WO2022059123A1 - 監視システム、カメラ、解析装置及びａｉモデル生成方法 - Google Patents

監視システム、カメラ、解析装置及びａｉモデル生成方法 Download PDF

Info

Publication number
WO2022059123A1
WO2022059123A1 PCT/JP2020/035229 JP2020035229W WO2022059123A1 WO 2022059123 A1 WO2022059123 A1 WO 2022059123A1 JP 2020035229 W JP2020035229 W JP 2020035229W WO 2022059123 A1 WO2022059123 A1 WO 2022059123A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
analysis
camera
cameras
detecting
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/JP2020/035229
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
海斗 笹尾
圭吾 長谷川
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kokusai Denki Electric Inc
Original Assignee
Hitachi Kokusai Electric Inc
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 Hitachi Kokusai Electric Inc filed Critical Hitachi Kokusai Electric Inc
Priority to US18/043,198 priority Critical patent/US20230306711A1/en
Priority to JP2022550253A priority patent/JP7399306B2/ja
Priority to PCT/JP2020/035229 priority patent/WO2022059123A1/ja
Publication of WO2022059123A1 publication Critical patent/WO2022059123A1/ja
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/94—Hardware or software architectures specially adapted for image or video understanding
    • G06V10/95—Hardware or software architectures specially adapted for image or video understanding structured as a network, e.g. client-server architectures
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources

