Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020063189 - VIDEO TARGET TRAJECTORY EXTRACTION METHOD AND DEVICE

Publication Number WO/2020/063189
Publication Date 02.04.2020
International Application No. PCT/CN2019/101273
International Filing Date 19.08.2019
IPC
G06K 9/62 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
CPC
G06K 9/00744
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
00624Recognising scenes, i.e. recognition of a whole field of perception; recognising scene-specific objects
00711Recognising video content, e.g. extracting audiovisual features from movies, extracting representative key-frames, discriminating news vs. sport content
00744Extracting features from the video content, e.g. video "fingerprints", or characteristics, e.g. by automatic extraction of representative shots or key frames
G06T 7/13
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
13Edge detection
G06T 7/136
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
136involving thresholding
G06T 7/155
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
155involving morphological operators
G06T 7/194
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
10Segmentation; Edge detection
194involving foreground-background segmentation
G06T 7/90
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
7Image analysis
90Determination of colour characteristics
Applicants
  • 惠州学院 HUIZHOU UNIVERSITY [CN]/[CN]
Inventors
  • 蔡昭权 CAI, Zhaoquan
  • 蔡映雪 CAI, Yingxue
  • 陈伽 CHEN, Jia
  • 胡松 HU, Song
  • 黄思博 HUANG, Sibo
  • 李慧 LI, Hui
  • 胡辉 HU, Hui
  • 陈明阳 CHEN, Mingyang
Agents
  • 北京中济纬天专利代理有限公司 BEIJING WITAL PATENT AGENCY LTD.
Priority Data
PCT/CN2018/10751426.09.2018CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) VIDEO TARGET TRAJECTORY EXTRACTION METHOD AND DEVICE
(FR) PROCÉDÉ ET DISPOSITIF D'EXTRACTION DE TRAJECTOIRE DE CIBLE VIDÉO
(ZH) 一种视频目标轨迹提取方法及装置
Abstract
(EN)
Disclosed by the present disclosure are a video target trajectory extraction method and device. Firstly, an estimated transparency value is recalculated by measuring the credibility of foreground and background pixel pairs to obtain a first transparency mask of a first image; secondly, a new image is generated by superimposing gray information to obtain a second transparency mask of the first image and further modify the first transparency mask of the first image; finally, the modified first transparency mask is used to extract foreground targets of a certain frame of image in a video, a specified target is retrieved from all foreground targets, and the trajectory of the specified target is generated based on all images including the specified target in accordance with the time sequence of all frames. The disclosure can comprehensively utilize the credibility and gray information of the foreground and background pixel pairs in a certain frame of image in the video, so as to provide a new video target trajectory extraction scheme.
(FR)
La présente invention concerne un procédé et un dispositif d'extraction de trajectoire de cible vidéo. Tout d'abord, une valeur de transparence estimée est recalculée par mesure de la crédibilité de paires de pixels de premier plan et d'arrière-plan pour obtenir un premier masque de transparence d'une première image ; deuxièmement, une nouvelle image est générée en superposant des informations de gris pour obtenir un second masque de transparence de la première image et en modifiant en outre le premier masque de transparence de la première image ; enfin, le premier masque de transparence modifié est utilisé pour extraire des cibles de premier plan d'une certaine trame d'image dans une vidéo, une cible spécifiée est extraite à partir de toutes les cibles de premier plan, et la trajectoire de la cible spécifiée est générée sur la base de toutes les images comprenant la cible spécifiée conformément à la séquence temporelle de toutes les trames. L'invention peut utiliser de manière complète les informations de crédibilité et de gris des paires de pixels de premier plan et d'arrière-plan dans une certaine trame d'image dans la vidéo, de façon à fournir un nouveau schéma d'extraction de trajectoire de cible vidéo.
(ZH)
本公开揭示一种视频目标轨迹提取方法及装置,首先通过度量前景背景像素对的可信度重新计算透明度估计值以获得第一图像的第一透明度遮罩,然后通过叠加灰度信息来生成新图片并获得第一图像的第二透明度遮罩,并进一步修正第一图像的第一透明度遮罩,最后利用修正的所述第一透明度遮罩将视频中某一帧图像的前景目标提取出来,并在所有前景目标中检索指定目标,然后按照所有帧的时间顺序,根据包括所述指定目标的所有图像生成所述指定目标的轨迹。本公开能够综合利用视频中某一帧图像中前景背景像素对的可信度和灰度信息,提供一种新的视频目标轨迹提取的方案。
Latest bibliographic data on file with the International Bureau