A Moving IR Point Target Detection Algorithm Based on Reverse Phase Feature of Neighborhood in Difference Between Neighbor Frame Images

An algorithm for detecting moving IR point target in complex background is proposed, which is based on the Reverse Phase Feature of Neighborhood （RPFN） of target in difference between neighbor frame images that two positions of the target in the difference image are near and the gray values of them are close to in absolute value but with inverse sign. Firstly, pairs of points with RPFN are detected in the difference image between neighbor frame images, with which a virtual vector graph is made, and then the moving point target can be detected by the vectors＇ sequence cumulated in vector graphs. In addition, a theorem for the convergence of detection of target contrail by this algorithm is given and proved so as to afford a solid guarantee for practical applications of the algorithm proposed in this paper. Finally, some simulation results with 1000 frames from 10 typical images in complex background show that moving point targets with SNR not lower than 1.5 can be detected effectively.

A High Speed Detection Scheme for Point Targets in a Multitarget Environment

The purpose of this paper is to develop a high speed detection scheme for moving and / or stationary point targets in a multitarget environment as registered in an IR image sequence. An iterative approximate 3-D line searching algorithm based upon the geometric representation of lines (for non-maneuvering targets in space) in a 3-D space is derived. The convergency of the algorithm is proved. An analysis is performed of the theoretical detection performance of the algorithm. The statistical experiment results show high effectiveness and computational efficiency of the algorithm in the case of low SNR. The idea may be employed to satisfy the real-time processing requirement of an IR system.

The combined FSS-SPRT method for detection and tracking of low SNR point targets

For the detection and tracking of dim point targets with SNR 〈 2 dB, the combined SPRT and FSS method is given to accomplish detection in whicb likelihood testing are carried out twice to prune constantly. Firstly, the SPRT is developed aiming at the heuristic segments formed by correlation analysis. In order to avoid missing detection the threshold is chosen much lower. Secondly, by adding samples and choosing the one most similar to the heuristic segment to make state estimation FSS is implemented. This time we choose a higher threshold. Moreover in preprocessing the compound kernel estimation is designed to depress varying background clutter. Multiple experimental sequences validate that the method is more suitable for the dim targets detection and tracking compared with the scheme choosing the higher intensity pixel in tracking. It not only has perfect detection performance but also can greatly enhance tracking performance.

New Approach in Processing of the Infrared Image Sequence for Moving Dim Point Targets Detection

The development of an efficient moving target detection algorithm in IR-image sequence is considered one of the most critical research fields in modern IRST (Infrared Search and Track) systems, especially when dealing with moving dim point targets. In this paper we propose a new approach in processing of the Infrared image sequence for moving dim point targets detection built on the transformation of the IR-image sequence into 4-vectors for each frame in the sequence. The results of testing the proposed approach on a set of frames having a simple single pixel target performing a different motion patterns show the validity of the approach for detecting the motion, with simplicity in calculation and low time consumption.

基于改进PointPillars的3D目标检测算法

针对自动驾驶场景下远距离及遮挡目标识别问题,对PointPillars算法进行了改进。引入了并行的空间注意力和通道注意力机制,增强了目标的位置信息及有用特征通道权重,提高了远距离目标的检测精度。在2D CNN骨干网络中引入了自适应空间特征融合模块,解决了特征拼接的信息丢失问题,提高了遮挡目标的检测精度。基于KITTI数据集在3种不同场景难度下分别对SECOND、PointPillars、改进PointPillars这3种算法进行了定量分析验证,并将改进的PointPillars算法进行可视化分析。定量分析表明,改进PointPillars算法在鸟瞰图模式下目标检测精度最大提升2.75%;在三维模式下目标检测精度最大提升2.93%;在AOS模式下目标检测精度最大提升4.05%,可视化结果表明改进的PointPillars算法能有效检测远距离及遮挡目标。

基于改进PointPillars的点云车辆目标检测

针对自动驾驶领域中,激光雷达获取的点云数据存在稀疏性以及边缘噪点误检等问题,提出一种基于改进PointPil-lars的点云车辆目标检测方法。首先,基于SimAM注意力机制改进体素化特征输入,使得网络特征提取阶段能更加关注关键信息,提高特征学习的全局性。其次,基于卷积块注意力模块(CBAM)改进骨干网络结构,提出全新的轻量化通道注意力模块Tiny-CAM和可变形空间注意力模块Deformable-SAM,构建Multi-CBAM骨干网络,提升网络特征提取及特征融合能力。在KITTI数据集以及非公开车库点云数据集上进行验证,实验结果表明,与原网络相比,改进PointPillars方法具有更高的检测精度,平均检测精度提升2.98%,针对遮挡小于30%的点云车辆目标检测精度提升6.51%,证明了该方法的有效性。

