-
公开(公告)号：KR101402206B1
公开(公告)日：2014-05-30
申请号：KR1020140043001
申请日：2014-04-10
Applicant: 국방과학연구소
IPC: G01S17/66
Abstract: The present invention relates to a multiple target tracking method, more particularly, to a multiple target tracking method using distance information of a target, feature information-based Kalman filter, and joint probability data-association (JPDA) method. According to the present invention, a JPDA method generates not only normalized distance squared (NDS) distance information but also an effective matrix through dual conditions based on the correlation information of an image. By using this, the identification of other objects, which closely exist, can be possible and the complexity decreases as the number of cases which are taken into account decrease.
Abstract translation: 多目标跟踪方法技术领域本发明涉及多目标跟踪方法，更具体地，涉及使用目标的距离信息，基于特征信息的卡尔曼滤波器和联合概率数据关联（JPDA）方法的多目标跟踪方法。 根据本发明，JPDA方法不仅产生归一化距离平方（NDS）距离信息，而且还通过基于图像的相关信息的双重条件生成有效矩阵。 通过使用这一点，密切存在的其他对象的识别是可能的，并且随着考虑到的案例数量的减少，复杂性降低。
-
公开(公告)号：KR101280348B1
公开(公告)日：2013-07-01
申请号：KR1020120005611
申请日：2012-01-18
Applicant: 국방과학연구소
Abstract: PURPOSE: A multiple target tracking method is provided to improve tracking performance by accurately classifying target and clutter using geometric information extracted from a target image. CONSTITUTION: A multiple target tracking method includes a target state prediction step predicting a kinetic state vector of a target and a state vector of a geometric image characteristic, an NN measurement value selection step selecting an NN measurement value through a sum of the kinetic state vector and state vector of geometric image characteristic provided in the target state prediction step and a state vector estimation update step updating state vector estimation using the NN measurement value selected by the NN measurement selection step. [Reference numerals] (AA) Start; (BB) End; (S301) Target state prediction step predicting a kinetic state vector/a state vector of a geometric image characteristic; (S302) NN measurement value selection step selecting an NN measurement value through (the kinetic state vector + state vector of geometric image characteristic); (S303) State vector estimation update step updating state vector estimation using the NN measurement value
Abstract translation: 目的：提供多目标跟踪方法，通过使用从目标图像提取的几何信息，精确地对目标和杂波进行分类，提高跟踪性能。 构成：多目标跟踪方法包括预测目标的动态状态向量和几何图像特征的状态向量的目标状态预测步骤，NN测量值选择步骤，通过动态状态矢量的和选择NN测量值 以及在目标状态预测步骤中提供的几何图像特性的状态矢量和使用由NN测量选择步骤选择的NN测量值来更新状态矢量估计的状态矢量估计更新步骤。 （附图标记）（AA）开始; （BB）结束; （S301）预测几何图像特征的动态状态向量/状态向量的目标状态预测步骤; （S302）NN测量值选择步骤通过（动态状态向量+几何图像特征的状态向量）来选择NN测量值; （S303）状态向量估计更新步骤使用NN测量值来更新状态向量估计
-
公开(公告)号：KR101590889B1
公开(公告)日：2016-02-02
申请号：KR1020140025506
申请日：2014-03-04
Applicant: 국방과학연구소
Abstract: 본발명에서는레이더(20)가추적하는다수의제1,2,3 타겟(100-1,100-2,100-3)에대한영상을공중에서촬영하는영상획득장치(10), 영상획득장치(10)에서촬영한감시영상을이용해레이더(20)의위치이동추적성능이판단되는원격모니터(30)로위치추정시스템이구성되고, 공중에서획득한영상을기반으로레이더(20)가추적하는표적을자동으로검출하고, 검출된표적의위치를주변참고점을이용하여추정하고위치를계산하며, 이동하는표적의궤적을분석함으로써기동하는표적의위치가효과적으로추출될수 있고, 특히이러한과정에서다른센서의추적결과와비교가수행됨으로써레이더(20)의추적성능에대한평가및 분석수행이정확하게이루어지는특징을갖는다.
