WO2017071084A1 - 报警方法及装置 - Google Patents

报警方法及装置 Download PDF

Info

Publication number
WO2017071084A1
WO2017071084A1 PCT/CN2015/099562 CN2015099562W WO2017071084A1 WO 2017071084 A1 WO2017071084 A1 WO 2017071084A1 CN 2015099562 W CN2015099562 W CN 2015099562W WO 2017071084 A1 WO2017071084 A1 WO 2017071084A1
Authority
WO
WIPO (PCT)
Prior art keywords
moving target
specified
video
monitoring video
pixel
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/CN2015/099562
Other languages
English (en)
French (fr)
Inventor
张涛
陈志军
汪平仄
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.)
Xiaomi Inc
Original Assignee
Xiaomi 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 Xiaomi Inc filed Critical Xiaomi Inc
Priority to KR1020167014501A priority Critical patent/KR101825045B1/ko
Priority to MX2016004500A priority patent/MX361526B/es
Priority to JP2016535034A priority patent/JP2017537357A/ja
Priority to RU2016117335A priority patent/RU2629469C1/ru
Publication of WO2017071084A1 publication Critical patent/WO2017071084A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/0202—Child monitoring systems using a transmitter-receiver system carried by the parent and the child
    • G08B21/0205—Specific application combined with child monitoring using a transmitter-receiver system
    • G08B21/0208—Combination with audio or video communication, e.g. combination with "baby phone" function
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19606—Discriminating between target movement or movement in an area of interest and other non-signicative movements, e.g. target movements induced by camera shake or movements of pets, falling leaves, rotating fan
    • 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
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19654—Details concerning communication with a camera
    • G08B13/19656—Network used to communicate with a camera, e.g. WAN, LAN, Internet
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/0202—Child monitoring systems using a transmitter-receiver system carried by the parent and the child
    • G08B21/0205—Specific application combined with child monitoring using a transmitter-receiver system
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/18—Status alarms
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B3/00—Audible signalling systems, e.g. audible personal calling systems
    • G08B3/10—Audible signalling systems, e.g. audible personal calling systems using electric transmission; using electromagnetic transmission
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19654—Details concerning communication with a camera
    • G08B13/1966—Wireless systems, other than telephone systems, used to communicate with a camera

Definitions

  • the present disclosure relates to the field of Internet technologies, and in particular, to an alarm method and apparatus.
  • the camera can use the scene change detection technology to monitor the indoor real-time. If an abnormal scene change is detected, the alarm is performed.
  • the scene change detection technology is based on the image currently captured by the camera. Pre-set scene models to determine abnormal scene changes, so when some objects that do not pose danger to the user, such as pets walking indoors, the camera is also very easy to determine the pet's walking as abnormal scene changes, Furthermore, an error alarm is issued, which causes trouble to the user. Therefore, an alarm method for avoiding false alarms is urgently needed.
  • the embodiments of the present disclosure provide an alarm method and apparatus.
  • an alarm method comprising:
  • the moving target When the moving target is a designated object, it is prohibited to send an alarm message to the terminal, so that the terminal performs an alarm.
  • the detecting whether the moving target exists in the monitoring video includes:
  • the method further includes:
  • the specified background model is updated based on pixel values of each pixel in the video image.
  • the determining whether the moving target is a specified object includes:
  • the category is a specified category, it is determined that the moving target is a specified object.
  • the determining, according to the specified classification model, the category to which the moving target belongs includes:
  • a category to which the moving target belongs is determined based on the specified classification model and the processed target image.
  • the specified object includes a pet .
  • an alarm device comprising:
  • a detecting module configured to detect whether there is a moving target in the monitoring video
  • a determining module configured to determine whether the moving target is a designated object when there is a moving target in the monitoring video
  • the sending module is prohibited from being configured to prohibit sending the alarm information to the terminal when the moving target is the designated object.
