-
公开(公告)号：KR101523740B1
公开(公告)日：2015-05-28
申请号：KR1020140000936
申请日：2014-01-03
Applicant: (주)유디피 , 브이씨에이 테크놀러지 엘티디
CPC classification number: G06K9/6232 , G06K9/00771 , G06K9/6269 , G06T5/40 , G06T7/269 , G06T7/277 , G06T2207/10016 , G08B13/19678
Abstract: 본발명은공간매핑을이용한객체추적장치및 방법에관한것으로서, 더욱상세히는입력영상에포함된객체를추적하기위한객체식별과정에서공간매핑을이용하여객체의식별에필요한연산부하를감소시키고, 객체의식별정확도를높여객체를용이하게추적할수 있는공간매핑을이용한객체추적장치및 방법에관한것이다. 본발명은 3차원공간매핑을통해추출된객체중 원하는크기의관심객체후보를선별하여전경중에서관심객체후보에대응되는객체후보영역에대해서만트레이닝세트를적용하고, 객체후보영역을트레이닝세트의크기에대응되도록리사이즈하여상기트레이닝세트를통해학습된관심객체에대한학습특징정보와객체후보영역으로부터추출된객체특징정보의비교가용이하게이루어지도록하여관심객체를식별할수 있도록함으로써, 기존과같이전경을피라미드화하여복수의이미지를생성할필요가없어객체식별에요구되는연산량을대폭줄일수 있을뿐 아니라리사이즈를통해학습특징정보가정확히적용되도록하여관심객체에대한식별정확도및 신뢰성을크게향상시킬수 있다.
Abstract translation: 本发明涉及使用空间映射跟踪对象的装置和方法。 更具体地，该装置使用空间映射来减少在识别用于跟踪包含在输入图像中的对象的对象的处理期间识别对象的计算负荷，并且该装置具有增加的用于识别对象的精度，并且因此可以容易地追踪对象 。 根据本发明，该装置以用户期望的尺寸从三维空间映射提取的对象中选择兴趣对象候选。 训练集仅适用于从前景对应于兴趣对象候选的对象候选区域。 对象候选区域被调整为对应于训练的大小。 因此，可以容易地将通过训练集学习的兴趣对象的学习特征信息与从对象候补区域提取的对象特征进行比较。 同样，该装置识别兴趣对象。 现有装置通过将前景变成金字塔来生成多个图像。 然而，本发明的装置需要做同样的事情。 因此，可以显着地减少用于识别对象所需的计算量。 此外，通过调整大小，装置可以更精确地应用学习特征信息，可以显着增加关于兴趣对象的识别精度和可靠性。
-
公开(公告)号：KR101480750B1
公开(公告)日：2015-01-12
申请号：KR1020140079177
申请日：2014-06-26
Applicant: (주)유디피
Inventor： 이규상
IPC: G06T7/20
CPC classification number: G06T5/20 , G06K9/00 , G06K9/00355 , G06K9/00771 , G06K9/38 , G06K9/6212 , G06K2009/6213 , G06T5/002 , G06T7/246 , G06T7/254 , G06T2207/10016 , G06T2207/20021 , G06T2207/30196
Abstract: 본 발명은 모션 검출 장치 및 방법에 관한 것으로서, 더욱 상세히는 프레임간 비교를 통해 픽셀 단위로 변화량을 판단하여 움직임이 발생한 객체를 정확하게 검출할 수 있는 모션 검출 장치 및 방법에 관한 것이다. 본 발명은 입력 영상을 픽셀 단위로 객체, 가비지 및 배경에 해당하는지 여부를 판단하여 움직임이 있는 객체의 영역을 정확하게 검출할 수 있을 뿐만 아니라, 기존 프레임 단위의 비교에 따른 객체 검출에서 발생하는 중복 영역에 따라 검출된 객체에 손실 영역이 발생하는 것을 방지하여 움직임이 발생한 객체 검출의 정확도를 크게 높이는 효과가 있다.
