基于邻帧差分近邻反相特征的红外运动点目标检测算法

基于运动点目标在邻帧差分图像中所具有的近邻反相特征,即运动点目标的两个位置相邻近、灰度值一正一负,提出一种在复杂背景下,基于红外序列图像的运动点目标检测算法.本算法利用该特征在邻帧差分图像中检测反相点对,进而构造反相点对矢量图,最后依据累积反相点对矢量图中多矢量首位相接的连续性检测出运动的点目标.文中给出并证明应用本算法能以概率1检测到运动点目标的收敛性定理.对典型复杂背景下10幅1000帧图像的仿真结果表明,当信噪比大于或等于1.5时,可以有效检测出运动点目标.     

一种新的红外背景抑制滤波算法

结合保护带和Dempster-Shafer(D-S)证据理论,提出了一种新的红外背景抑制滤波算法。首先分析了小目标的特性,论证了小目标在局部背景中呈现出极值性的条件和可能性,并在此基础上采用D-S证据理论计算最佳滤波尺寸。然后在最佳滤波尺寸下,根据保护带的思想进行背景抑制滤波,该算法能够适用不同尺寸的小目标。以全局信噪比增益、局部信噪比增益和背景抑制系数作为算法性能评价标准,实验结果表明,在低信噪比的情况下该算法对典型的不同红外背景具有良好的适用性,能够有效地抑制红外背景并增强目标信息。     

Doppler Compensation & Single Pulse Imaging using Adaptive Pulse Compression

The effects of target Doppler are addressed in relation to adaptive receive processing for radar pulse compression. To correct for Doppler-induced filter mismatch over a single pulse, the Doppler-compensated adaptive pulse compression (DC-APC) algorithm is presented whereby the respective Doppler shifts for large target returns are jointly estimated with the illuminated range profile and subsequently incorporated into the original APC adaptive receive filter formulation. As a result, the Doppler-mismatch-induced range sidelobes can be suppressed thereby regaining a significant portion of the sensitivity improvement that is possible when applying adaptive pulse compression (APC) without the existence of significant Doppler mismatch. In contrast, instead of compensating for Doppler mismatch, the single pulse imaging (SPI) algorithm generalizes the APC formulation for a bank of Doppler-shifted matched filters thereby producing a sidelobe-suppressed range-Doppler image from the return signal of a single radar pulse which is applicable for targets with substantial variation in Doppler. Both techniques are based on the recently proposed APC algorithm and its generalization, the multistatic adaptive pulse compression (MAPC) algorithm, which have been shown to be effective for the suppression of pulse compression range sidelobes thus dramatically increasing the sensitivity of pulse compression radar.     

Collaborative image compression and classification with multi-task learning for visual Internet of Things

Widespread deployment of the Internet of Things(Io T) has changed the way that network services are developed, deployed, and operated. Most onboard advanced Io T devices are equipped with visual sensors that form the so-called visual Io T. Typically, the sender would compress images, and then through the communication network, the receiver would decode images, and then analyze the images for applications. However, image compression and semantic inference are generally conducted separately, and t...     

基于DSP和FPGA的视频编码器协同设计与算法优化实现

采用DSP和FPGA协同技术设计实现了一个高性能的MPEG-4视频编码器。FPGA模块完成视频采集、YUV分离、数据I/O等功能,而使用DSP专一进行视频压缩编码。针对DSP片内资源特点设计了片内存储器数据分配方案,并根据该方案优化了MPEG-4视频压缩的数据流模式。提出了基于宏块空间复杂度的宏块类型判断算法,有效地降低了视频压缩算法的计算复杂度。测试结果表明,采用MPEG-4视频标准该视频编码器每秒能够压缩39.2帧CIF图像。     

融合背景图像信息和多特征压缩的相关滤波跟踪算法

针对相关滤波类跟踪算法目标背景图像信息利用率较低、目标特征表达能力较弱的问题,提出了一种融合背景图像信息的多特征压缩跟踪算法。首先,在上下文感知滤波器的基础上,将背景图像信息加入位置滤波器。其次,提取颜色名(Color Name, CN)特征与梯度直方图(Histogram of Oriented Gradient, HOG)特征,使用最大响应因子及平均峰相关能量(Average Peak-to-Correlation Energy, APCE)评估跟踪结果的可信度,实现两种特征的自适应融合。最后,利用特征降维简化模型的复杂度,实现算法运行速度的提升。实验结果表明,改进后的算法在遮挡、形变、尺度变化等复杂环境下均具有较高的鲁棒性,其跟踪精度和成功率指标均优于DSST及其他主流的跟踪算法,并且仍保持了实时性。     