-
公开(公告)号：KR1020140124145A
公开(公告)日：2014-10-24
申请号：KR1020130041499
申请日：2013-04-16
Applicant: 국방과학연구소
Abstract: The present invention relates to a device for correcting mounting angles of multiple radar for a vehicle. The device, which comprises two radar mounted in a vehicle, is capable of: calculating mounting angles of the two radar using obstacle location values measured from the two radar by putting an arbitrary obstacle in front of the vehicle; and correcting the obstacle location values from the calculated mounting angles. According to the present invention, the device can more accurately track the location of a moving obstacle in front of the vehicle by estimating the mounting angles of multiple radar for the vehicle and by correcting the estimated mounting angles.
Abstract translation: 本发明涉及一种用于校正车辆的多雷达的安装角度的装置。 该装置包括安装在车辆中的两个雷达，其能够：使用从两个雷达测量的障碍物位置值，在车辆前方放置任意障碍物来计算两个雷达的安装角度; 以及根据所计算的安装角度来校正障碍物位置值。 根据本发明，通过估计车辆的多个雷达的安装角度并且通过校正估计的安装角度，该装置可以更精确地跟踪车辆前方的移动障碍物的位置。
-
公开(公告)号：KR101628154B1
公开(公告)日：2016-06-08
申请号：KR1020150030943
申请日：2015-03-05
Applicant: 국방과학연구소
CPC classification number: G01S13/726 , G01S7/28 , G01S13/53
Abstract: 본발명은수신신호세기를이용한다중표적추적방법에관한것으로, 더욱상세하게는, 레이더가탐지한다수의표적신호로부터수신신호세기데이터를생성하고상기수신신호세기데이터가생성된표적신호중 소정의유효거리이내에존재하는표적신호만을추출한후, 상기수신신호세기데이터를이용하여상기추출된표적신호의데이터연관확률을생성하고, 상기데이터연관확률이적용된추적필터를이용하여상기표적에대한상태벡터연산치및 오차공분산을구함으로써상기표적의항적을갱신하는수신신호세기를이용한다중표적추적방법에관한것이다.
Abstract translation: 本发明涉及使用接收信号强度指示符（RSSI）的多目标跟踪方法，更具体地说，涉及使用RSSI的多目标跟踪方法，RSSI从雷达检测到的多个跟踪信号产生RSSI数据，提取 仅在产生RSSI数据的跟踪信号之间存在一定有效距离内的跟踪信号，通过使用RSSI数据产生提取的跟踪信号的数据相关概率，并获得状态向量计算值和误差协方差值 通过使用应用了数据相关概率的跟踪滤波器来对目标进行更新，从而更新目标的轨道。 根据本发明，将目标的不同RSSI数据应用于联合概率数据关联（JPDA）或廉价联合概率数据关联（CJPDA），从而提高数据相关技术的性能。
-
Patent Agency Ranking
公开(公告)号：KR1020140104131A
公开(公告)日：2014-08-28
申请号：KR1020130017880
申请日：2013-02-20
Applicant: 국방과학연구소
Abstract: After a target obtained from an electro-optical scope is visually recognized through an image on a display screen, the type and parameters of the recognized target are selected and then a reference lane in case when the target is positioned in a reference distance, 1 km, is generated from an established DB. A distance to the currently obtained target can be calculated without laser or radio waves by comparing a pixel of the lane, controlled according to the parameters of the target selected through a process of controlling the size of the reference lane to be matched with the recognized target, with a pixel at the reference distance. Especially, the present invention can calculate the distance to the target in a state where the target is partially hidden by a forest by selecting the parameters of a recognizable target area even if the target image is partially obtained.
