共查询到20条相似文献，搜索用时 171 毫秒
1.
为了将图像处理技术应用于智能客流检测,准确及时地获取运动信息,对视频图像中运动物体的分割与跟踪技术进行研究,将图像分割与跟踪技术有机地结合在一起。采用帧间与帧内相结合的分割算法,将图像填充技术应用于图像分割,消除了背景区域的干扰以及图像处理过程中造成的失真,达到了准确获取视频流中运动物体的目的。在图像跟踪过程,引入了边缘衍生点的概念,在此基础上,提出基于部分轮廓线特征点集的多线程跟踪算法,把运动物体跟踪转变为连续帧间特征点集匹配的问题,极大地减少了运算量,解决了图像跟踪处理领域中目标物体自遮挡与不完全遮挡的难题。经客流测试证明,文中方法能够对上下车客流数量进行准确统计。 相似文献
2.
海底影像存在着对比度低、噪声污染严重、图像质量差等问题,采用传统算子的海底管线边缘图像中含有大量的无用和断裂边缘信息。文中将多尺度边缘检测和匹配跟踪相结合,提出利用平稳小波变换的海底管线边缘检测方法;在提取边缘的同时利用匹配跟踪手段对噪声干扰进行抑制,提高图像目标边缘检测质量。通过对海底管线和测试图像边缘检测实验表明,文中所提出的方法在抑制图像噪声的干扰、提高水下目标边缘完整性方面明显优于传统的边缘提取算子,证明该算法的有效性。 相似文献
3.
4.
本文以卫星云图的热带气旋云系为研究对象,基于Canny边缘检测、contour轮廓提取等方法,研究了通过图像的形状特征和范围大小对热带气旋云系进行自动识别,得到了较好的效果。这种方法有助于提高热带气旋云系识别的自动化程度,实现热带气旋的自动跟踪,从而为热带气旋的预报提供了便利。 相似文献
5.
《中国海洋大学学报(自然科学版)》2017,(2)
目前针对背景复杂的海上舰船作为跟踪对象的跟踪算法研究较少。另外,多目标跟踪算法在变化频繁的背景条件下的精确性和实时性不足。为了提高多目标舰船跟踪的视觉跟踪算法的实时性和鲁棒性,本文提出了一种改进的基于TLD(Tracking-Learning-Detection)多目标舰船自动跟踪算法。该算法实现了海上多个目标舰船准确实时跟踪。首先,算法应用海天线检测与提取方法,利用最大类间方差阈值分割和Hough变换提取海天线;其次,算法利用Kalman滤波原理对海天线上的目标舰船进行定位与检测,分离出目标舰船;最后通过提取出来的目标用其最小外接矩形生成目标初始跟踪框,跟踪器利用初始跟踪框的位置坐标信息对目标进行实时自动跟踪。算法运行过程中都在海天线附近进行扫描检测,缩小了图像遍历范围,提高了算法的实时性。实验表明,通过对比典型的Mean-shift算法以及原始TLD目标跟踪算法,本文算法跟踪结果的精度较高,实时性较好。 相似文献
6.
7.
8.
北冰洋航路是未来全球航运的重点开拓领域,海冰运动对北冰洋航路开发有重要影响。本文利用风云三号卫星中分辨率光谱成像仪(FY-3/MERSI)数据的特点和优势,研究北冰洋流冰自动提取和运动跟踪的方法。首先在分析流冰灰度分布特征的基础上,提出分区域阈值分割与梯度差分相结合的方法实现块状流冰提取,然后根据块状流冰的多种几何特征匹配同名流冰并计算其运动速度。应用这种方法跟踪2011年6月弗雷姆海峡流冰运动,跟踪结果与美国国家冰雪数据中心提供的极地网格化日均海冰运动矢量整体趋势一致,验证了方法的有效性。这种方法获取各块流冰实际运动速度,可以有效弥补网格化海冰运动平均速度场分布和细节的不足,为北冰洋航路开发提供更详细的流冰运动信息。 相似文献
9.
