiguity surfaces minimized over certain predetermined regions of the ambiguity plane. The weighted ambiguity surface is utilized as the weighted error criterion. This error criterion is mathematically tractable and pertinent to radar system performance but is not unduly restrictive as some orientation parameters are left unspecified for subsequent cost or penalty function analysis. The signal optimization is approached by variational techniques augmented by equality and inequality constraints, for example, limiting the amount of bandwidth or frequency modulation to be less than some system requirement. Several examples are presented demonstrating the optimization techniques and providing a minimum error for the stated problem. It is shown that for any given type of amplitude modulation of the radar signal, the variance or dispersion of the ambiguity surface is not decreased for any type of phase modulation added. The optimum signal for an elliptical weighting function is derived for several cases. The minimum error is shown to depend upon the constraints and the unspecified orientation parameters and, for one case, on the second moment of the signal.     

Optimum Frequencies for Aircraft Classification

The selection of the optimum features for aircraft classification from radar returns was studied. A constraint was imposed that a minimum number of simultaneous frequencies be used. The results indicate that a high reliability of classification is achievable with only two frequencies, and even a single frequency can provide over 95 percent reliability if both phase and amplitude at two orthogonal polarizations are used. The data set tested involved eight military aircraft whose sizes varied over a three-to-one range.     

一个实用的WIN95启动界面

本文通过用简单的方法修改ＭＳＤＯＳ．ＳＹＳ和重新配置ＡＵＴＯＥＸＥＣ．ＢＡＴ文件。     

Optimum Adaptive Reception for Binary Sequences

The a posteriori probability density function p(?|X1, X2,...,Xk), where the Xi, i=1, 2, ..., K, represent Kvector-valued observations statistically related to the random vector ?, appears in many applications of the methods of statistical inference to problems in pattern recognition and statistical communication theory. In this paper, it is shown that for equally likely binary sequences (M= 2) of anticorrelated patterns for signals observed in additive Gaussian noise, a device that computes p?|X1, X2, XK) can be synthesized from a correlator, a simple instantaneous nonlinearity, and a multiplier. These results are used to derive some equally simple structures for various optimum nonsupervised estimators, pattern recognition machines, and signal detectors.     

Optimum Cross-Coupled Tracker for Pulse-Doppler Radar

Although there is a well-defined relation between range and range rate, in conventional pulse-Doppler tracking radars the range-gate and frequency control loops are usually closed independently. In this paper, the optimum cross- coupled stationary tracking filter is derived, in which the range error signal is used to boost the frequency loop, and vice versa. The filter is derived for a general case of target dynamics and uncorrelated white noise in both the range gate and frequency control channels. It is shown that its tracking performance is superior to that of the conventional uncoupled tracking loop for reliable frequency channel operation. In solving the usually difficult degenerated Riccati matrix equation, a simple solution is obtained by applying the method of completion of the square to matrixes.     

Minimal-Order Models for False-Alarm Calculations on Sliding Windows

A procedure is developed to obtain practical numerical results in connection with the m-out-of-n sliding-window detection problem. This effort was motivated by difficulties with previous approaches involving approximation, Markov models, and Monte Carlo simulation. Generating-function methods were found to be unsatisfactory for window lengths greater than 6 due to their complexity. Instead, a Markov model is described that is then constructively reduced to the minimum number of state variables. The results are derived for binary strings with intersymbol correlation. Computational aids are discussed for obtaining design information, such as quantiles, from the minimal-order Markov models. Numerical results are given comparing the methods of the paper with a "jumping" window approximation for an 8/10 problem.     

考虑半径补偿的三坐标测量数据精确配准方法

叶片模型的配准定位是叶片形状检测分析的一个核心环节,针对涡轮叶片的三坐标测量数据在基于迭代最近点的配准定位中与CAD模型表面点的配准关系不对应的问题,提出一种考虑半径补偿的精确配准方法。在求取配准点集的过程当中考虑到半径补偿,使得测量点集与待配准点集之间的关系相对应,进而获取更高的配准精度。进行基于仿真数据的配准试验及对比,给出了基于实测数据的应用与分析,验证了该方法的精确性与实用性。结果表明,考虑半径补偿的三坐标测量数据精确配准方法是可行的。     

民用航空器注册号识别的鲁棒性方法研究

为了提高空管自动化水平并降低运行成本,提出一种基于计算机图像处理技术的民用航空器注册号识别方法.首先将全局阈值分割算法与形态学分割算法结合,实现对航空器图像中注册号的精确定位.然后采用主成分分析法解决注册号倾斜校正问题,并分别针对粘连和非粘连情况的注册号字符进行有效切割.最后利用基于特征提取的模板匹配算法,完成对注册号字符的准确识别.在此基础上,基于MATLAB搭建实验测试平台对上述方法进行仿真验证.实验结果表明,方法能够针对不同角度拍摄的航空器图像快速、准确地识别其注册号,单幅图像识别耗时在0.5 s左右.相比于其他方法,在图像拍摄方面要求较低,有效性与鲁棒性较高,计算速度较快,可以满足实时监控的需求.     

Asymptotically Optimum Sample Size for Quickest Detection

A method is presented for selecting the asymptotically optimum sample size M for detecting a sudden change in the statistics of an observed process. The test statistic is assumed to be a sum of M consecutive values of some single sample detector and the optimization criterion is to minimize the mean time to detection MD for a given mean time between false alarms MF. It is shown that for large MF and MD the solution can be expressed as a function of the single variable ?MF? (or alternatively ?MD?) where ? is a measure of the signal-to-noise ratio (SNR).     

Optimum Signals for High-Resolution Radio Transmission Systems

This paper reports on progress in signal design that has led to improvedresolution capability in radar and communication systems without theuse of complicated signal-processing techniques.Two approaches to the problem of improving resolution capabilityare made. The first approach emphasizes the need to produce sharplypeaked autocorrelation functions. The optimum signal amplitude infrequency is specified to accomplish this, and the spectral density ofthe deterministic signal is shown to satisfy a homogeneous Wiener-Hopfequation. The second approach emphasizes the need to producelow and flattened cross-correlation functions, in order to distinguishthem (since they correspond to error outputs) from the sharply peakedautocorrelation functions. With the use of stationary phase integration,a detailed method for producing any desired cross-correlationamplitude is presented. In particular, the techniques necessary to producesinusoidally modulated cross-correlation functions are discussed.These tools are applied to a realistic N-signal processing system,and the resulting optimum signals are shown to be amplitude-modulatedchirped sinusoids. Detailed examples for physically justifiablesystem parameters are included.