Definitions

  • the present invention relates to a surveillance system including a plurality of cameras having a function of video analysis using an AI model.
  • Video surveillance systems that provide 24-hour video surveillance using image pickup devices such as surveillance cameras are widespread.
  • AI Artificial Intelligence
  • the video monitoring system using AI has a function to save only the minimum necessary information such as the video and time when a specific object appears, a warning icon is displayed by a display device, and a buzzer sounds. It has a function to encourage. Therefore, the video surveillance system using AI is useful for reducing the burden of monitoring work as compared with the conventional method that requires constant confirmation.
  • Patent Document 1 discloses a camera having a plurality of learning models that can be switched.
  • AI analysis accuracy (for example, detection accuracy of a specific object) is a very important factor.
  • a system for performing AI analysis on a server or a cloud for images obtained from each of a plurality of cameras, and a system for performing AI analysis inside each camera have been constructed.
  • it is difficult to track the detected object across a plurality of cameras because the cooperation between the cameras is not taken into consideration.
  • the present invention has been made in view of the above-mentioned conventional circumstances, and is intended to improve cooperation between cameras in a surveillance system equipped with a plurality of cameras having a function of video analysis using an AI model. The purpose.
  • the monitoring system is configured as follows. That is, the surveillance system according to the present invention includes a plurality of cameras having a function of performing video analysis using an AI model, and an analysis device for generating an AI model used in the cameras. Specific behavior is detected by video analysis using the first AI model for detecting specific behavior, and the analysis device detects the specific behavior when any camera detects the specific behavior, and the analysis device performs the specific behavior. A second AI model for detecting is generated, the second AI model is transmitted to one or more of a plurality of cameras, and one or more cameras are identified by video analysis using the second AI model. It is characterized by detecting a person.
  • the analysis device regenerates the second AI model and transmits it to one or more of a plurality of cameras. May be good.
  • the one or more cameras may include a camera whose distance from the camera that detected the specific behavior is smaller than the threshold value.
  • the present invention can also be realized as a camera having a function of performing video analysis using an AI model. That is, the camera according to the present invention has a process of detecting the specific behavior by video analysis using the first AI model for detecting the specific behavior, and a process of notifying the analysis device when the specific behavior is detected. It is characterized by performing a process of receiving a second AI model for detecting a specific object that has performed a specific behavior from an analysis device and a process of detecting a specific object by video analysis using the second AI model. do.
  • the present invention can also be realized as an analysis device that generates an AI model used for video analysis in a camera. That is, the analysis device according to the present invention performs the specific behavior when the specific behavior is detected by the image analysis using the first AI model for detecting the specific behavior with any of the plurality of cameras. It is characterized by performing a process of generating a second AI model for detecting a specific object and a process of transmitting the second AI model to one or more of a plurality of cameras.
  • the present invention can also be realized as an AI model generation method for generating an AI model used for video analysis in a camera. That is, the AI model generation method according to the present invention is when the analysis device detects the specific behavior by video analysis using the first AI model for detecting the specific behavior with any of a plurality of cameras.
  • the analysis device has a step of generating a second AI model for detecting a specific object that has performed a specific behavior, and a step of transmitting the second AI model to one or more of a plurality of cameras.
  • the present invention it is possible to improve the cooperation between cameras in a surveillance system equipped with a plurality of cameras having a function of video analysis using an AI model.
  • the monitoring system according to the embodiment of the present invention will be described with reference to the drawings.
  • the suspicious person here is a person who suddenly takes suspicious behavior (including abnormal behavior) and is not normally detected.
  • Examples of the suspicious person include a person who has taken suspicious behavior such as entering a restricted area or leaving behind his / her belongings.
  • FIG. 1 shows a configuration example of a monitoring system according to an embodiment of the present invention.
  • the monitoring system of this example includes one or more AI cameras 101 (N in FIG. 1), an analysis device 102, one or more PC 103 (M in FIG. 1), an alarm device 104, and a mobile terminal. It is equipped with 105 and is connected to each other so as to be able to communicate with each other via a communication network such as the Internet.
  • one or more alarming devices 104, mobile terminals 105, and the like may be provided.
  • the AI camera 101 is an imaging device having a function of not only shooting a surveillance area but also analyzing the image based on the AI model.
  • the AI camera 101 is preset with an AI model for detecting specific behavior for detecting suspicious behavior. Further, as will be described later, when a suspicious behavior is detected by any of the AI cameras 101, an AI model for detecting a specific person suitable for tracking a specific person (that is, a suspicious person) who has performed the behavior is additionally added. Set.
  • Each AI camera 101 not only detects a suspicious person using the AI model for detecting specific behavior (first detection of suspicious person), but also tracks the suspicious person using the AI model for detecting a specific person (second and subsequent times). Suspicious person detection) is also performed. Further, the AI camera 101 is associated with an identifier (for example, a camera number) that identifies each of the AI cameras 101, and is added to communication data with other devices such as the analysis device 102 and the PC 103.
  • an identifier for example, a camera number
  • the analysis device 102 stores the images and analysis results obtained by each AI camera 101, and integrates and reanalyzes these analysis results. Further, when the analysis device 102 detects a suspicious behavior in the video analysis of any of the plurality of AI cameras 101, the analysis device 102 generates an AI model for detecting a specific person for detecting the suspicious person who performed the behavior. And the process of transmitting the AI model for detecting a specific person to each AI camera 101.