Monocular trajectory intersection method for 3D motion measurement of a point target

This article proposes a monocular trajectory intersection method,a videometrics measurement with a mature theoretical system to solve the 3D motion parameters of a point target.It determines the target’s motion parameters including its 3D trajectory and velocity by intersecting the parametric trajectory of a motion target and series of sight-rays by which a motion camera observes the target,in contrast with the regular intersection method for 3D measurement by which the sight-rays intersect at one point.The method offers an approach to overcome the technical failure of traditional monocular measurements for the 3D motion of a point target and thus extends the application fields of photogrammetry and computer vision.Wide application is expected in passive observations of motion targets on various mobile beds.

Efficient generation of targeted point mutations in the Brassica oleracea var.botrytis genome via a modified CRISPR/Cas9 system

In this study,we used the modified CRISPR/Cas9 system to produce targeted point mutations in cauliflower.Acetolactate synthase(ALS)and Centromere-specific histone H3 variant(CENH3)genes were selected as the base-editing targets and hypocotyls of cauliflower were used as explants.For ALS gene,a C-to-T conversion in the Pro182 codon(CCT)can alter the encoded amino acid,likely resulting in herbicide resistance,and a C-to-T mutation in the Leu133 codon(CTT)in the CENH3 gene may produce a haploid inducer.Results indicated that the transformation efficiency was 1.8%–4.5%and the mutation efficiencies for the ALS and CENH3 genes were approximately 22%and 87%,respectively.The ALS mutant cauliflower showed strong herbicide resistance,with possible immediate implications for broadleaf weed control in cauliflower fields.

Point association analysis of vessel target detection with SAR, HFSWR and AIS

A space-borne synthetic aperture radar （SAR）, a high frequency surface wave radar （HFSWR）, and a ship automatic identification system （AIS） are the main remote sensors for vessel monitoring in a wide range. These three sensors have their own advantages and weaknesses, and they can complement each other in some situations. So it would improve the capability of vessel target detection to use multiple sensors including SAR, HFSWR, and A/S to identify non-cooperative vessel targets from the fusion results. During the fusion process of multiple sensors＇ detection results, point association is one of the key steps, and it can affect the accuracy of the data fusion and the efficiency of a non-cooperative target＇s recognition. This study investigated the point association analyses of vessel target detection under different conditions： space- borne SAR paired with AIS, as well as HFSWR, paired with AIS, and the characteristics of the SAR and the HFSWR and their capability of vessel target detection. Then a point association method of multiple sensors was proposed. Finally, the thresholds selection of key parameters in the points association （including range threshold, radial velocity threshold, and azimuth threshold） were investigated, and their influences on final association results were analyzed.

Application of Adaptive Reduced Sigma Points Unscented Kalman Filter to the Tracking of Maneuvering Target

Based on the principle of statistical linear regression, a set of n ＋ 2 sigma points instead of 2n ＋ 1 sigma points used in the unscented Kalman filter （UKF）, is constructed to approximate the system state. And filter accuracy is second order. Real-time of modified UKF is improved. In order to describe accurately the maneuvering target, the ＂current＂ statistical model is used. And the equation of acceleration error covariance is modified at every sample time of the filter. The modified adaptive UKF is presented for estimating the position, velocity and acceleration of maneuvering target. Monte Carlo simulations show the modified adaptive UKF acquires good performance for tracking position of maneuvering target. The modified adaptive UKF has better computational efficiency than UKF.

远距离和遮挡下三维目标检测算法研究

针对现有三维目标检测算法对存在遮挡及距离较远目标检测效果差的问题,以基于点云的三维目标检测算法(3D object proposal generation and detection from point cloud,PointRCNN)为基础,对网络进行改进,提高三维目标检测精度。对区域生成网络(region proposal network,RPN)获取的提议区域(region of interest,ROI)体素化处理,同时构建不同尺度的区域金字塔来捕获更加广泛的兴趣点;加入点云Transformer模块来增强对网格中心点局部特征的学习;在网络中加入球查询半径预测模块,使得模型可以根据点云密度自适应调整球查询的范围。最后,对所提算法的有效性进行了试验验证,在KITTI数据集下对模型的性能进行评估测试,同时设计相应的消融试验验证模型中各模块的有效性。

Quantum search for unknown number of target items by hybridizing fixed-point method with trail-and-error method