Efficient target tracking using dynamic programming

A dynamic programming (DP) algorithm has been developed for the detection and tracking of subpixel-sized, low signal-to-noise ratio (SNR) targets observed by side-or forward-looking imaging sensors. A distinguishing feature of this approach is that target detection and tracking are combined into a single optimization procedure that takes into account statistical models of target motion, background noise, and clutter. Current work has led to a number of technical innovations that improve the performance and efficiency of the DP tracking algorithm, including the development of a new track scoring function, and an extension to the basic DP algorithm that reduces computation requirements by over an order of magnitude. A prototype infrared (IR) target tracking system incorporating these enhancements has been implemented for a step-starting IR camera application. Sensitivity improvements of several decibels over conventional sequential detection and tracking algorithms were realized     

基于约束序贯M估计的时空域融合红外杂波抑制

针对红外弱小目标检测中的强背景杂波干扰抑制问题进行研究,提出了一种基于参数约束序贯M估计的时空域融合自适应杂波抑制算法.该算法首先在分析序列图像帧间失配的基础上建立了一种改进的时空域融合背景预测模型,结合二维离散傅里叶快速变换图像配准和双线性插值方法进行灰度值估计;然后,基于约束序贯M估计方法进行模型参数的自适应估计,...     

Method for searching bridge in IR images

A method for searching bridges in IR images is presented in this paper. It is a time-consuming image understanding task to recognize targets in low SCR images. As in IR images, the bridges usually have almost the same gray level range of the background, the typical segmentation algorithm cannot get correct segmentation labels and edge sites of the bridges. The gray level characters of both the lands, which connect the bridge, and the river are fully considered resulting in a method that can be realized effortlessly and quickly. The criteria are also presented to decide whether there exist bridges or not in the given images, so the method endows the system with the ability to seize the target automatically. Some experimental results are presented demonstrating that the method can effectively recognize bridges in IR images     

受限资源下制导武器末制导机器视觉技术研究

针对精确制导武器末制导机器视觉技术应用需求,研究了基于卷积神经网络的、针对复杂背景及小目标的自主目标检测识别算法,并分别进行了网络性能评估和硬件资源需求定量评估。针对最优算法,提出了基于嵌入式受限资源下的高精度神经网络压缩算法,并对算法进行了普适性评估。基于GPU嵌入式平台,实现TensorRT路线网络优化,并在速度和精度两方面均衡考虑下,对裁剪与量化算法进行了详细实验验证。实验结果表明,高精度神经网络压缩算法在硬件资源受限条件下,可以有效提升推理速度,最终经算法优化后的网络结构,可以获得3倍以上的速度提升,网络精度损失小于5%。     

Improving performance in pulse radar detection using neuralnetworks

A new approach using a multilayered feed forward neural network for pulse compression is presented. The 13 element Barker code was used as the signal code. In training this network, the extended Kalman filtering (EKF)-based learning algorithm which has faster convergence speed than the conventional backpropagation (BP) algorithm was used. This approach has yielded output peak signal to sidelobe ratios which are much superior to those obtained with the BP algorithm. Further, for use of this neural network for real time processing, parallel implementation of the EKF-based learning algorithm is indispensable. Therefore, parallel implementation has also been developed     

DGPS/IMU integration-based geolocation system: Airborne experimental test results

This paper addresses differential global positioning system (DGPS)/inertial measurement unit (IMU) integration-based geolocation system developed for airborne remote sensing cameras. First, we provide a brief review on sensor calibration, alignment and sensor fusion as background material of this research. After presenting those background material, as a main part of this paper we present a geolocation algorithm designed for an airborne imaging system. The geolocation system developed is tested through actual airborne experiments. For the verification of the geolocation system developed, we compare initial stationary states of the airplane before-taking off with states after-landing. From the actual test results, we find that it is critical to do an accurate time synchronization between IMU, DGPS, and airborne images, and to compensate for the data delay occurred during the network transfer.     