随着雷达成像技术和高分辨率光栅显示技术的发展和应用,基于船载雷达图像的船只检测成为可能.海上船只检测的主要困难之一是雷达图像中包含固有的海面背景杂波.传统的雷达船只检测方法,如恒虚警率法(CFAR),以杂波分布模型为基础,计算待检测窗口中的信号统计分布来确定自适应阈值,取得了一些成果.但是,当海面背景杂波和船只目标的回波强度在同一数量级,甚至船只目标淹没在海面背景杂波中时,就难以确定一个有效的阈值将船只目标从雷达图像中提取出来.在分析海面背景杂波和船只目标的相关差异性基础上,提出了一种基于船载雷达序列图像的海上船只快速检测方法.该方法首先对相邻两幅图像进行互相关性分析,在两幅图像中的同一位置提取一定尺寸的移动窗口,计算其互相关函数值,窗口移动一个步长,重复操作直至遍布整幅图像,形成一幅由灰度图像互相关函数值组成的相关图像.然后使用概率神经网络模型(PNN模型)来估计相关图像背景杂波的灰度概率密度分布函数(PDF),应用CFAR技术,使用二分法求解一个区分船只和背景噪声的自适应整体阈值,并根据阈值将相关图像二值化,其中大于阈值的像元作为候选的船只目标信息,小于阈值的像元则为海面背景杂波.最后使用连通性8-邻域准则统计各个候选船只目标区域的像元数,并与预先定义的最小船只目标像元数进行比较,偏小的候选船只目标区域作为虚警去除,保留下来的候选船只目标区域即为船只检测结果.研究显示,如果图像序列中没有船只目标信息,则三维相关图像比较平整.相反,如果图像序列中含有船只目标信息,则三维相关图像上有峰值被检测出,通过测量峰值的高度,就能判断存在可能的船只目标.运用X波段船载雷达序列图像对本文提出的海上船只检测方法进行了测试.测试结果表明,该检测方法具有很好的船只检测效果,得到的船只检测结果与目视判别的结果一致.而且该检测算法原理简单,计算速度快,易于实时处理,具有广阔的应用前景. 相似文献
10.
基于数学形态学的侧扫声呐图像轮廓自动提取 总被引：1，自引：0，他引：1
侧扫声呐图像特征自动提取的难点在于特征地貌边缘检测较困难,依据图像灰度突变检测得到的边缘比较粗糙、不连续,而且有断口和小洞。本文在对图像进行预处理和阈值化的基础上,采用数学形态学方法对图像进行处理,即用具有一定形态的结构元素去量度和提取图像中的对应形状,得到连续化、粗化、圆滑的特征区域边缘填充目标内部阴影且消除背景噪声。基于数学形态学的侧扫声呐图像特征自动提取的主要步骤为:首先对侧扫声呐图像进行预处理,然后进行灰度阈值化,接着采用数学形态学方法进行处理,最后对处理后的图像进行边缘检测,提取出特征地貌边缘。实验表明,采用数学形态学方法进行处理后,错断、离散的海底目标物变得连续,背景噪声大大减少,自动提取结果准确可靠。 相似文献
11.
以7 000 m载人潜水器的工程需求为背景,以水下单目摄像机为视觉传感器,进行了水下机器人动力定位方法研究。该动力定位方法利用视觉系统测量得到水下机器人与被观察目标之间的三维位姿关系,通过路径规划、位置控制和姿态控制分解,逐步使机器人由初始位姿逼近期望位姿并最终定位于期望位姿,从而实现了机器人的4自由度动力定位。通过水池实验验证了提出的动力定位方法,并且机器人能够抵抗恒定水流干扰和人工位置扰动。同时,该动力定位方法还可以实现机器人对被观察目标的自动跟踪。 相似文献
12.
This paper describes a new framework for detection and tracking of underwater pipeline,which includes software system and hardware system.It is designed for vision system of AUV based on monocular CCD camera.First,the real-time data flow from image capture card is pre-processed and pipeline features are extracted for navigation.The region saturation degree is advanced to remove false edge point group after Sobel operation.An appropriate way is proposed to clear the disturbance around the peak point in the process of Hough transform.Second,the continuity of pipeline layout is taken into account to improve the efficiency of line extraction.Once the line information has been obtained,the reference zone is predicted by Kalman filter.It denotes the possible appearance position of the pipeline in the image.Kalman filter is used to estimate this position in next frame so that the information of pipeline of each frame can be known in advance.Results obtained on real optic vision data in tank experiment are displayed and discussed.They show that the proposed system can detect and track the underwater pipeline online,and is effective and feasible. 相似文献
13.
针对固定粒子数PF-TBD算法计算量大、复杂环境下地波雷达海上船只目标检测与跟踪性能不佳的问题,本文将粒子滤波方法应用于地波雷达船只目标检测与跟踪中,提出了基于自适应粒子滤波的地波雷达目标检测与跟踪联合处理方法。该方法结合地波雷达回波谱中目标展宽特性,充分利用了地波雷达回波谱中面目标的粒子权重信息来设置粒子自适应采样策略,提高了目标检测和跟踪联合处理的效果。通过地波雷达实测数据的目标跟踪结果及与同步AIS信息的比对分析,结果表明:提出的检测跟踪联合处理方法在对低信噪比、快速机动等复杂环境下的多目标跟踪时,可提高目标整体跟踪性能。 相似文献
14.
针对海洋环境下自主水下机器人（AUV）的路径规划问题,提出了一种基于框架四叉树的改进量子粒子群算法（QPSO）,首先使用框架四叉树的方法对障碍物建模,该方法提高了建模的精度且对后续算法的效率也有极大的改进,之后设计改进的量子粒子群算法,并且结合水下环境的特殊性设计适应度函数,综合考虑航线路径长度、偏转角度以及海流影响,使得算法可以在水下环境中寻得能耗最短的解路径。最后通过仿真试验验证,相比于传统的栅格法和粒子群算法,改进量子粒子群算法的运算时间更短,收敛速度更快,其独特的适应度函数可以使AUV能更好适应水下多变的环境,且能利用海流设计能耗更小的路径,具有很大的实用价值。 相似文献
15.