  • the detecting module includes:
  • An acquiring unit configured to acquire, for each video image in the monitoring video, a pixel value of each pixel in the video image
  • a determining unit configured to determine whether there is a foreground pixel in the video image based on the pixel value of each pixel and the specified background model
  • the first determining unit is configured to: when there is a foreground pixel in the video image, determine that there is a moving target in the monitoring video; otherwise, determine that there is no moving target in the monitoring video.
  • the detecting module further includes:
  • an update unit configured to update the specified background model based on pixel values of each pixel in the video image.
  • the The broken module includes:
  • a second determining unit configured to determine, according to the specified classification model, a category to which the moving target belongs when there is a moving target in the monitoring video
  • a third determining unit configured to determine that the moving target is a specified object when the category is a specified category.
  • the second determining unit includes:
  • a cropping unit configured to: when a moving target exists in the monitoring video, crop the area where the moving target is located in the video image of the monitoring video to obtain a target image;
  • Processing the subunit configured to process the size of the target image to a preset size
  • the determining subunit is configured to determine a category to which the moving object belongs based on the specified classification model and the processed target image.
  • the specified object includes a pet.
  • an alarm device comprising:
  • a memory configured to store processor executable instructions
  • processor is configured to:
  • the server acquires the monitoring video and detects whether the monitoring video is saved.
  • the moving target when there is a moving target in the monitoring video, it is judged whether the moving target is a designated object, and when the moving target is the designated object, it is prohibited to send an alarm message to the terminal, thereby avoiding the false alarm caused by the specified object moving, and reducing the error.
  • the false alarm rate improves the accuracy of the alarm.
  • FIG. 1 is a schematic diagram of an implementation environment involved in an alarm method according to an exemplary embodiment
  • FIG. 2 is a flow chart showing an alarm method according to an exemplary embodiment
  • FIG. 3 is a flowchart of another alarm method according to an exemplary embodiment
  • FIG. 4 is a block diagram of an alarm device according to an exemplary embodiment
  • FIG. 5 is a block diagram of a detection module according to an exemplary embodiment
  • FIG. 6 is a block diagram of another detection module according to an exemplary embodiment
  • FIG. 7 is a block diagram of a determination module according to an exemplary embodiment
  • FIG. 8 is a block diagram of a second determining unit, according to an exemplary embodiment
  • FIG. 9 is a block diagram of another alarm device, according to an exemplary embodiment.
  • FIG. 1 is a schematic diagram of an implementation environment involved in an alarm method according to an exemplary embodiment.
  • the implementation environment may include a server 101, a smart camera device 102, and a terminal 103.
  • the server 101 can be a server, or a server cluster composed of several servers, or a cloud computing service center.
  • the smart camera device 102 can be a smart camera.
  • the terminal 103 can be a mobile phone, a computer, a tablet device, or the like.
  • the server 101 and the smart camera device 102 can be connected through a network, and the server 101 and the terminal 103 can also be connected through a network.
  • the server 101 is configured to receive the surveillance video transmitted by the smart camera device and send the alarm information to the terminal.
  • the smart camera device 102 is configured to collect monitoring video within the monitoring area and send the monitoring video to the server.
  • the terminal 103 is configured to receive alarm information sent by the server and perform an alarm.
  • FIG. 2 is a flowchart of an alarm method according to an exemplary embodiment. As shown in FIG. 2, the method is used in a server, and includes the following steps.
  • step 201 a surveillance video is obtained.
  • step 202 it is detected whether there is a moving target in the surveillance video.
  • step 203 when there is a moving target in the monitoring video, it is determined whether the moving target is a designated object.
  • step 204 when the moving target is the designated object, it is prohibited to transmit the alarm information to the terminal.
  • the server acquires the monitoring video, and detects whether there is a moving target in the monitoring video.
  • the server determines whether the moving target is a specified object, and when the moving target is a specified object, the terminal is prohibited.
  • Sending alarm information thus avoiding the false alarm caused by the specified object movement, reducing the false alarm rate and improving the alarm accuracy.
  • detecting whether there is a moving target in the monitoring video includes:
  • the specified background model is configured to characterize the distribution characteristics of the pixel values of each background pixel in the video image in the time domain, based on the pixel values of each pixel in the video image and the specified background model, the video image can be effectively determined. Whether there is a foreground pixel, that is, it can effectively judge whether there is a moving target in the monitoring video.