Superresolution HRR ATR with high definition vector imaging

A new 1-D template-based automatic target recognition (ATR) algorithm is developed and tested on high range resolution (HRR) profiles formed from synthetic aperture radar (SAR) images of targets taken from the Moving and Stationary Target Acquisition and Recognition (MSTAR) data set. In this work, a superresolution technique known as High Definition Vector Imaging (HDVI) is applied to the HRR profiles before the profiles are passed through ATR classification. The new I-D ATR system using HDVI demonstrates significantly improved target recognition compared with previous I-D ATR systems that use conventional image processing techniques. This improvement in target recognition is quantified by improvement in probability of correct classification (PCC). More importantly, the application of HDVI to HRR profiles helps to maintain the same ATR performance with reduced radar resource requirements     

基于加强模糊聚类的航空行李图像超像素分割

在自助行李托运系统拍摄的行李图像库中检索错误运输的行李时,图像背景会严重影响检索精度。针对该问题提出了一种加强模糊聚类算法（EnFCM）的超像素分割方法,实现了行李目标区域的提取。通过多尺度形态学重建梯度图像,设计了自适应上限尺度的分水岭超像素预分割算法,得到多个独立的超像素区域。对超像素图像进行直方图统计,并结合分水岭分割参数和实际行李图像内容的类别数量进行超像素的加强模糊聚类,得到行李区域。通过多个实际行李图像的分割实验验证了算法的有效性,平均分割精度达到93%,超过多个典型的分割算法。     

一种基于Curvelet变换的红外图像去噪方法

针对红外图像中目标与背景的对比度差、边缘模糊、纹理信息缺失等特点,提出了一种基于Curvelet变换的红外图像去噪方法。算法原理简单,易于实现,具有较高的实时性。试验结果表明,算法能够有效去除红外图像中的噪声和杂波。     

Super resolution feature extraction of moving targets

High range resolution (HRR) moving target indicator (MTI) is becoming increasingly important for many military and civilian applications such as the detection and classification of moving targets in strong clutter background. We consider the problem of extracting the HRR features of moving targets with very closely spaced scatterers in the presence of strong stationary clutter, where the range migration and Doppler frequency are taken into account. A relaxation-based algorithm, which is robust and computationally simple, is proposed to deal with the above problem. Numerical results have shown that the proposed algorithm exhibits super resolution and excellent estimation performance     

Noise effects on centroid tracker aim point estimation

An investigation of errors due to noise in centroid tracker aim-point estimation is presented. The centroid tracker discussed is similar to the tracker described by A.L. Gilbert et al. (1980). Simplifications to this algorithm were made so that the derived models would be consistent with the actual tracker algorithm. Two statistical models are derived which relate image noise effects to computation of the target centroid. The first model, the simplified aim-point error model, is derived by assuming that the probabilities of incorrectly classifying target and background pixels are equal. The second model, the extended aim-point error model, is derived by assuming that the probability of incorrectly classifying a target pixel can differ from the probability of incorrectly classifying a background pixel. These models are described and their mathematical implication is discussed. Simulation results which verify the models are presented     

High resolution reconstruction of ISAR images

A 2-D high-resolution spectral analysis algorithm with application to an inverse synthetic aperture radar (ISAR) imaging, is presented. This algorithm is based on two-dimensional linear prediction using autoregressive (AR) coefficients. Stability is guaranteed by AR process pole adjustment. An ISAR target is modeled for a complex scatterer geometry. Computer simulation results are provided for the high-resolution reconstruction of ISAR images     

一种新的目标跟踪特征融合方法

在粒子滤波目标跟踪框架下,给出一种基于颜色直方图和边缘直方图的新特征融合算法．该算法是一种通过颜色和边缘直方图的特征波和熵实现特征之间粒子数重分配的方法。通过实验比较本文算法与单特征跟踪算法的跟踪误差,结果表明本文方法较传统依靠单一特征进行目标跟踪的粒子滤波方法,在运算时间增幅不明显的情况下,实现了更为准确的目标跟踪,并可稳定的应用于运动背景下的动目标跟踪、背景光照变化下的多目标跟踪以及多目标发生相互遮挡等情况。     

SAR image formation via semiparametric spectral estimation

A new algorithm, referred to as the SPAR (Semiparametric) algorithm, is presented herein for target feature extraction and complex image formation via synthetic aperture radar (SAR). The algorithm is based on a flexible data model that models each target scatterer as a two-dimensional (2-D) complex sinusoid with arbitrary unknown amplitude and constant phase in cross-range and with constant amplitude and phase in range. By attempting to deal with one corner reflector, such as one dihedral or trihedral, at a time, the algorithm can be used to effectively mitigate the artifacts in the SAR images due to the flexible data model. Another advantage of SPAR is that it can be used to obtain initial conditions needed by other parametric target feature extraction methods to reduce the total amount of computations needed. Both numerical and experimental examples are provided to demonstrate the performance of the proposed algorithm