  • the AI model for detecting specific behavior is generated by deep learning based on past images of specific behavior labeled as suspicious behavior, for example.
  • the AI model for detecting a specific person is, for example, the physical characteristics (age, gender, height, face, body shape, skin color, hairstyle, hair color, etc.) and clothes of a suspicious person obtained by analyzing an image in which suspicious behavior is detected. , Generated based on feature data such as personal belongings. If the video of the suspicious behavior at the time of detection does not sufficiently obtain the characteristic data of the suspicious person, the characteristic data of the suspicious person may be acquired from the video before or after the detection time.
  • the specific behavior detection AI model is generated before the start of operation of the monitoring system, set in each AI camera 101, and updated as necessary. On the other hand, the AI model for detecting a specific person is automatically generated in response to the discovery of a suspicious person and set in each AI camera 101.
  • the PC 103 accesses the AI camera 101 or the analysis device 102 according to the user operation, and displays the past or present monitoring area image, the analysis result, and the like.
  • the alarm device 104 issues an alarm in response to a detection signal transmitted from the AI camera 101 when a suspicious person is detected.
  • the mobile terminal 105 displays an image transmitted from the AI camera 101 when a suspicious person is detected and an analysis result.
  • the detection coordinates of the suspicious person the number of detected persons, the detection attributes (gender, age, height, nationality, etc. of the suspicious person), the detection date and time, and the detection location are included in the image. included.
  • FIG. 2 shows a configuration example of the AI camera 101.
  • the AI camera 101 includes functional units such as a video receiving unit 201, an AI analysis unit 203, a data transmitting unit 204, a data storage unit 205, an AI model receiving unit 207, and an AI model updating unit 208, an AI model memory 202, and video / analysis. It has a storage unit such as a result storage device 206. Further, as hardware, the AI camera 101 includes an ISP (Image Signal Processor) that performs correction processing and sharpening processing of an optical system such as a lens, and a GPU (Graphics Processing Unit) or FPGA (Field) for performing AI analysis at high speed. -Programmable Gate Array), or other AI chips to replace them.
  • ISP Image Signal Processor
  • GPU Graphics Processing Unit
  • FPGA Field
  • the video receiving unit 201 acquires video from an image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal-Oxide-Semiconductor).
  • the AI analysis unit 203 analyzes the video acquired by the video reception unit 201 by the AI model in the AI model memory 202 (that is, the AI model for specific behavior detection and the AI model for specific person detection).
  • the data transmission unit 204 transmits data such as the video acquired by the video reception unit 201 and the analysis result by the AI analysis unit 203.
  • the data storage unit 205 stores the video acquired by the video reception unit 201 and data such as the analysis result by the AI analysis unit 203 in the video / analysis result storage device 206.
  • the AI model receiving unit 207 receives the AI model from the analysis device 102.
  • the AI model update unit 208 updates the AI model in the AI model memory 202 with the AI model received by the AI model reception unit 207.
  • FIG. 3 shows a configuration example of the analysis device 102.
  • the analysis device 102 includes functional units such as a data reception unit 301, an analysis unit 302, a data storage unit 303, a learning data generation unit 305, an AI model generation unit 307, and an AI model transmission unit 309, and a video / analysis result storage device 304. It has a storage unit such as a learning data memory 306 and an AI model memory 308. Further, the analysis device 102 is configured by a computer system or a cloud equipped with a CPU (Central Processing Unit), GPU, FPGA, etc. as hardware, and executes processing (described later) of each functional unit.
  • CPU Central Processing Unit
  • GPU GPU
  • FPGA field-programmable gate array
  • the data receiving unit 301 receives data such as video and analysis results transmitted from the AI camera 101.
  • the analysis unit 302 performs reanalysis based on the video received by the data reception unit 301 and the analysis result.
  • the re-analysis by the analysis unit 302 is, for example, a more detailed analysis than the analysis by the AI camera 101, or a comprehensive or cross-sectional analysis using images of a plurality of AI cameras 101.
  • the data storage unit 303 stores the video received by the data reception unit 301, the analysis result, the data reanalyzed by the analysis unit 302, and the like in the video / analysis result storage device 304.
  • the learning data generation unit 305 generates learning data for an AI model using the video and analysis results stored in the video / analysis result storage device 304, and stores the learning data in the training data memory 306.
  • the AI model generation unit 307 generates an AI model using the training data in the training data memory 306 and stores it in the AI model memory 308.
  • the AI model transmission unit 309 transmits the AI model in the AI model memory 308 to the AI camera 101.
  • FIG. 4 shows an example of a processing flow for detecting a suspicious person by the monitoring system of this example.
  • the suspicious person detection process includes a specific behavior detection process for discovering a suspicious person (first detection of a suspicious person) using an AI model for detecting a specific behavior, and an AI model for detecting a suspicious person.
  • the AI camera (1) and the AI camera (2) are suspicious by AI analysis of the captured video acquired by the video receiving unit 201 using the AI model for specific behavior detection in the AI model memory 202 at the video analysis unit 203. Behavior detection processing is performed (steps S101 and S102). As a result, it is assumed that the AI camera (1) detects suspicious behavior (step S103). In this case, the AI camera (1) transmits a detection signal and an analysis result to the alarm device 104 in order to notify the observer and the like of the discovery of the suspicious person (first detection of the suspicious person) (step S104). Further, the video and the analysis result for several seconds before and after the moment when the suspicious person is detected are transmitted to the mobile terminal 105 and the analysis device 102 (steps S105 and S106).
  • a wired LAN for example, a wired LAN, a wireless LAN including Wi-Fi, LTE, 5G, etc., a contact, and the like can be used.
  • the image transmitted by the AI camera may be a snapshot (still image) at the moment when a suspicious person is detected.
  • the AI camera (1) saves the same transmitted video and analysis result data in the built-in video / analysis result storage device 206 (step S107).
