Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020135392 - METHOD AND DEVICE FOR DETECTING ABNORMAL BEHAVIOR

Publication Number WO/2020/135392
Publication Date 02.07.2020
International Application No. PCT/CN2019/127797
International Filing Date 24.12.2019
IPC
G06K 9/00 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
CPC
G06K 9/00
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
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
G06K 9/00758
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
00758Matching video sequences
G06K 9/00771
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
00771Recognising scenes under surveillance, e.g. with Markovian modelling of scene activity
G06K 9/62
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
62Methods or arrangements for recognition using electronic means
G06K 9/6256
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
62Methods or arrangements for recognition using electronic means
6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting
Applicants
  • 杭州海康威视数字技术股份有限公司 HANGZHOU HIKVISION DIGITAL TECHNOLOGY CO., LTD. [CN]/[CN]
Inventors
  • 赵飞 ZHAO, Fei
Agents
  • 北京三高永信知识产权代理有限责任公司 BEIJING SAN GAO YONG XIN INTELLECTUAL PROPERTY AGENCY CO., LTD.
Priority Data
201811581954.024.12.2018CN
Publication Language Chinese (ZH)
Filing Language Chinese (ZH)
Designated States
Title
(EN) METHOD AND DEVICE FOR DETECTING ABNORMAL BEHAVIOR
(FR) PROCÉDÉ ET DISPOSITIF DE DÉTECTION DE COMPORTEMENT ANORMAL
(ZH) 异常行为检测方法及装置
Abstract
(EN)
The present disclosure relates to the field of video surveillance, and disclosed thereby are a method and device for detecting abnormal behavior. The method comprises: acquiring behavior data to be detected; inputting the behavior data into a feature extraction model and outputting a behavior feature of the behavior data, wherein the feature extraction model is used to output a behavior feature within a feature space range according to normal behavior data and output a behavior feature outside of the feature space range according to abnormal behavior data, and the distance between each behavior feature within the feature space range is less than a distance threshold; according to the distance between the behavior feature of the behavior data and a normal behavior feature center and the distance threshold, acquiring a detection result of the behavior data, the detection result being used to indicate whether the behavior data is abnormal behavior data, and the normal behavior feature center being used to represent the behavior features within the feature space range. The described method for detecting abnormal behavior on the basis of distance measurement described in the present disclosure has high accuracy.
(FR)
La présente invention se rapporte au domaine de la surveillance vidéo, et concerne un procédé et un dispositif de détection de comportement anormal. Le procédé consiste à : acquérir des données de comportement devant subir une détection ; fournir les données de comportement en entrée d'un modèle d'extraction de caractéristiques et produire en sortie une caractéristique de comportement des données de comportement, le modèle d'extraction de caractéristiques étant utilisé pour produire une caractéristique de comportement dans une plage spatiale de caractéristiques selon des données de comportement normal et produire une caractéristique de comportement hors de la plage spatiale de caractéristiques selon des données de comportement anormal, et la distance entre chaque caractéristique de comportement dans la plage spatiale de caractéristiques est inférieure à un seuil de distance ; selon la distance entre la caractéristique de comportement des données de comportement et un centre de caractéristiques de comportement normal et le seuil de distance, acquérir un résultat de détection des données de comportement, le résultat de détection étant utilisé pour indiquer si les données de comportement sont des données de comportement anormal, et le centre de caractéristiques de comportement normal étant utilisé pour représenter les caractéristiques de comportement dans la plage spatiale de caractéristiques. Le procédé décrit pour détecter un comportement anormal en fonction d'une mesure de distance décrit dans la présente invention offre une précision élevée.
(ZH)
本公开公开了一种异常行为检测方法及装置,属于视频监控领域。所述方法包括:获取待检测的行为数据;将所述行为数据输入特征提取模型,输出所述行为数据的行为特征,所述特征提取模型用于根据正常行为数据输出特征空间范围内的行为特征以及根据异常行为数据输出所述特征空间范围外的行为特征,所述特征空间范围内各个行为特征之间的距离小于距离阈值;根据所述行为数据的行为特征与正常行为特征中心的距离和所述距离阈值,获取所述行为数据的检测结果,所述检测结果用于指示所述行为数据是否为异常行为数据,所述正常行为特征中心用于代表所述特征空间范围内的行为特征。本公开这种基于距离度量的异常行为检测方法,准确性较高。
Also published as
Latest bibliographic data on file with the International Bureau