Abstract translation: 在通过显示屏幕上的图像从电光范围获得的目标被视觉识别之后，选择识别的目标的类型和参数，然后在目标位于参考距离的情况下选择参考车道1km ，从已建立的DB生成。 可以通过比较通过将通过控制参考车道的尺寸与所识别的目标匹配的过程所选择的目标的参数来控制的通道的像素来比较当前获得的目标的距离而不使用激光或无线电波 ，具有参考距离处的像素。 特别地，即使部分获得目标图像，本发明也可以通过选择可识别的目标区域的参数来计算目标被森林部分隐藏的状态的目标的距离。
-
公开(公告)号：KR101426864B1
公开(公告)日：2014-08-07
申请号：KR1020130005253
申请日：2013-01-17
Applicant: 국방과학연구소
Abstract: 본 발명은 표적의 열상정보를 획득하여 표적의 궤적을 추적하기 위해 추적창 조절을 이용하는 영상추적기의 근거리 열상 표적추적을 위한 영역기반 경계향상 방법에 관한 것으로, 상기 발명은 상기 획득된 열상정보를 휘도 기울기 생성을 통해 경계 영상으로 변환하고 배경의 경계를 약화시킨 기준영상을 크기가 동일한 사전에 정해진 개수의 영역으로 분할하는 단계; 및 상기 분할된 영역들 각각에 속하는 경계의 세기를 비교하여 세기의 차이가 나는 영역들 간에는 보간을 수행하고 상기 영역들 각각에 속한 픽셀의 평균값을 이용하여 상기 추적창 조절을 수행하는 단계를 포함하는 것을 특징으로 한다. 이에 의해 본 발명은 영상추적을 위한 영역기반 경계영상을 향상시킴으로써 근거리 열상 표적 추적의 정확성과 지속성을 향상시킬 수 있다.
-
18.发明授权
公开(公告)号：KR101394164B1
公开(公告)日：2014-05-16
申请号：KR1020130005247
申请日：2013-01-17
Applicant: 국방과학연구소
CPC classification number: G06T7/248 , G06T3/4023 , G06T3/4053 , G06T7/32
Abstract: The present invention relates to a super-resolution-based high-precision video tracking method for a video tracker, and a high-precision video tracking apparatus using the same. The high-precision video tracking method comprises the steps of: extracting a target image from a input image (t-1) and determining the size of the extracted target image; performing a first interpolation of creating 1/2 subpixels and a second interpolation of creating 1/4 subpixels, according to the determined size of the target image; selecting one among a target image created by pixels, a target image created by the first interpolation or a target image created by the second interpolation according to the determined size of the target image and updating a reference image; extracting, from a current input image (t), a search area set to find a target image and determining the size of the extracted search area; performing the first and second interpolations according to the determined size of the search area; and selecting one among the search area created by pixels, a search area created by the first interpolation, and a search area created by the second interpolation according to the size of the updated reference image, matching the selected one with a corresponding reference image, and determining the location of a target in the search area. Accordingly, the present invention can improve real-time properties by minimizing load according to the generation of a super-resolution image while enhancing the accuracy of displacement calculation of a target by amplifying only high-frequency components of the target except for a background.
Abstract translation: 本发明涉及一种用于视频跟踪器的基于超分辨率的高精度视频跟踪方法以及使用其的高精度视频跟踪装置。 高精度视频跟踪方法包括以下步骤：从输入图像（t-1）中提取目标图像并确定所提取的目标图像的大小; 根据确定的目标图像的大小执行创建1/2子像素的第一内插和创建1/4子像素的第二内插; 选择由像素创建的目标图像之一，通过第一内插创建的目标图像或根据所确定的目标图像大小创建的目标图像，并更新参考图像; 从当前输入图像（t）提取设置为找到目标图像并确定所提取的搜索区域的大小的搜索区域; 根据确定的搜索区域的大小执行第一和第二内插; 并且根据所更新的参考图像的大小，将所选择的参考图像与对应的参考图像相匹配，并且选择由像素创建的搜索区域，由第一内插创建的搜索区域和由第二内插创建的搜索区域，以及 确定目标在搜索区域中的位置。 因此，本发明可以通过根据超分辨率图像的产生来最小化负载来提高实时性，同时通过仅放大除了背景的目标的高频分量来增强目标的位移计算的精度。
-
公开(公告)号：KR101348680B1
公开(公告)日：2014-01-09
申请号：KR1020130002389
申请日：2013-01-09
Applicant: 국방과학연구소
CPC classification number: G06T7/223 , G06T7/13 , G06T2207/30241