  • the timing for saving the data may be before the data such as the video and the analysis result are transmitted.
  • the analysis device 102 receives the video and the analysis result from the AI camera (1)
  • the analysis device 102 saves the received video and the analysis result in the built-in video / analysis result storage device 304 (step S108).
  • the alarm device 104 issues an alarm (step S109).
  • the alarm can be issued by using, for example, a pre-recorded voice, an alarm sound, an emergency bell, light emission using a rotating light, or the like.
  • the mobile terminal 105 displays the received video and a message notifying the discovery of the suspicious person on the screen of the mobile terminal 105 (step S110).
  • the analysis device 102 generates an AI model for detecting a specific person in order to track a suspicious person found by the suspicious behavior detection process in response to receiving a video and an analysis result from the AI camera (1) (1).
  • Step S111 The analysis device 102 transmits the generated AI model for detecting a specific person to the AI camera (1) and the AI camera (2) (steps S112 and S113).
  • the AI model updating unit 208 transfers the received AI model for specific person detection to the AI model memory 202.
  • Store steps S114, S115).
  • the AI camera (1) and the AI camera (2) use the AI model for specific person detection in the AI model memory 202 to perform specific person detection processing by AI analysis of the captured video acquired by the video receiving unit 201. (Steps S116 and S117). As a result, it is assumed that a suspicious person is detected by the AI camera (2) (step S118). In this case, the AI camera (2) transmits the detection signal and the analysis result to the alarm device 104 in order to notify the observer and the like of the detection of the suspicious person (the second and subsequent detections of the suspicious person) (step S119). Further, the video and the analysis result for several seconds before and after the moment when the suspicious person is detected are transmitted to the mobile terminal 105 and the analysis device 102 (steps S120 and S121). Further, the AI camera (2) saves the same transmitted video and analysis result data in the built-in video / analysis result storage device 206 (step S122). Subsequent operations of each device are basically the same as when a suspicious person is found.
  • the analysis device 102 detects the specific person AI based on the image at the time of the first suspicious person detection (when the suspicious person is found) and the image at the time of the second suspicious person detection. Regenerate the model and send it to each AI camera. Similarly, at the time of subsequent detection of a suspicious person, the analysis device 102 regenerates the AI model for detecting a specific person based on the images at the time of detecting the suspicious person so far, and transmits the AI model to each AI camera. As a result, the learning data used to generate the AI model for detecting a specific person gradually increases, so that it becomes possible to more accurately detect (track) a suspicious person.
  • the PC 103 can always access the AI camera 101 and the analysis device 102. Therefore, the PC 103 can be used to refer to the video and analysis results delivered in real time by the AI camera 101, or the video / analysis result storage device 206 in the AI camera 101 or the video / analysis result storage device in the analysis device 102. You can refer to the past images and analysis results stored in 304.
  • the monitoring system of this example includes a plurality of AI cameras 101 having a function of performing video analysis using an AI model, and an analysis device 102 for generating an AI model used in the AI camera 101.
  • Each AI camera 101 detects suspicious behavior by video analysis using an AI model for detecting specific behavior for detecting suspicious behavior, and the analysis device 102 detects suspicious behavior with any of the AI cameras 101.
  • an AI model for detecting a specific person (suspicious person) who has performed suspicious behavior is generated and transmitted to each AI camera 101, and each AI The camera 101 is configured to detect a suspicious person by video analysis using an AI model for detecting a specific person.
  • each AI camera when a suspicious person is found by video analysis using the AI model for detecting specific behavior, each AI camera has an AI model for detecting a specific person suitable for detecting the suspicious person. Provided to. As a result, a plurality of AI cameras can cooperate to detect a suspicious person, so that the suspicious person can be tracked efficiently. In addition, since each AI camera analyzes its own captured image, it is possible to improve the real-time performance of suspicious person detection. In the above explanation, one suspicious person is found and tracked, but it is also possible to find and track a plurality of suspicious persons.
  • an AI model for detecting a specific person is provided for all AI cameras so that a suspicious person can be tracked using all AI cameras, but only some AI cameras detect a specific person. You may also send the AI model for.
  • the AI model for detecting a specific person may be transmitted only to the AI camera whose distance from the AI camera that has detected the suspicious behavior is smaller than the threshold value.
  • the threshold value for example, a distance that a person can move within a predetermined time can be used.
  • the suspicious person can be efficiently tracked by using only the AI camera within the range in which the suspicious person can move within a predetermined time. In addition, it becomes possible to avoid unnecessary processing with an AI camera outside the above range.
  • the present invention is a monitoring system using an AI model for detecting a specific behavior of an object to be monitored such as a vehicle, an animal, or a luggage, and an AI model for detecting a specific object that has performed the specific behavior.
  • an AI model for detecting a specific object that has performed the specific behavior.
  • the present invention also provides, for example, a method including a technical procedure relating to the above processing, a program for executing the above processing by a processor, a storage medium for storing such a program in a computer-readable manner, and the like. Is also possible.
  • the present invention can be used in a surveillance system equipped with a plurality of cameras having a function of video analysis using an AI model.
  • AI camera 101: AI camera, 102: analysis device, 103: PC, 104: alarm device, 105: mobile terminal, 201: video receiver, 202: AI model memory, 203: AI analysis unit, 204: data transmission unit, 205 : Data storage unit, 206: Video / analysis result storage device, 207: AI model reception unit, 208: AI model update unit, 301: Data reception unit, 302: Analysis unit, 303: Data storage unit, 304: Video / analysis unit Result storage device, 305: training data generation unit, 306: training data memory, 307: AI model generation unit, 308: AI model memory, 309: AI model transmission unit