Abstract translation: 本发明涉及一种用于检测运动的装置和方法，更具体地，涉及通过帧之间的比较确定像素单位的变化量来准确地检测移动物体的装置和方法。 根据本发明，对于输入图像，输入图像是否对应于物体，以像素为单位确定垃圾或背景，使得能够准确地检测运动对象的区域，检测对象为 防止当以现有的帧单元的比较方法检测到对象时产生的重叠区域具有丢失区域。 因此，可以大大提高检测物体的移动精度。
-
公开(公告)号：KR1020140008731A
公开(公告)日：2014-01-22
申请号：KR1020120075685
申请日：2012-07-11
Applicant: (주)유디피
IPC: G03B17/55
CPC classification number: G03B17/55 , G03B17/08 , G03B17/561 , G03B17/566 , H04N5/2252
Abstract: The present invention relates to a dual camera device with a radiating structure. The present invention comprises a dual camera part having a camera box of which the front end is connected to a camera fixating end for fixating a thermal imaging camera and a general camera; a radiating body having a heat sink on the top surface and a radiating housing connected to the camera box on the front end; and an insulation part as an insulation member between the camera box and the radiating housing. [Reference numerals] (B3) Area of 220
Abstract translation: 本发明涉及具有辐射结构的双相机装置。 本发明包括具有相机盒的双相机部件，其前端连接到相机固定端，用于固定热成像相机和普通相机; 在顶面上具有散热器的辐射体和在前端与摄像机盒连接的辐射壳体; 以及绝缘部件，作为相机盒和散热壳体之间的绝缘部件。 （附图标记）（B3）220的面积
-
公开(公告)号：KR101339026B1
公开(公告)日：2013-12-09
申请号：KR1020130121763
申请日：2013-10-14
Applicant: (주)유디피
CPC classification number: G06T7/277 , G06T5/002 , G06T2207/10048 , G06T2207/30232 , H04N21/42202
Abstract: The present invention relates to a thermographic camera visibility improving method and an apparatus thereof which improves object identification performance by increasing data quantity about an object. The thermographic camera visibility improving method and the apparatus thereof improves the object identification quality by obtaining a precise image for an object of interest to observe while removing the air temperature noise of the thermographic camera. Moreover, the present invention improves the identification efficiency for an object by identifying the object based on an area, in which a temperature range which is set for the object of interest is well shown, while increasing the data quantity of the object of interest by compensating the temperature range of the object of interest for a weighted value while decreasing the noise of the thermographic camera through an equalization. [Reference numerals] (810) Thermographic camera;(820) Object temperature setting unit;(840) Image converting unit;(850) Processor;(860) Object identifying unit
Abstract translation: 本发明涉及一种通过增加关于物体的数据量来提高对象识别性能的热像摄影机可见度改善方法及其装置。 热像摄影机可见度提高方法及其装置通过在消除热像摄影机的空气温度噪声的同时观察感兴趣物体的精确图像来提高对象识别质量。 此外，本发明通过基于区域来识别对象来提高对象的识别效率，在该区域中，对于感兴趣对象设定的温度范围被很好地示出，同时通过补偿来增加感兴趣对象的数据量 通过均衡降低热成像摄像机的噪声，同时加权值的感兴趣对象的温度范围。 （810）热像仪;（820）物体温度设定单元;（840）图像转换单元;（850）处理器;（860）对象识别单元
-
公开(公告)号：KR101284842B1
公开(公告)日：2013-07-10
申请号：KR1020130040619
申请日：2013-04-12
Applicant: (주)유디피 , 브이씨에이 테크놀러지 엘티디
CPC classification number: Y02A90/32
Abstract: PURPOSE: A water level measuring device and a method thereof are provided to calculate the actual water level by compensating a change in the water level based on the inclined surface information of a monitoring region and the three-dimensional spatial information with respect to the monitoring region, thereby obtaining the accuracy and the convenience of measuring the water level. CONSTITUTION: A water level measuring device comprises a receiving unit (200), a water change determining unit (521), a calibration unit (522), and a water level measuring unit (523). The receiving unit receives image data from a camera or an external device. The water change determining unit determines a change in a water region and a land region of a monitoring region based on a water level model in which the land region and the water region are distinguished as similar property regions. The calibration unit makes the two-dimensional image data correspond with the three-dimensional spatial information by a preset calibration method. The water level measuring unit corrects the change in the water region of the monitoring region of the image data by using the based on the two-dimensional image data of the monitoring region and the three-dimensional spatial information, thereby measuring the water level. [Reference numerals] (100) Camera unit; (200) Receiving unit; (300) User interface providing unit; (400) Setting unit; (500) Control unit; (510) Filtering unit; (520) Measuring unit; (521) Water change determining unit; (522) Calibration unit; (523) Water level measuring unit; (600) Storing unit; (700) Output unit; (AA) Water level; (BB) Alarm
Abstract translation: 目的：提供一种水位测量装置及其方法，用于通过基于监视区域的倾斜表面信息和相对于监视区域的三维空间信息来补偿水位的变化来计算实际水位 ，从而获得测量水位的准确性和便利性。 构成：水位测量装置包括接收单元（200），换水确定单元（521），校准单元（522）和水位测量单元（523）。 接收单元从相机或外部设备接收图像数据。 水变化判定部基于将陆地区域和水域区分为相似的特性区域的水位模型来判定监视区域的水域和陆地区域的变化。 校准单元通过预设的校准方法使二维图像数据与三维空间信息对应。 水位测量单元通过使用基于监视区域的二维图像数据和三维空间信息来校正图像数据的监视区域的水域的变化，从而测量水位。 （附图标记）（100）相机单元; （200）接收单位; （300）用户界面提供单元; （400）设定单位; （500）控制单元; （510）过滤单元; （520）测量单元; （521）换水判定单元; （522）校准单位; （523）水位测量单元; （600）存储单元; （700）输出单元; （AA）水位; （BB）报警
-
Patent Agency Ranking
公开(公告)号：KR101262044B1
公开(公告)日：2013-05-08
申请号：KR1020120012267
申请日：2012-02-07
Applicant: (주)유디피 , 브이씨에이 테크놀러지 엘티디
Abstract: PURPOSE: An apparatus and method for detecting a vehicle using a thermal image are provided to improve efficiency by easily identifying and tracking the vehicle. CONSTITUTION: A camera is installed around a road and photographs a driving space of the vehicle with an thermal image. A data input unit(110) receives the thermal image. A pattern information unit(140) patterns a specific temperature distribution of an engine and a glass window. A vehicle sensing unit includes a vehicle recognizing unit(130). [Reference numerals] (110) Data input unit; (120) Calibration information unit; (130) Vehicle recognizing unit; (140) Pattern information unit; (150) Traffic information analysis unit
Abstract translation: 目的：提供一种用于使用热图像检测车辆的装置和方法，以通过容易地识别和跟踪车辆来提高效率。 规定：摄像机安装在道路周边，并用热图像拍摄车辆的驾驶空间。 数据输入单元（110）接收热图像。 图案信息单元（140）对发动机和玻璃窗的特定温度分布进行图案化。 车辆感测单元包括车辆识别单元（130）。 （附图标记）（110）数据输入单元; （120）校准信息单元; （130）车辆识别单元; （140）模式信息单元; （150）交通信息分析单位
-
公开(公告)号：KR101222482B1
公开(公告)日：2013-01-15
申请号：KR1020110061282
申请日：2011-06-23
Applicant: (주)유디피 , 브이씨에이 테크놀러지 엘티디
CPC classification number: G06K9/00369 , G06F3/04847 , G06F2203/04803 , G06Q90/00 , G07C9/00
Abstract: 본 발명은 설정 인터페이스를 구비한 피플 카운터 및 그 설정 방법에 관한 것으로서, 더욱 상세하게는 영상 내 이동 객체의 카운트를 위한 기준 폭을 시각적으로 배치하여 화면에 표시함으로써 이동 객체의 검출 폭을 상기 기준 폭과 비교하도록 하여 카운트를 위한 설정이 매우 용이하고, 마우스와 같은 포인팅 장치를 이용한 상기 기준 폭의 조정 및 비교를 위한 이동이 자유로운 인터페이스를 제공함으로써 종래의 수동 조정 방식에 비해 직관적이고 실용성이 높은 설정이 가능하도록 하여 영상 내 상황 또는 소정 이동 객체의 종류에 따른 다양한 환경에서 용이하게 카운트의 정확도를 향상시킬 수 있으며, 실시간 물체 측정 크기를 설정치에 근접시켜 표시함으로써 직관적으로 검출 신뢰성을 확인하면서 설정을 최적으로 조정하도록 하여 이동 객체의 종류에 따른 카운터의 설정 및 검증을 더욱 용이하게 하는 효과가 있다.