Autonomous vehicles are being developed to replace the conventional, manned surface vehicles that tow mine hunting towed platforms. While a wide body of work exists that describes numerical models of towed systems, they usually include relatively simple models of the towed bodies and neglect the dynamics of the towing vehicle. For systems in which the mass of the towing vehicle is comparable to that of the towed vehicle, it becomes important to consider the dynamics of both vehicles. In this work, we describe the development of a numerical model that accurately captures the dynamics of these new mine hunting systems. We use a lumped mass approximation for the towcable and couple this model to non-linear numerical models of an autonomous surface vehicle and an actively controlled towfish. Within the dynamics models of the two vehicles, we include non-linear controllers to allow accurate maneuvering of the towed system. 相似文献
16.
为了最终实现对海洋中尺度涡流(简称中尺度涡)的自动采样,首先应该发展中尺度涡动态特征识别技术。本文基于SLA(Sea Level Anomaly)数据,实现了对中尺度涡动态特征的检测算法。主要内容是制定了一个判别相邻两组SLA数据中的涡流,是否为同一涡流子在不同时刻的状态的标准,即判别下一时刻SLA数据中是否存在涡流是由上一时刻确定的被检测涡流演化而来的。通过确定这种进化关系,可以得到被检测涡流的一系列动态状态信息,例如:面积变化速率、中心移动情况以及其他情况。本算法的计算量不大,从而可以应用到实时涡流跟踪的环境中。值得注意的是,本文中的算法不仅仅局限于应用SLA数据,SSH(Sea Surface Height)等大部分反映海洋高度的数据也可以使用。 相似文献
17.
This paper proposes a saturated tracking controller for underactuated autonomous marine surface vehicles with limited torque. First, a second-order open-loop error dynamic model is developed in the actuated degrees of freedom to simplify the design procedure. Then, a saturated tracking controller is designed by utilizing generalized saturation functions to reduce the risk of actuator saturation. This, in turn, improves the transient performance of the control system. A multi-layer neural network and adaptive robust control techniques are also employed to preserve the controller robustness against unmodeled dynamics and environmental disturbances induced by waves and ocean currents. A Lyapunov stability analysis shows that all signals of the closed-loop system are bounded and tracking errors are semi-globally uniformly ultimately bounded. Finally, simulation results are provided for a hovercraft vehicle to illustrate the effectiveness of the proposed controller as a qualified candidate for real implementations in offshore applications. 相似文献
18.
Depth-trim mapping control of underwater vehicle with fins 总被引：1，自引：1，他引：0
Underwater vehicle plays an important role in ocean engineering.Depth control by fin is one of the difficulties for underwater vehicle in motion control.Depth control is indirect due to the freedom coupling between trim and axial motion.It includes the method of dynamic analysis and lift-resistance-coefficient experiment and theory algorithm.By considering the current speed and depth deviation,comprehensive interpretation is used in object-planning instruction.Expected depth is transformed into expected trim.Dynamic output fluctuation can be avoided,which is caused by linear mapping of deviation.It is steady and accurate for the motion of controlled underwater vehicles.The feasibility and efficiency of the control method are testified in the pool and natural area for experiments. 相似文献
19.
This paper presents an integrated navigational algorithm for unmanned underwater vehicles (UUV) using two acoustic range transducers and strap-down inertial measurement unit (SD-IMU). A range measurement model is derived for a UUV having one acoustic transducer and cruising around two reference transponders at sea floor or surface. The proposed algorithm, called pseudo long base line (PLBL), estimates the position of the vehicle integrating the SD-IMU signals corrected with the two range measurements. Extended Kalman filter was applied to propagate error covariance, to update measurement errors and to correct state equation whenever the external measurements are available. Simulations were conducted to illustrate the effectiveness of the PLBL using the 6-d.o.f. nonlinear numerical model of a UUV at current flow, excluding bottom-fixed DVL. This paper also shows the error convergence of the vehicle's initial position by the additional range measurements without velocity information. 相似文献
20.
In this study, a dynamic modeling method for foil-like underwater vehicles is introduced and experimentally verified in different sea tests of the Hadal ARV. The dumping force of a foil-like underwater vehicle is sensitive to swing motion. Some foil-like underwater vehicles swing periodically when performing a free-fall dive task in experiments. Models using conventional modeling methods yield solutions with asymptotic stability, which cannot simulate the self-sustained swing motion. By improving the ridge regression optimization algorithm, a grey-box modeling method based on 378 viscous drag coefficients using the Taylor series expansion is proposed in this study. The method is optimized for over-fitting and convergence problems caused by large parameter matrices. Instead of the PMM test data, the unsteady computational fluid dynamics calculation results are used in modeling. The obtained model can better simulate the swing motion of the underwater vehicle. Simulation and experimental results show a good consistency in free-fall tests during sea trials, as well as a prediction of the dive speed in the swing state. 相似文献