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Human Computer Interaction (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Alarm Systems (AREA)
  • Image Analysis (AREA)
PCT/JP2020/035229 2020-09-17 2020-09-17 監視システム、カメラ、解析装置及びａｉモデル生成方法 Ceased WO2022059123A1 (ja)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US18/043,198 US20230306711A1 (en) 2020-09-17 2020-09-17 Monitoring system, camera, analyzing device, and ai model generating method
JP2022550253A JP7399306B2 (ja) 2020-09-17 2020-09-17 監視システム、カメラ、解析装置及びａｉモデル生成方法
PCT/JP2020/035229 WO2022059123A1 (ja) 2020-09-17 2020-09-17 監視システム、カメラ、解析装置及びａｉモデル生成方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/035229 WO2022059123A1 (ja) 2020-09-17 2020-09-17 監視システム、カメラ、解析装置及びａｉモデル生成方法

Publications (1)

Publication Number Publication Date
WO2022059123A1 true WO2022059123A1 (ja) 2022-03-24

Family

ID=80776577

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/035229 Ceased WO2022059123A1 (ja) 2020-09-17 2020-09-17 監視システム、カメラ、解析装置及びａｉモデル生成方法

Country Status (3)

Country Link
US (1) US20230306711A1 (https=)
JP (1) JP7399306B2 (https=)
WO (1) WO2022059123A1 (https=)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115439787A (zh) * 2022-09-07 2022-12-06 长扬科技(北京)股份有限公司 粮库的ai视觉检测方法、装置、电子设备及存储介质
WO2025253542A1 (ja) * 2024-06-05 2025-12-11 日本電気株式会社 情報処理装置、情報処理方法、及び記録媒体