  • the method further includes:
  • the specified background model is updated based on the pixel values of each pixel in the video image.
  • real-time updating of the specified background model can make the specified background model adaptive, that is, the pixel value of the specified background model can be kept close to the current background pixel. Distribution characteristics in the time domain, thereby improving the accuracy of moving target detection.
  • determining whether the moving target is a specified object includes:
  • the motion target is determined to be the specified object.
  • the server may determine whether the moving target is a specified object, to avoid a false alarm caused by the specified object during the movement.
  • determining a category to which the moving target belongs based on the specified classification model includes:
  • the area where the moving target is located is cropped to obtain a target image
  • the category to which the moving target belongs is determined based on the specified classification model and the processed target image.
  • the server crops the area where the moving target is located, obtains the target image, and processes the size of the target image as
  • the preset size can facilitate the specified classification model to determine the category to which the moving target belongs based on the processed target image, and improve the efficiency of the category determination.
  • the specified object includes a pet.
  • the specified object includes a pet
  • the false alarm caused by the pet movement can be avoided, the false alarm rate is reduced, and the alarm accuracy is improved.
  • FIG. 3 is a flowchart of an alarm method according to an exemplary embodiment. As shown in FIG. 3, the method includes the following steps.
  • step 301 the server obtains a surveillance video.
  • the server can obtain the monitoring video from the smart camera device.
  • the smart camera device can also send the monitoring video to other devices, so that the server can obtain the monitoring video from the other device.
  • the disclosed embodiments do not specifically limit this.
  • the smart camera device is configured to collect the monitoring video in the monitoring area, and the process of the smart camera device collecting the monitoring video in the monitoring area may refer to related technologies, and the embodiments of the present disclosure are not described in detail herein.
  • the smart camera device can communicate with a server or other device through a wired network or a wireless network, and when the smart camera device communicates with a server or other device through a wireless network, the smart camera device can pass the built-in wireless fidelity (English: Wireless-Fidelity (WIFI), Bluetooth or other wireless communication chip to communicate with a server or other device, which is not specifically limited in the embodiment of the present disclosure.
  • WIFI Wireless-Fidelity
  • step 302 the server detects whether there is a moving target in the monitored video.
  • the smart camera device Since the smart camera device is generally fixed, that is, the smart camera device is a fixed monitoring area.
  • the monitoring video in the domain is collected.
  • a background model may be established in the background in the fixed monitoring area, so that each video image in the monitoring video can be compared with the background model.
  • the foreground image refers to an image of any meaningful moving object assuming the background is stationary.
  • the operation of the server to detect whether there is a moving target in the monitoring video may be: for each video image in the monitoring video, the server acquires the pixel value of each pixel in the video image; based on the pixel value and designation of each pixel
  • the background model determines whether there is a foreground pixel in the video image; when there is a foreground pixel in the video image, determining that there is a moving target in the monitoring video; otherwise, determining that there is no moving target in the monitoring video.
  • the background model is used to represent the distribution characteristics of the pixel values of each background pixel in the video image in the time domain, and the specified background model may be a mixed Gaussian model. Of course, the specified background model may also be other background models. The disclosed embodiments do not specifically limit this.
  • the specified background model may be pre-established, for example, the specified background model may be established according to the distribution of the pixel values of each pixel in the specified video image of the surveillance video in advance, and of course, the specified background may be established in other manners.
  • the model is not specifically limited in this embodiment.
  • the color feature is one of the essential features of the image
  • the color feature can be expressed as the pixel value of the pixel of the image, and the pixel value refers to the position, color, brightness, and the like of the pixel point of the image. Therefore, the server can be based on each of the video images.
  • the pixel value of the pixel and the specified background model determine whether there is a foreground pixel in the video image. When there is a foreground pixel in the video image, it indicates that there is a meaningful moving object in the video image, that is, there is a moving target in the monitoring video.
