WO2023124451A1 - Procédé et appareil de génération d'événement d'alarme, dispositif et support de stockage - Google Patents

Procédé et appareil de génération d'événement d'alarme, dispositif et support de stockage Download PDF

Info

Publication number
WO2023124451A1
WO2023124451A1 PCT/CN2022/126918 CN2022126918W WO2023124451A1 WO 2023124451 A1 WO2023124451 A1 WO 2023124451A1 CN 2022126918 W CN2022126918 W CN 2022126918W WO 2023124451 A1 WO2023124451 A1 WO 2023124451A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
target person
person
scene image
image
Prior art date
Application number
PCT/CN2022/126918
Other languages
English (en)
Chinese (zh)
Inventor
宋述铕
Original Assignee
上海商汤智能科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 上海商汤智能科技有限公司 filed Critical 上海商汤智能科技有限公司
Publication of WO2023124451A1 publication Critical patent/WO2023124451A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/70—Multimodal biometrics, e.g. combining information from different biometric modalities
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data

Definitions

  • the present disclosure at least discloses a method for generating an alarm event.
  • the method may include: identifying persons in one or more scene images collected at one or more monitoring sites, determining a target person appearing in the scene image, wherein the target person is identified
  • the on-site image of the target is called the target on-site image; determine the position where the target person is recognized when appearing in the target on-site image and the time when the target person appears in the target on-site image; Perform form recognition on the target person appearing in the target scene image to determine the target person's form in the target scene image; in response to the time when the target person appears in the target scene image, the target
  • the position where the person is recognized when appearing in the target scene image, and at least one of the appearance of the target person in the target scene image conforms to a preset rule, and an alarm event is generated.
  • the alarm event includes the target scene image and detection information of the target scene image.
  • the method further includes: sending the movement track of the target person to the N preset devices closest to the target person, where N is a positive integer; wherein, the target The movement trajectory of the person is generated according to the location where the target person is recognized when appearing in the multiple target scene images and the time when the target person appears in the multiple target scene images respectively.
  • the performing form recognition on the target person appearing in the target scene image, and determining the form of the target person in the target scene image include: performing personnel identification on the target scene image Object detection, obtaining a first person detection frame corresponding to the target person in the target scene image; determining a person image indicated by the first person detection frame in the scene image according to the first person detection frame;
  • the shape classification network is used to classify the person image to obtain the shape of the target person; wherein the shape classification network includes a neural network trained based on image samples marked with shape information of the person object.
  • the performing form recognition on the target person appearing in the target scene image to determine the target person's form in the target scene image includes: respectively performing Personnel object detection and vehicle detection, obtaining a second person detection frame and a vehicle detection frame corresponding to the target person; in response to the coincidence degree of the second person detection frame and a certain vehicle detection frame reaching a preset coincidence degree threshold, determine The form of the target person is driving a vehicle; in response to the coincidence degree of the second person detection frame and any vehicle detection frame not reaching the preset coincidence degree threshold, it is determined that the form of the target person is walking.
  • the preset rule includes that the form of the target person changes from walking or driving a non-target vehicle to driving a target vehicle; in response to the time when the target person appears in the target scene image, At least one of the position where the target person is identified when he appears in the target scene image and the shape of the target person in the target scene image conforms to a preset rule, and an alarm event is generated, including: acquiring the a first form possessed by the target person at a first time, and a second form possessed by the target person at a second time, wherein the second time is later than the first time; in response to the first form comprising Walking or driving a non-target vehicle, and the second form includes driving the target vehicle, generating an alarm event; and/or, acquiring the third position and the third form possessed by the target person at a third time, and the The fourth position and the fourth form possessed by the target person at the fourth time, wherein the fourth time is later than the third time; in response to the third form including walking or driving a non-
  • the preset rule includes that the duration of the target person staying in the parking area of the non-preset cell reaches a preset threshold; , the location where the target person is identified when appearing in the target scene image, and at least one of the target person’s form in the target scene image conforms to a preset rule, generating an alarm event, including: According to the time when the target person appears, where he is, and the shape of the target person, determine the length of time the target person stays in a parking area other than the preset community; in response to the length of stay reaching a preset threshold , generating an alert event.
  • a second determining module configured to perform form recognition on the target person appearing in the target scene image, and determine the form of the target person in the target scene image; generate A module, configured to respond to the time when the target person appears in the target scene image, the location where the target person is recognized when appearing in the target scene image, and the target person's location in the target scene At least one of the morphologies in the live image conforms to a preset rule, and an alarm event is generated.