Abstract: The present invention relates to a target acquisition method for an image tracking system which tracks the trace of a target by acquiring and analyzing image information of the target and a target acquisition apparatus using the same. The target acquisition method includes the following steps of: (a) Gaussian-filtering an input image and generating a gradient contour image of a target through gradient magnitude conversion; (b) generating a cumulative and positive difference of edges (PDOE) at a frame interval, accumulating the generated PDOE images, and producing an accumulated PDOE image; (c) detecting a cell which the target belongs to by measuring the change quantity of a pixel at a cell unit in the accumulated PDOE image by using a half-grid search; and (d) designating the detected cell as a candidate cell in case an average pixel value of the detected cell is greater than or equal to a predetermined threshold value and determining the effectiveness of the detected cell by using the number of the candidate cells. By the aforementioned, the present invention improves a real-time target acquisition and reliability because the noise of a difference image is minimized, and a response time is shortened by using the technology of the half-grid search and the accumulated PDOE made by changing an existing detection algorithm and the PDOE. [Reference numerals] (AA) Start; (BB) End; (S210) Gaussian-filtering + Gradient magnitude detection; (S220) Produce an accumulated PDOE; (S230) Grid search; (S235) Kmin
Abstract translation: 本发明涉及一种图像跟踪系统的目标获取方法，其通过获取和分析目标图像信息和使用其的目标获取装置跟踪目标轨迹。 目标获取方法包括以下步骤：（a）对输入图像进行高斯滤波，并通过梯度幅度转换产生目标的梯度轮廓图像; （b）以帧间隔产生边缘的累积和正差值（PDOE），累积产生的PDOE图像，并产生累积的PDOE图像; （c）通过使用半网格搜索来测量累积PDOE图像中的单元单元处的像素的变化量来检测目标所属的单元; 以及（d）在检测到的单元的平均像素值大于或等于预定阈值的情况下将所检测的单元指定为候选单元，并且通过使用候选单元的数量来确定检测单元的有效性。 通过上述，本发明改善了实时目标的获取和可靠性，因为差分图像的噪声被最小化，并且通过使用半网格搜索的技术和通过改变图像的累积PDOE来缩短响应时间 现有检测算法和PDOE。 （附图标记）（AA）开始; （BB）结束; （S210）高斯滤波+梯度幅度检测; （S220）生成累计PDOE; （S230）网格搜索; （S235）Kmin <细胞数
-
公开(公告)号：KR101196298B1
公开(公告)日：2012-11-06
申请号：KR1020110119244
申请日：2011-11-16
Applicant: 국방과학연구소
Abstract: PURPOSE: A high speed track merging method is provided to perform high speed clustering through a nearest measured value among measured values which is entered in a validation gate. CONSTITUTION: A unique number is given to detection information obtained from a sensor and a validation gate of each track is calculated(S310). A unique number of nearest detection information among detection information which is entered in the validation gate of each track is added to information property of each track(S320). Tracks in which the unique number of the nearest detection information is same are clustered from information property of each track(S330). Distance between the tracks which are clustered based on a selected track is calculated(S350). Track merging is performed(S360). [Reference numerals] (S310) A unique number is given; (S320) The nearest unique number is added; (S330) Track clustering; (S340) A track in which a determinant of covariance is the smallest is established; (S350) The distance between tracks is calculated; (S360) Track merging
Abstract translation: 目的：提供高速跟踪合并方法，通过在验证门中输入的测量值之间的最近测量值来执行高速聚类。 构成：对从传感器获得的检测信息给出唯一的编号，并计算每个轨道的验证门（S310）。 在每个轨道的验证门中输入的检测信息中唯一数量最近的检测信息被添加到每个轨道的信息属性（S320）。 其中最近的检测信息的唯一数量相同的跟踪是从每个轨道的信息属性聚集的（S330）。 计算基于所选轨道聚集的轨道之间的距离（S350）。 执行跟踪合并（S360）。 [参考号]（S310）给出唯一的号码; （S320）添加最近的唯一编号; （S330）跟踪聚类; （S340）建立其中协方差的行列式最小的轨道; （S350）计算轨道之间的距离; （S360）跟踪合并
-
-
-
-
-
-
-
-
-
Search Results
23
records
(took 0.022 seconds)