  • the server determines whether there is a foreground pixel in the video image, and the server may specify the pixel value of each pixel.
  • the background model is matched.
  • the pixel values of each pixel point are successfully matched with the specified background model, it is determined that there is no foreground pixel point in the video image. Otherwise, the foreground pixel point exists in the video image, and the foreground pixel point is determined.
  • the process of the server matching the specified background model based on the pixel value of each pixel may refer to the related art, which is not elaborated in this embodiment of the present disclosure.
  • the server may further update the specified background model based on the pixel value of each pixel in the video image.
  • the specified background model Since the specified background model is pre-established by the server, and due to uncontrollable factors such as illumination changes and camera shake, the background will change. Therefore, in order to avoid the accumulation of changes caused by the unmeasured factors, the specified background model is targeted to the moving target. The detection error occurs.
  • the server may update the specified background model in real time based on the pixel value of each pixel in the video image to make the specified background model adaptive. Sexuality can continuously keep close to the distribution characteristics of the pixel values of the current background pixels in the time domain, thereby improving the accuracy of moving target detection.
  • step 303 when there is a moving target in the monitoring video, the server determines whether the moving target is a designated object.
  • the server can determine whether the moving target is the designated object.
  • the operation of the server determining whether the moving target is the specified object may be: the server may determine the category to which the moving target belongs based on the specified classification model, and determine the moving target as the specified object when the moving target belongs to the specified category; otherwise, determine The moving target is not the specified object.
  • the specified object can be set in advance, and the specified object can be used without being used.
  • An object that brings danger to the household, such as a designated object, may include a pet.
  • the specified object may also include other items, which is not specifically limited in the embodiment of the present disclosure.
  • the specified category is a category to which the specified object belongs, and the specified category and the specified object are in one-to-one correspondence.
  • the specified category may include a cat, a dog, and the like, which is not specifically limited in the embodiment of the present disclosure.
  • the server determines, based on the specified classification model, that the category to which the moving target belongs is a cat, and if the specified category includes a cat and a dog, and therefore, the category to which the moving target belongs is the specified category, the server may determine that the moving target is the specified object.
  • the server determines, based on the specified classification model, that the category to which the moving target belongs is a person, and if the specified category includes a cat and a dog, and therefore, the category to which the moving target belongs is not the specified category, the server determines that the moving target is not the specified object.
  • the server determines, according to the specified classification model, the category to which the moving target belongs, the server may crop the area where the moving target is located in the video image of the monitoring video, obtain the target image, and process the size of the target image into a preset size. Then, based on the specified classification model and the processed target image, the category to which the moving target belongs is determined.
  • the server crops the region where the moving target is located, and when the target image is obtained, the circumscribed rectangle of the moving target may be intercepted from the video image in which the moving target is located, and the circumscribed rectangle is determined as the motion.
  • the target is the area of the image in which the video is being monitored, ie the target image.
  • the server may also acquire foreground pixels from the video image in which the moving target is located, and combine the acquired foreground pixels to obtain an image region in which the moving target is located in the monitoring video, that is, the target image.
  • the server may also clear the background pixel in the video image of the moving target, and obtain the image region where the moving target is located in the monitoring video, that is, the target image, wherein the background pixel is a pixel value that successfully matches the specified background model. Corresponding pixel points.
  • the specified classification model is used to determine the category corresponding to the image, and the specified points are specified.
  • the class model may distinguish a plurality of categories, the specified categories are included in the plurality of categories, and the specified classification model may be pre-established, and when the specified classification model is established, the server may obtain a sample image set corresponding to the plurality of categories in advance, wherein each The sample picture set corresponds to one category, and each sample picture in each sample picture set includes an item of a category corresponding to the sample picture set, and the server may process the size of the sample picture in the sample picture set to a preset size.
  • the server may also establish a specified classification model in other manners, which is not specifically limited in this embodiment of the present disclosure.
  • the specified classification model can be pre-established, and in order to improve the efficiency of the category determination, the specified classification model can generally process the image of the preset size to determine the category to which the object included in the image belongs, and therefore, the server determines the category to which the moving target belongs.