  • the present disclosure also proposes an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein, the processor executes the processor-executable instructions to implement any of the aforementioned embodiments.
  • Method for generating alert events including: a processor; a memory for storing processor-executable instructions; wherein, the processor executes the processor-executable instructions to implement any of the aforementioned embodiments.
  • the present disclosure also provides a computer-readable storage medium, where the storage medium stores a computer program, and the computer program is used to cause a processor to execute the method for generating an alarm event as shown in any one of the foregoing embodiments.
  • FIG. 2 is a flow chart of a method for identifying a target person shown in an embodiment of the present disclosure
  • FIG. 3 is a schematic flowchart of a method for identifying a person's form shown in an embodiment of the present disclosure
  • FIG. 4 is a schematic flowchart of a method for identifying a person's form shown in an embodiment of the present disclosure
  • Fig. 8 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present disclosure.
  • FIG. 1 is a schematic flowchart of a method for generating an alarm event according to an embodiment of the present disclosure.
  • the method shown in FIG. 1 can be applied to electronic equipment.
  • the electronic device may implement the method by carrying software logic corresponding to the method for generating the alarm event.
  • the electronic device can be a notebook computer, a computer, a server, a mobile phone, a personal digital assistant (Personal Digital Assistant, PDA) and the like.
  • PDA Personal Digital Assistant
  • the electronic device may also be a client device or a server device, and the type of the electronic device is not particularly limited in the present disclosure.
  • a target detection network may be used to detect personnel objects.
  • the target detection network may be a network obtained by pre-training based on image samples marked with human object detection frame information.
  • the first person detection frame corresponding to the target person can be detected through the network.
  • At least one of the target person's time, location, and target person's appearance in the live image may conform to a preset rule.
  • the target preset rule may be a modified rule of an existing preset rule, or a newly added preset rule.
  • the distance between the acquired device location and the location corresponding to the target person can be determined through a third-party map (such as Gaode map).
  • a third-party map such as Gaode map
  • the movement trajectory of the target person may be sent to the N preset devices with the closest distance to the target person.
  • the target behavior in this example includes the behavior of stealing electric vehicles.
  • the target person is a person who has stolen an electric vehicle.
  • the target person C it is found that it is tracked to ride a shared bicycle (with a third form) at the gate of the community (third position) at the third time, and it is tracked at the gate of the community (the third position) at the fourth time afterwards.
  • the fourth position) riding an electric vehicle (with the fourth form) then the state of the target person C hits the preset rule, indicating that he may have carried out the behavior of stealing the electric vehicle, and an alarm event can be generated for alarm.
  • the preset rule instructs the target person to drive the target vehicle in a non-daily activity area.
  • FIG. 7 is a schematic structural diagram of an apparatus for generating an alarm event according to an embodiment of the present disclosure.
  • the device 700 for generating an alarm event shown in FIG. 7 may include: an identification module 710, configured to identify the person in one or more scene images collected at one or more monitoring sites, and determine that the person appearing in the scene image The target person in , wherein, the on-site image that is recognized as the target person is called the target on-site image; the first determining module 720 is configured to determine that when the target person is identified to appear in the target on-site image The location and the time when the target person appears in the target scene image; the second determination module 730 is used to perform morphological recognition on the target person appearing in the target scene image, and determine the target person Morphology in the target scene image; generating module 740, configured to respond to the time when the target person appears in the target scene image, the location where the target person is recognized when appearing in the target scene image At least one of the position of the target person and the shape of the target person in
  • the alarm event includes a target scene image and detection information of the target scene image.
  • the preset rule indicates that the form of the target person changes from walking or driving a non-target vehicle to driving a target vehicle; the generating module 740 is specifically used for:
  • the present disclosure provides an electronic device, which may include: a processor and a memory for storing instructions executable by the processor.
  • the instruction corresponding to the device generating the alarm event may also be directly stored in the memory, which is not limited herein.
  • one or more embodiments of the present disclosure may be provided as a method, system or computer program product. Accordingly, one or more embodiments of the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of the present disclosure may employ a computer implemented on one or more computer-usable storage media (which may include, but are not limited to, disk storage, CD-ROM, optical storage, etc.) with computer-usable program code embodied therein. The form of the Program Product.
  • the processes and logic flows described in this disclosure can be performed by one or more programmable computers executing one or more computer programs to perform corresponding functions by operating on input data and generating output.
  • the processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).
  • FPGA Field Programmable Gate Array
  • ASIC Application Specific Integrated Circuit

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Medical Informatics (AREA)
  • Databases & Information Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Human Computer Interaction (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Alarm Systems (AREA)
  • Burglar Alarm Systems (AREA)
  • Image Analysis (AREA)