For the unsorted database quantum search with the unknown fraction λ of target items, there are mainly two kinds of methods, i.e., fixed-point and trail-and-error.(i) In terms of the fixed-point method, Yoder et al. [Phys. Rev. Lett.113 210501(2014)] claimed that the quadratic speedup over classical algorithms has been achieved. However, in this paper, we point out that this is not the case, because the query complexity of Yoder’s algorithm is actually in O(1/λ01/2)rather than O(1/λ1/2), where λ0 is a known lower bound of λ.(ii) In terms of the trail-and-error method, currently the algorithm without randomness has to take more than 1 times queries or iterations than the algorithm with randomly selected parameters. For the above problems, we provide the first hybrid quantum search algorithm based on the fixed-point and trail-and-error methods, where the matched multiphase Grover operations are trialed multiple times and the number of iterations increases exponentially along with the number of trials. The upper bound of expected queries as well as the optimal parameters are derived. Compared with Yoder’s algorithm, the query complexity of our algorithm indeed achieves the optimal scaling in λ for quantum search, which reconfirms the practicality of the fixed-point method. In addition, our algorithm also does not contain randomness, and compared with the existing deterministic algorithm, the query complexity can be reduced by about 1/3. Our work provides a new idea for the research on fixed-point and trial-and-error quantum search.

改进人工势场算法的路径规划

人工势场法是一种适用于局部路径规划的算法,针对人工势场法存在的目标不可达、局部极小值等固有缺陷,提出了一种改进后的人工势场法。首先针对目标不可达问题,将引力影响因子添加到斥力场的生成中,改进斥力场函数,从而避免引力与斥力合力为零的情况。针对局部极小值问题,通过设立虚拟目标点来引入额外外力,打破机器人的平衡问题。通过与其他算法的对比实验,仿真结果显示,改进人工势场法规划的路径长度和消耗时间都更短,稳定性更强。

基于三维激光点云的序列运动图像点目标快速跟踪方法

由于现有方法未充分考虑小目标和遮挡目标检测的问题,导致目标跟踪效果不佳,由此针对点目标跟踪较为困难的问题,提出基于三维激光点云的序列运动图像点目标快速跟踪方法。首先下采样序列运动图像三维激光点云并去除其中的地面数据,通过动态阈值改进欧式聚类方法,分割图像为目标和背景,然后改进SECOND算法,在目标检测阶段融入自适应空间特征融合模块和二维卷积神经网络,同时采用3D DIoU替代Smooth L1作为损失函数,提升SECOND算法的目标检测性能,最后在激光雷达坐标系中构建各点目标的三维中心,利用3D卡尔曼滤波器连续跟踪目标,以交并比和欧式距离为度量标准,通过贪婪算法匹配最近邻目标,实现序列运动图像点目标快速跟踪。实验结果表明,所提方法IoU值更高,可达到0.971,且对目标的跟踪更理想。

面向对海监视的舰船目标跟踪与航迹融合数据集

对海监视中航迹实时关联与轨迹融合任务是安全防控、区域态势监视、远程精确打击等军民应用领域的热点和难点问题,高质量的数据集对推动目标跟踪与融合技术在该领域的研究有重要作用。本研究针对目标跟踪与融合领域技术研究的数据需求以及目前公开数据集所存在的数据缺乏、场景设计针对性差、数据格式单一、数据描述不全等问题,通过仿真软件对复杂场景中多传感器多目标探测数据进行仿真,提供了一套面向典型对海监视场景(以舰船为探测对象的2D雷达与侦察传感器﹝ESM﹞)的目标跟踪与航迹融合数据集。其中仿真软件包括剧情发生器和传感器模拟器两部分,是一套成熟的目标跟踪场景仿真环境,提供逼真的探测数据模拟能力。本数据集涵盖的传感器对象包括2D雷达与侦察传感器,目标包括典型的海上舰船类别,并支持携带辐射源配置,设计了高速运动、密集交通、多传感器数据融合、特定舰船侦测和交叉定位等多种典型场景。本数据集中共包含368155条目标点迹,舰船数量为101条,时间范围15000秒,数据格式符合实际设备上报情景、探测误差模型符合实际。本数据集通过对数据误差进行正态性检验、对检测率、虚警率的场景检验以及实地调研,完成了对数据的准确性评估与数据完备性控制,可为舰船目标跟踪、轨迹融合等算法研究与验证提供基础数据。