  • the area where the moving target is located needs to be cropped, the target image is obtained, and the size of the target image is processed to a preset size, so that the specified classification model determines the category to which the moving target belongs based on the processed target image.
  • the preset size may be preset, for example, the preset size may be 224*224 pixels, 300*300 pixels, and the like, which is not specifically limited in the embodiment of the present disclosure.
  • the server may extract the feature vector from the target image according to the specified classification model, and then according to the feature vector and the specified classification model,
  • the calculation target image belongs to the probability values of the plurality of categories included in the specified classification model, and the category having the largest probability value is determined as the category to which the moving target belongs.
  • step 304 when the moving target is the designated object, the server prohibits sending the alarm information to the terminal.
  • the terminal can connect to the server through a wired network or a wireless network, when the target is moving
  • the server prohibits sending the alarm information to the terminal, thereby avoiding the false alarm caused by the specified object movement, reducing the false alarm rate, and improving the alarm. Precision.
  • the server may further determine whether to send the alarm information to the terminal based on the specified alarm policy, so that the terminal alarms.
  • the specified alarm policy is used to indicate whether the alarm information is sent to the terminal when the moving target is not the specified object, and the specified alarm policy can be preset. For example, if the specified alarm policy can determine whether the moving target is a person, when the moving target is a person, Send an alarm message to the terminal.
  • the specified alarm policy may also include other policies, which are not specifically limited in this embodiment of the present disclosure.
  • the alarm information is used to remind the user of an abnormal moving target in the monitoring area.
  • the terminal when the terminal performs an alarm, the alarm information can be directly played through the speaker set on the terminal.
  • the terminal can also perform the alarm in other manners, which is not specifically limited in the embodiment of the present disclosure.
  • the server acquires the monitoring video, and detects whether there is a moving target in the monitoring video.
  • the server determines whether the moving target is a specified object, and when the moving target is a specified object, the terminal is prohibited.
  • Sending alarm information thus avoiding the false alarm caused by the specified object movement, reducing the false alarm rate and improving the alarm accuracy.
  • FIG. 4 is a block diagram of an alarm device, according to an exemplary embodiment.
  • the apparatus includes an acquisition module 401, a detection module 402, a determination module 403, and a transmission prohibition module 404.
  • the obtaining module 401 is configured to acquire a monitoring video.
  • the detecting module 402 is configured to detect whether there is a moving target in the monitoring video
  • the determining module 403 is configured to determine whether the moving target is a designated object when there is a moving target in the monitoring video;
  • the sending module 404 is configured to be configured to prohibit sending to the terminal when the moving target is the specified object Alarm information.
  • the detecting module 402 includes an obtaining unit 4021, a determining unit 4022, and a first determining unit 4023.
  • the obtaining unit 4021 is configured to acquire, for each video image in the monitoring video, a pixel value of each pixel in the video image;
  • the determining unit 4022 is configured to determine whether there is a foreground pixel point in the video image based on the pixel value of each pixel point and the specified background model;
  • the first determining unit 4023 is configured to determine that there is a moving target in the monitoring video when there is a foreground pixel in the video image, otherwise, it is determined that there is no moving target in the monitoring video.
  • the detection module 402 further includes an update unit 4024.
  • Update unit 4024 is configured to update the specified background model based on pixel values for each pixel in the video image.
  • the determining module 403 includes a second determining unit 4031, and a third determining unit 4032.
  • a second determining unit 4031 configured to determine, according to the specified classification model, a category to which the moving target belongs when there is a moving target in the monitoring video;
  • the third determining unit 4032 is configured to determine that the moving target is the specified object when the category is the specified category.
  • the second determining unit 4031 includes a cropping subunit 40311, a processing subunit 40312, and a determining subunit 40313.
  • the cropping subunit 40311 is configured to: when there is a moving target in the monitoring video, crop the area where the moving target is located in the video image of the monitoring video to obtain the target image;
  • Processing subunit 40312 configured to process a size of the target image to a preset size
  • Determining subunit 40313 configured to be based on the specified classification model and the processed target image, The category to which the moving target belongs.