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12549593B2 (en) 2023-02-23 2026-02-10 Reliaquest Holdings, Llc Threat mitigation system and method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170076576A1 (en) * 2014-05-04 2017-03-16 Seow Loong Tan Activity monitoring method and system
JP2018207222A (ja) * 2017-05-31 2018-12-27 パナソニックＩｐマネジメント株式会社 カメラ及びパラメータ登録方法
JP6573297B1 (ja) * 2019-01-16 2019-09-11 パナソニック株式会社 監視カメラおよび検知方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160180239A1 (en) * 2014-12-17 2016-06-23 Cloudtalk Llc Motion detection and recognition employing contextual awareness
US11164041B2 (en) * 2019-08-14 2021-11-02 Toyota Research Institute, Inc. Semi-supervised learning with infrastructure
KR102143029B1 (ko) * 2020-04-02 2020-08-10 아성글로벌(주) 인공지능형 cctv 시스템
US12417632B2 (en) * 2020-05-15 2025-09-16 Tyco Fire & Security Gmbh Systems and methods of identifying persons-of-interest
KR102152237B1 (ko) * 2020-05-27 2020-09-04 주식회사 와치캠 상황 분석 기반의 cctv 관제 방법 및 시스템

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170076576A1 (en) * 2014-05-04 2017-03-16 Seow Loong Tan Activity monitoring method and system
JP2018207222A (ja) * 2017-05-31 2018-12-27 パナソニックＩｐマネジメント株式会社 カメラ及びパラメータ登録方法
JP6573297B1 (ja) * 2019-01-16 2019-09-11 パナソニック株式会社 監視カメラおよび検知方法

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115439787A (zh) * 2022-09-07 2022-12-06 长扬科技(北京)股份有限公司 粮库的ai视觉检测方法、装置、电子设备及存储介质
CN115439787B (zh) * 2022-09-07 2023-08-04 长扬科技(北京)股份有限公司 粮库的ai视觉检测方法、装置、电子设备及存储介质
WO2025253542A1 (ja) * 2024-06-05 2025-12-11 日本電気株式会社 情報処理装置、情報処理方法、及び記録媒体

Also Published As

Publication number Publication date
JPWO2022059123A1 (https=) 2022-03-24
JP7399306B2 (ja) 2023-12-15
US20230306711A1 (en) 2023-09-28

Similar Documents

Publication Publication Date Title
CN117749995B (zh) 一种基于多场景识别与语音交互的视频监控方法及系统
EP3410413B1 (en) Improved generation of alert events based on a detection of objects from camera images
JP4617269B2 (ja) 監視システム
KR20230004421A (ko) 인공지능 기반의 이상행동 감지 시스템
JP7399306B2 (ja) 監視システム、カメラ、解析装置及びａｉモデル生成方法
WO2017071084A1 (zh) 报警方法及装置
JP7363838B2 (ja) 異常挙動通知装置、異常挙動通知システム、異常挙動通知方法、およびプログラム
CA3227189A1 (en) Methods and apparatus for detecting pets
CN118334811A (zh) 一种非授权人员入侵检测系统及其入侵检测方法
CN110060441A (zh) 用于终端防盗的方法和装置
JP7492595B2 (ja) 監視システム、解析装置及びａｉモデル生成方法
KR102472369B1 (ko) 긴급도움 요청 음원을 통한 객체 추적시스템 및 추적방법
CN121153065A (zh) 用于检测未识别移动对象的方法和设备
KR20220000873A (ko) 인공 지능을 이용한 관제 서비스 시스템
CN103227916B (zh) 一种监控视频备份方法、装置及系统
JP7447626B2 (ja) 情報処理方法、情報処理装置、プログラム及び情報処理システム
CN113128447A (zh) 口罩识别方法及装置、无人机、存储介质
AG et al. Development of a portable community video surveillance system
KR102527208B1 (ko) 반려동물 관리 서비스 융합장치, 및 이를 이용한 서비스 융합 시스템
JP2022081882A (ja) 野生動物監視システム、野生動物監視方法、及びプログラム
WO2021199311A1 (ja) 監視装置、監視方法、および記録媒体
JP2003329762A (ja) 対象物定位システム
CN112804492B (zh) 一种电子猫眼的通信提示方法及装置
CN113963501A (zh) 拍摄报警方法、拍摄装置及可读存储介质
TWI831727B (zh) 影像模式應用方法及其監控設備

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: 20954109

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022550253

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20954109

Country of ref document: EP

Kind code of ref document: A1