Abstract translation: 公开了包括设置界面和其设置方法的人员计数器。 由于用于对图像内的运动物体进行计数的参考宽度被明显地布置并显示在屏幕上，使得可以将检测到的移动物体的宽度与参考宽度进行比较，所以计数的设定和验证非常容易。 此外，由于可以使用诸如鼠标的指示装置来自由移动用于调整和比较参考宽度的界面，从而提供比常规手动调整方案更直观和实用的验证和重置，因此可以容易地增加计数精度 不同的环境取决于图像内的移动物体的条件或类型。
-
公开(公告)号：KR1020130000633A
公开(公告)日：2013-01-03
申请号：KR1020110061282
申请日：2011-06-23
Applicant: (주)유디피 , 브이씨에이 테크놀러지 엘티디
CPC classification number: G06K9/00369 , G06F3/04847 , G06F2203/04803 , G06Q90/00 , G07C9/00
Abstract: PURPOSE: A people counter having a setting interface and a setting method thereof are provided to compare the detection width of a moving object with a reference width by visually arranging the reference width for the count of a moving object in an image and displaying the reference width on a screen. CONSTITUTION: A setting interface unit(270) provides an interface in order to select more than one set visual element for the count of a moving object on a screen(300). A line providing unit(240) provides the passage boundary line of the moving object. A reference width providing unit(260) provides a reference width. The reference width indicates a standard about the width of the moving object of a count target passing a line. A counting unit(230) detects the width of the moving object passing the line from an input image. The counting unit counts the moving object. [Reference numerals] (100) Camera; (210) Image input unit; (220) Image analyzing unit; (230) Counting unit; (240) Line providing unit; (250) Counter providing unit; (260) Reference width providing unit; (270) Setting interface unit; (300) Screen
Abstract translation: 目的：提供一种具有设定界面及其设定方法的人物计数器，通过在图像中可视地设置运动物体的计数的基准宽度来比较运动物体的检测宽度与参考宽度，并显示参考宽度 在屏幕上 构成：设置界面单元（270）为屏幕（300）上的运动对象的计数选择多于一组的视觉元素提供界面。 行提供单元（240）提供移动物体的通过边界线。 参考宽度提供单元（260）提供参考宽度。 参考宽度表示通过线的计数目标的移动物体的宽度的标准。 计数单元（230）从输入图像检测经过线的移动物体的宽度。 计数单元对运动对象进行计数。 （附图标记）（100）相机; （210）图像输入单元; （220）图像分析单元; （230）计数单位; （240）线路提供单元; （250）柜台提供单位; （260）参考宽度提供单元; （270）设置接口单元; （300）屏幕
-
公开(公告)号：KR100923747B1
公开(公告)日：2009-10-27
申请号：KR1020080116129
申请日：2008-11-21
Applicant: (주)유디피
CPC classification number: H04N7/18 , G06F1/3234 , H04N5/23241
Abstract: PURPOSE: An apparatus and a method for controlling a low power security system are provided to comprise a security system with small storage medium which records images only required for security. CONSTITUTION: A control device and a method of the low power security system are as follows. The predetermined index value is selected by matching the running time of active mode and sleep mode of security system. The active mode or the sleep mode is performed based on the index value. When the security system is performed in a sleep mode, it is determined whether an operation event is generated to make the security system into active mode or not. As a result of the determination, if an operation event is occurred, the security system is switched to the active mode during the predetermined time.
Abstract translation: 目的：提供一种用于控制低功率安全系统的装置和方法，其包括具有小的存储介质的安全系统，其记录仅为安全性所需的图像。 构成：低功率安全系统的控制装置和方法如下。 通过匹配安全系统的活动模式和休眠模式的运行时间来选择预定的索引值。 基于索引值执行活动模式或睡眠模式。 当在睡眠模式下执行安全系统时，确定是否产生操作事件以使安全系统成为活动模式。 作为确定的结果，如果发生操作事件，则在预定时间内将安全系统切换到活动模式。
-
公开(公告)号：KR100865531B1
公开(公告)日：2008-10-29
申请号：KR1020070054261
申请日：2007-06-04
Applicant: (주)유디피
CPC classification number: G06T7/194 , G06K9/00369
Abstract: A method for separating individual pedestrians by clustering foreground pixels is provided to separate the discrete pedestrian accurately even though pedestrians can be overlapped in the foreground image by grouping each pixel about the foreground image to a plurality of groups and effectively merging and removing groups. A method for separating individual pedestrians by clustering foreground pixels relates to the method for separating the discrete pedestrian from the source image. More particularly, it is about the discrete pedestrian separation method by the grouping of the foreground pixel separating discrete pedestrians from foreground pixels separated from the source image for the behavioral analysis of pedestrian. Each pixel of the source image is classified into the background pixel and foreground pixel. And the foreground image comprising the silhouette image of the pedestrians with a plurality of foreground pixels is produced(S101). Each pixel comprising the foreground image is grouped to one or more groups of each pedestrian unit. A plurality of groups overlappingly mapped on one discrete pedestrian is merged. And after error groups which are not included in the discrete pedestrian are removed, the discrete pedestrian is distinguished(S103).
Abstract translation: 提供了通过聚类前景像素来分离个体行人的方法，即使行人可以通过将前景图像周围的每个像素分组到多个组并且有效地合并和移除组来重叠在前景图像中来精确地分离离散步行者。 通过聚类前景像素分离个体行人的方法涉及将离散行人与源图像分开的方法。 更具体地，关于离散行人分离方法，通过将离散行人与从源图像分离的前景像素分开的前景像素的分组用于行人的行为分析。 源图像的每个像素被分类为背景像素和前景像素。 产生包括具有多个前景像素的行人的轮廓图像的前景图像（S101）。 包括前景图像的每个像素被分组到每个行人单元的一个或多个组。 在一个离散行人上重叠映射的多个组合被合并。 并且在不包括在离散行人中的错误组之后，区分离散行人（S103）。
-
-
-
-
-
-
-
-
-
Search Results
42
records
(took 0.033 seconds)