  • the specified object includes a pet.
  • the server acquires the monitoring video, and detects whether there is a moving target in the monitoring video.
  • the server determines whether the moving target is a specified object, and when the moving target is a specified object, the terminal is prohibited.
  • Sending alarm information thus avoiding the false alarm caused by the specified object movement, reducing the false alarm rate and improving the alarm accuracy.
  • FIG. 9 is a block diagram of an apparatus 900 for alerting, according to an exemplary embodiment.
  • device 900 can be provided as a server.
  • apparatus 900 includes a processing component 922 that further includes one or more processors, and memory resources represented by memory 932, configured to store instructions executable by processing component 922, such as an application.
  • An application stored in memory 932 may include one or more modules each corresponding to a set of instructions.
  • Device 900 may also include a power supply component 926 configured to perform power management of device 900, a wired or wireless network interface 950 configured to connect device 900 to the network, and an input/output (I/O) interface 958.
  • Device 900 can operate based on an operating system stored in memory 932, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or the like.
  • processing component 922 is configured to execute instructions to perform the alerting method described below, the method comprising:
  • detecting whether there is a moving target in the monitoring video includes:
  • the method further includes:
  • the specified background model is updated based on the pixel values of each pixel in the video image.
  • determining whether the moving target is a specified object includes:
  • the motion target is determined to be the specified object.
  • determining a category to which the moving target belongs based on the specified classification model includes:
  • the area where the moving target is located is cropped to obtain a target image
  • the category to which the moving target belongs is determined based on the specified classification model and the processed target image.
  • the specified object includes a pet.
  • the server acquires the monitoring video, and detects whether there is a moving target in the monitoring video.
  • the server determines whether the moving target is a specified object, and when the moving target is a specified object, the terminal is prohibited.