基于场景点云重建的移动焊接机器人作业轨迹提取方法

为保证移动作业焊接机器人能够在建筑、石化、船舶等非结构化场景中精准高效的完成作业,提出一种基于点云重建的焊接轨迹提取方法。焊接机器人变换作业场景后,首先通过深度相机实时检测引导标志物,明确作业范围和起点并重建场景点云。然后基于点云分割和边界提取算法获取焊缝点云轨迹,并提出一种焊接机器人位姿调整策略。最后,针对点云精度不足的问题,采用激光视觉传感器对焊缝进行精度修正,提取精确焊接轨迹。通过曲线焊缝场景实验,表明机器人运动平滑,焊缝轨迹提取误差小于0.5 mm,精度满足生产要求。

基于ROS-Matlab的移动机器人仿真研究

结合Linux下ROS系统高性能的仿真工具和跨平台兼容性,以及Matlab强大的计算能力、丰富的工具箱和函数库,通过建立ROS-Matlab的仿真控制系统,对移动机器人的关键技术进行研究。文中介绍了ROS-Matlab系统的框架和功能,首先,使用Gazebo构建2D仿真室内环境,通过Solidworks创建移动机器人模型,并将该模型以URDF文件形式导入ROS环境中;其次,利用SLAM_gmapping算法对仿真室内环境进行扫描建图,由Simulink发布目标点坐标和目标点导航话题,控制移动机器人完成导航任务;最后,由Simulink发布轨迹跟踪话题,控制移动机器人进行轨迹跟踪。通过仿真验证,由所建地图、目标点导航和轨迹跟踪曲线说明仿真系统的可行性,为机器人研究提供参考。

基于偏转角抑制的改进人工势场法路径规划

针对传统人工势场法在实际应用中存在的局部最极小值、目标不可达以及路径震荡等问题,文种提出了一种基于偏转角抑制的改进人工势场法。该方法以传统人工势场法为基础,采用改进斥力势场函数保证目标点为全局势能最低点。在路径解算中引入偏转角抑制因子来抑制无人车行驶过程中过大的偏转角,在无人车陷入局部极小值点后每一步的路径解算中都添加新的虚拟目标点,直至无人车累积一定的偏转角摆脱障碍物群。仿真实验结果表明,该算法可以在不影响无人车避障的情况下减少规划路径的震荡性,并能够使无人车逃脱复杂障碍物群顺利到达目标点,规划出一条有效、简短以及震荡较少的路径。

一种多层多模态融合3D目标检测方法

在自动驾驶感知系统中视觉传感器与激光雷达是关键的信息来源,但在目前的3D目标检测任务中大部分纯点云的网络检测能力都优于图像和激光点云融合的网络,现有的研究将其原因总结为图像与雷达信息的视角错位以及异构特征难以匹配,单阶段融合算法难以充分融合二者的特征.为此,本文提出一种新的多层多模态融合的3D目标检测方法:首先,前融合阶段通过在2D检测框形成的锥视区内对点云进行局部顺序的色彩信息(Red Green Blue,RGB)涂抹编码;然后将编码后点云输入融合了自注意力机制上下文感知的通道扩充PointPillars检测网络;后融合阶段将2D候选框与3D候选框在非极大抑制之前编码为两组稀疏张量,利用相机激光雷达对象候选融合网络得出最终的3D目标检测结果.在KITTI数据集上进行的实验表明,本融合检测方法相较于纯点云网络的基线上有了显著的性能提升,平均mAP提高了6.24%.

基于EP-YOLO v8的瓶栽金针菇最优抓取位置定位方法

针对工厂化瓶栽金针菇自动切根过程中,夹持末端因结构设计导致行程固定,进而影响抓取效果甚至切根质量的问题,本文基于YOLO v8(You only look once)构建改进的Enoki-pick_region-YOLO v8(EP-YOLO v8),实现瓶栽金针菇整体及最佳受力区域(关键抓取区域)的精准定位与轮廓提取,保障抓取参数的可靠性。该方法在网络优化基础上,基于最小欧几里得距离(Euclidean distance,ED)构建掩膜关系归属与判断模型,明确金针菇菇体与关键抓取区域掩膜间父子关系并合并优化。通过解析合并前后关键抓取区域的相对位置编码,确定抓取参数并进行坐标转换,为建立末端控制映射模型实现末端机械手运动行程的精确控制提供基础。实验结果表明,本文所提算法的金针菇菇体掩膜识别精确率达99.3%,关键抓取区域掩膜识别精确率达99.6%。同时,对比发现掩膜质量得到了提高,获取的参数抓取区域宽度与实际宽度之间的误差仅为0.7%,抓取参数基本满足抓取条件,能有效实现最优抓取位置的精准识别与定位。