-
Patent Agency Ranking
公开(公告)号：KR101372944B1
公开(公告)日：2014-03-25
申请号：KR1020120073856
申请日：2012-07-06
Applicant: 한국과학기술원
IPC: G06T7/00
Abstract: 본발명은기울인비디오단면도를이용한비디오시그니처추출장치및 방법에관한것이다. 이를위한비디오시그니처추출장치는, 비디오또는학습비디오가입력되는비디오입력부와, 비디오또는학습비디오로부터시공간큐브를추출한후에추출된시공간큐브로부터시공간서브큐브를생성하는시공간비디오처리부와, 생성된시공간서브큐브를대상으로하여모션량을계산한후에계산된모션량에따른기울기를가지는기울어진비디오단면도를추출하는단면도추출부와, 추출된기울어진비디오단면도로부터질감과에지를추출하여서술자를생성하는서술자처리부와, 학습비디오로부터미리생성하여둔 VoVW(Vocabulary of Visual Words)를저장하는 VoVW 저장부와, 비디오로부터생성된서술자에게 VoVW와의비교를통해근접워드를할당하여 BoSTW(Bag of Spatio-Temporal Words)를출력하는 BoSTW 출력부를포함한다.
-
公开(公告)号：KR1020140006566A
公开(公告)日：2014-01-16
申请号：KR1020120073856
申请日：2012-07-06
Applicant: 한국과학기술원
IPC: G06T7/00
CPC classification number: G06K9/00744 , G06T7/13 , G06T7/254 , G06T7/40
Abstract: The present invention relates to apparatus and method for extracting a video signature using an inclined video cross section. The apparatus comprises: a video input unit for receiving videos or learning videos; a visuospatial video processing unit for extracting a visuospatial cube from the videos or the learning videos and generating a visuospatial subcube from the extracted visuospatial cube; a cross section extracting unit for calculating the amount of motion for the generated visuospatial subcube and extracting an inclined video cross section with a gradient depending on the calculated among of motion; a descriptor processing unit for extracting texture and edges from the extracted inclined video cross section to generate descriptor; a vocabulary of visual words (VoVW) storage unit for storing VoVW which has been previously generated from the learning videos; and a bag of spatio-temporal words (BoSTW) output unit for assigning a proximity word to the descriptor generated from the videos by comparison between the descriptor and the VoVW to output BoSTW. [Reference numerals] (110) Video input unit; (120) Visuospatial video processing unit; (130) Cross section extracting unit; (140) Descriptor processing unit; (150) VoVW storage unit; (160) BoSTW output unit
Abstract translation: 本发明涉及使用倾斜视频横截面提取视频签名的装置和方法。 该装置包括：用于接收视频或学习视频的视频输入单元; 视觉空间视频处理单元，用于从视频或学习视频中提取视觉空间立方体，并从提取的视觉空间立方体生成视觉空间子多维数据集; 横截面提取单元，用于计算所产生的视空间子分支的运动量，并且根据运动中计算的梯度提取倾斜视频横截面; 描述符处理单元，用于从所提取的倾斜视频横截面中提取纹理和边缘以产生描述符; 用于存储先前从学习视频产生的VoVW的视觉词（VoVW）存储单元的词汇表; 以及一袋时空词（BoSTW）输出单元，用于通过描述符和VoVW之间的比较将接近词分配给从视频产生的描述符，以输出BoSTW。 （附图标记）（110）视频输入单元; （120）视空间视频处理单元; （130）截面提取单元; （140）描述符处理单元; （150）VoVW存储单元; （160）BoSTW输出单元
-
公开(公告)号：KR1020110056999A
公开(公告)日：2011-05-31
申请号：KR1020090113523
申请日：2009-11-23
CPC classification number: G06K9/00718 , G06T7/11 , G06T7/41 , G06T7/90
Abstract: PURPOSE: A harmful image discriminating and blocking apparatus is provided to efficiently model big harmful meaning through a meaning based feature and to efficiently block a harmful image. CONSTITUTION: A pre-processing unit(2000) divides an input image into one or more areas. A visual feature generator(3000) generates visual feature about the divided areas. A meaning feature generator(4000) generates the meaning feature from visual features by area. A harmful image blocking unit(5000) decides the final harmfulness of the image. A meaning feature modeling unit(6000) models meaning feature models which are stored in a meaning feature model storage(4300).
Abstract translation: 目的：提供有害的图像识别和阻挡装置，通过基于含义的特征有效地建模大的有害意义，并有效地阻止有害的图像。 规定：预处理单元（2000）将输入图像分成一个或多个区域。 视觉特征生成器（3000）生成关于分割区域的视觉特征。 意义特征生成器（4000）根据区域的视觉特征生成意义特征。 有害图像阻挡单元（5000）决定图像的最终有害性。 意义特征建模单元（6000）对存储在意义特征模型存储（4300）中的特征模型进行建模。
-
公开(公告)号：KR101730964B1
公开(公告)日：2017-04-27
申请号：KR1020150077219
申请日：2015-06-01
Applicant: 한국과학기술원
IPC: G07D5/00
Abstract: 본발명에따른이진화패턴을이용한동전인식방법은, 동전영역에대응하게세그먼테이션된사각형의동전영상을획득하는과정과, 획득된상기동전영상으로부터미분영상을생성하는과정과, 생성된상기미분영상으로부터동심원형태의고리(ring) 내분할된서브영역들에대한에지강도변화패턴의추출을통해영역이진화패턴을추출하는과정과, 추출된상기영역이진화패턴과트레이닝용동전영상의기준이진화패턴간의비유사도측정을통해동전의종류를분류하는과정을포함할수 있다.
Abstract translation: 根据本发明的使用二值图案的硬币识别方法包括以下步骤：获取与硬币区域对应的分段四边形的四边形图像;从获得的硬币图像生成差分图像; 通过提取同心环中划分的子区域的边缘强度变化图案并且将提取的二值化图案与提取的二值化图案的参考二值化图案进行比较并训练硬币图像来提取区域二值化图案， 它可能包括通过测量分类硬币类型的过程。
-
公开(公告)号：KR1020160141489A
公开(公告)日：2016-12-09
申请号：KR1020150077219
申请日：2015-06-01
Applicant: 한국과학기술원
IPC: G07D5/00
Abstract: 본발명에따른이진화패턴을이용한동전인식방법은, 동전영역에대응하게세그먼테이션된사각형의동전영상을획득하는과정과, 획득된상기동전영상으로부터미분영상을생성하는과정과, 생성된상기미분영상으로부터동심원형태의고리(ring) 내분할된서브영역들에대한에지강도변화패턴의추출을통해영역이진화패턴을추출하는과정과, 추출된상기영역이진화패턴과트레이닝용동전영상의기준이진화패턴간의비유사도측정을통해동전의종류를분류하는과정을포함할수 있다.
-
-
-
-
Search Results
5
records
(took 0.02 seconds)