  • Sending alarm information thus avoiding the false alarm caused by the specified object movement, reducing the false alarm rate and improving the alarm accuracy.
  • the server acquires the monitoring video, and detects whether there is a moving target in the monitoring video.
  • the server determines whether the moving target is a specified object, and when the moving target is a specified object, the terminal is prohibited.
  • Sending alarm information thus avoiding the false alarm caused by the specified object movement, reducing the false alarm rate and improving the alarm accuracy.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Child & Adolescent Psychology (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Multimedia (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Electromagnetism (AREA)
  • Signal Processing (AREA)
  • Alarm Systems (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Burglar Alarm Systems (AREA)

Abstract

一种报警方法及装置，属于互联网技术领域。报警方法包括：获取监控视频（201）；检测该监控视频中是否存在运动目标（202）；当该监控视频中存在运动目标时，判断该运动目标是否为指定对象（203）；当该运动目标是指定对象时，禁止向终端发送报警信息（204）。

Description

报警方法及装置
相关申请的交叉引用
本申请基于申请号为201510711332.5、申请日为2015年10月28日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本公开涉及互联网技术领域，尤其涉及一种报警方法及装置。
背景技术
随着摄像头的普及，利用摄像头进行实时监控越来越流行。而当用户不在家或者用户睡着时，摄像头可以利用场景变化检测技术对室内进行实时监控，如果检测到异常的场景变化，就进行报警，由于场景变化检测技术是基于摄像头当前采集到的图像和预先设置的场景模型，来确定异常的场景变化，所以，当某些不会给用户带来危险的对象，如宠物在室内走动时，摄像头也极易将宠物的走动确定为异常的场景变化，进而进行错误报警，给用户造成困扰，因此，亟需一种避免误报警的报警方法。
发明内容
为克服相关技术中存在的问题，本公开实施例提供一种报警方法及装置。
根据本公开实施例的第一方面，提供一种报警方法，所述方法包括：
获取监控视频；
检测所述监控视频中是否存在运动目标；
当所述监控视频中存在运动目标时，判断所述运动目标是否为指定对象；
当所述运动目标是指定对象时，禁止向终端发送报警信息，使所述终端进行报警。
结合第一方面，在上述第一方面的第一种可能的实现方式中，所述检测所述监控视频中是否存在运动目标，包括：
对于所述监控视频中的每帧视频图像，获取所述视频图像中每个像素点的像素值；
基于所述每个像素点的像素值和指定背景模型，判断所述视频图像中是否存在前景像素点；
当所述视频图像中存在前景像素点时，确定所述监控视频中存在运动目标，否则，确定所述监控视频中不存在运动目标。
结合第一方面的第一种可能的实现方式，在上述第一方面的第二种可能的实现方式中，所述确定所述监控视频中不存在运动目标之后，还包括：
基于所述视频图像中每个像素点的像素值，更新所述指定背景模型。
结合第一方面，在上述第一方面的第三种可能的实现方式中，所述判断所述运动目标是否为指定对象，包括：
基于指定分类模型，确定所述运动目标所属的类别；
当所述类别为指定类别时，确定所述运动目标为指定对象。
结合第一方面的第三种可能的实现方式，在上述第一方面的第四种可能的实现方式中，所述基于指定分类模型，确定所述运动目标所属的类别，包括：
在所述监控视频的视频图像中，对所述运动目标所在的区域进行裁剪，得到目标图像；
将所述目标图像的尺寸处理为预设尺寸；
基于指定分类模型和处理后的目标图像，确定所述运动目标所属的类别。
结合第一方面至第一方面的第四种可能的实现方式中的任一可能的实现方式，在上述第一方面的第五种可能的实现方式中，其特征在于，所述指定对象包括宠物。
根据本公开实施例的第二方面，提供一种报警装置，所述装置包括：
获取模块，配置为获取监控视频；
检测模块，配置为检测所述监控视频中是否存在运动目标；
判断模块，配置为当所述监控视频中存在运动目标时，判断所述运动目标是否为指定对象；
禁止发送模块，配置为当所述运动目标是指定对象时，禁止向终端发送报警信息。
结合第二方面，在上述第二方面的第一种可能的实现方式中，所述检测模块包括：
获取单元，配置为对于所述监控视频中的每帧视频图像，获取所述视频图像中每个像素点的像素值；
判断单元，配置为基于所述每个像素点的像素值和指定背景模型，判断所述视频图像中是否存在前景像素点；