Abstract

Procédé et appareil de génération d'alarme, dispositif et support de stockage. Le procédé consiste : à effectuer une reconnaissance d'identité sur un individu dans une image sur site acquise au niveau d'un site surveillé afin de déterminer un individu cible dans l'image sur site, l'image sur site dans laquelle l'individu cible est reconnu étant appelée image sur site cible ; à déterminer un emplacement de l'individu cible dans l'image cible lorsque l'individu cible est reconnu comme apparaissant dans l'image sur site cible, et un instant auquel l'individu cible apparaît dans l'image sur site cible ; à réaliser une reconnaissance de morphologie sur l'individu cible qui apparaît dans l'image sur site cible, afin de déterminer une morphologie de l'individu cible dans l'image sur site cible ; à générer un événement d'alarme en réponse au moment où l'individu cible apparaît dans l'image sur site cible et/ou à l'emplacement de l'individu cible dans l'image sur site cible lorsque l'individu cible est reconnu comme apparaissant dans l'image sur site cible, et/ou au fait que la morphologie de l'individu cible dans l'image sur site cible satisfait une règle prédéfinie.
PCT/CN2022/126918 2021-12-30 2022-10-24 Procédé et appareil de génération d'événement d'alarme, dispositif et support de stockage WO2023124451A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111655210.0 2021-12-30
CN202111655210.0A CN114333079A (zh) 2021-12-30 2021-12-30 生成告警事件的方法、装置、设备和存储介质

Publications (1)

Publication Number Publication Date
WO2023124451A1 true WO2023124451A1 (fr) 2023-07-06

Family

ID=81018833

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/126918 WO2023124451A1 (fr) 2021-12-30 2022-10-24 Procédé et appareil de génération d'événement d'alarme, dispositif et support de stockage

Country Status (2)

Country Link
CN (1) CN114333079A (fr)
WO (1) WO2023124451A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114333079A (zh) * 2021-12-30 2022-04-12 北京市商汤科技开发有限公司 生成告警事件的方法、装置、设备和存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113034826A (zh) * 2021-03-10 2021-06-25 深圳市兴海物联科技有限公司 基于视频的异常事件告警方法及其系统、设备、存储介质
CN113128414A (zh) * 2021-04-22 2021-07-16 北京房江湖科技有限公司 人员跟踪方法、装置、计算机可读存储介质及电子设备
CN113723257A (zh) * 2021-08-24 2021-11-30 江苏范特科技有限公司 事件短视频生成方法、系统、设备和存储介质
CN114333079A (zh) * 2021-12-30 2022-04-12 北京市商汤科技开发有限公司 生成告警事件的方法、装置、设备和存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113034826A (zh) * 2021-03-10 2021-06-25 深圳市兴海物联科技有限公司 基于视频的异常事件告警方法及其系统、设备、存储介质
CN113128414A (zh) * 2021-04-22 2021-07-16 北京房江湖科技有限公司 人员跟踪方法、装置、计算机可读存储介质及电子设备
CN113723257A (zh) * 2021-08-24 2021-11-30 江苏范特科技有限公司 事件短视频生成方法、系统、设备和存储介质
CN114333079A (zh) * 2021-12-30 2022-04-12 北京市商汤科技开发有限公司 生成告警事件的方法、装置、设备和存储介质

Also Published As

Publication number Publication date
CN114333079A (zh) 2022-04-12

Similar Documents

Publication Publication Date Title
CN108062349B (zh) 基于视频结构化数据及深度学习的视频监控方法和系统
US11354901B2 (en) Activity recognition method and system
CN108009473B (zh) 基于目标行为属性视频结构化处理方法、系统及存储装置
TWI749113B (zh) 在視訊監控系統中產生警示之方法、系統及電腦程式產品
EP3654285B1 (fr) Suivi d'objets utilisant des attributs d'objets
US10152858B2 (en) Systems, apparatuses and methods for triggering actions based on data capture and characterization
WO2019153193A1 (fr) Procédé et système de surveillance d'exploitation d'un taxi, dispositif et support de stockage
CN108052859B (zh) 一种基于聚类光流特征的异常行为检测方法、系统及装置
CN105913559B (zh) 一种基于体感技术的银行atm机智能监控方法
JP4924607B2 (ja) 不審行動検知装置および方法、プログラムおよび記録媒体
KR101116273B1 (ko) 교통사고 인식장치 및 그 방법
CN103699677B (zh) 一种基于人脸识别技术的罪犯行踪地图绘制系统及其方法
CN106448160B (zh) 结合车辆行驶轨迹和监控视频数据的目标人物追踪方法
US20070035622A1 (en) Method and apparatus for video surveillance
CN103049459A (zh) 一种基于特征识别的快速录像检索方法
JP2006011728A (ja) 不審者対策システム及び不審者検出装置
WO2020024552A1 (fr) Procédé et système de surveillance de sécurité routière, et support de stockage lisible par ordinateur
WO2023124451A1 (fr) Procédé et appareil de génération d'événement d'alarme, dispositif et support de stockage
CN105227918B (zh) 一种智能监控方法
Agrawal et al. Automatic traffic accident detection system using ResNet and SVM
CN116311166A (zh) 交通障碍物识别方法、装置及电子设备
KR101154350B1 (ko) 고해상도 영상의 객체 검지 및 분할 인식을 이용한 다중 영상처리시스템 및 방법
Masood et al. Identification of anomaly scenes in videos using graph neural networks
Sankar et al. Intelligent Door Assist system using Chatbot and Facial Recognition
US20240037761A1 (en) Multimedia object tracking and merging

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22913714

Country of ref document: EP

Kind code of ref document: A1