第一确定单元，配置为当所述视频图像中存在前景像素点时，确定所述监控视频中存在运动目标，否则，确定所述监控视频中不存在运动目标。
结合第二方面的第一种可能的实现方式，在上述第二方面的第二种可能的实现方式中，所述检测模块还包括：
更新单元，配置为基于所述视频图像中每个像素点的像素值，更新所述指定背景模型。
结合第二方面，在上述第二方面的第三种可能的实现方式中，所述判 断模块包括：
第二确定单元，配置为当所述监控视频中存在运动目标时，基于指定分类模型，确定所述运动目标所属的类别；
第三确定单元，配置为当所述类别为指定类别时，确定所述运动目标为指定对象。
结合第二方面的第三种可能的实现方式，在上述第二方面的第四种可能的实现方式中，所述第二确定单元包括：
裁剪子单元，配置为当所述监控视频中存在运动目标时，在所述监控视频的视频图像中，对所述运动目标所在的区域进行裁剪，得到目标图像；
处理子单元，配置为将所述目标图像的尺寸处理为预设尺寸；
确定子单元，配置为基于指定分类模型和处理后的目标图像，确定所述运动目标所属的类别。
结合第二方面至第二方面的第四种可能的实现方式中的任一可能的实现方式，在上述第一方面的第五种可能的实现方式中，所述指定对象包括宠物。
根据本公开实施例的第三方面，提供一种报警装置，所述装置包括：
处理器；
配置为存储处理器可执行指令的存储器；
其中，所述处理器被配置为：
获取监控视频；
检测所述监控视频中是否存在运动目标；
当所述监控视频中存在运动目标时，判断所述运动目标是否为指定对象；
当所述运动目标是指定对象时，禁止向终端发送报警信息。
在本公开实施例中，服务器获取监控视频，并检测监控视频中是否存 在运动目标，当监控视频中存在运动目标时，判断运动目标是否为指定对象，当运动目标是指定对象时，禁止向终端发送报警信息，从而避免了指定对象运动时引起的错误报警，降低了误报警率，提高了报警精度。
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，并不能限制本公开。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本发明的实施例，并与说明书一起用于解释本发明的原理。
图1是根据一示例性实施例示出的一种报警方法所涉及的实施环境的示意图；
图2是根据一示例性实施例示出的一种报警方法的流程图；
图3是根据一示例性实施例示出的另一种报警方法的流程图；
图4是根据一示例性实施例示出的一种报警装置的框图；
图5是根据一示例性实施例示出的一种检测模块的框图；
图6是根据一示例性实施例示出的另一种检测模块的框图；
图7是根据一示例性实施例示出的一种判断模块的框图；
图8是根据一示例性实施例示出的一种第二确定单元的框图；
图9是根据一示例性实施例示出的另一种报警装置的框图。
具体实施方式
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本发明相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本发明的一些方面相一致的装置和方法的例子。
图1是根据一示例性实施例示出的一种报警方法所涉及的实施环境的示意图。如图1所示，该实施环境可以包括：服务器101、智能摄像设备102和终端103。服务器101可以是一台服务器，或者是由若干台服务器组成的服务器集群，或者是一个云计算服务中心，智能摄像设备102可以是智能摄像机，终端103可以是移动电话，计算机，平板设备等。服务器101和智能摄像设备102之间可以通过网络进行连接，服务器101与终端103之间也可以通过网络进行连接。服务器101配置为接收智能摄像设备发送的监控视频，并向终端发送报警信息。智能摄像设备102配置为采集监控区域内的监控视频，并将监控视频发送给服务器。终端103配置为接收服务器发送的报警信息，并进行报警。
图2是根据一示例性实施例示出的一种报警方法的流程图，如图2所示，该方法用于服务器中，包括以下步骤。
在步骤201中，获取监控视频。
在步骤202中，检测监控视频中是否存在运动目标。
在步骤203中，当监控视频中存在运动目标时，判断运动目标是否为指定对象。
在步骤204中，当运动目标是指定对象时，禁止向终端发送报警信息。
在本公开实施例中，服务器获取监控视频，并检测监控视频中是否存在运动目标，当监控视频中存在运动目标时，判断运动目标是否为指定对象，当运动目标是指定对象时，禁止向终端发送报警信息，从而避免了指定对象运动时引起的错误报警，降低了误报警率，提高了报警精度。
在本公开的另一实施例中，检测监控视频中是否存在运动目标，包括：
对于监控视频中的每帧视频图像，获取视频图像中每个像素点的像素值；
基于每个像素点的像素值和指定背景模型，判断视频图像中是否存在 前景像素点；
当视频图像中存在前景像素点时，确定监控视频中存在运动目标，否则，确定监控视频中不存在运动目标。
由于指定背景模型配置为表征视频图像中每个背景像素点的像素值在时域上的分布特征，因此，基于视频图像中每个像素点的像素值和指定背景模型，可以有效判断视频图像中是否存在前景像素点，也即是可以有效判断监控视频中是否存在运动目标。
在本公开的另一实施例中，确定监控视频中不存在运动目标之后，还包括：
基于视频图像中每个像素点的像素值，更新指定背景模型。
其中，基于视频图像中每个像素点的像素值，对指定背景模型进行实时更新，可以使指定背景模型具有自适应性，也即是可以使指定背景模型不断贴近当前背景像素点的像素值在时域上的分布特征，从而提高运动目标检测的准确性。
在本公开的另一实施例中，判断运动目标是否为指定对象，包括：
基于指定分类模型，确定运动目标所属的类别；
当类别为指定类别时，确定运动目标为指定对象。
其中，当服务器确定监控视频中存在运动目标后，服务器可以判断该运动目标是否为指定对象，以避免指定对象在运动时引起的误报警。
在本公开的另一实施例中，基于指定分类模型，确定运动目标所属的类别，包括：
在监控视频的视频图像中，对运动目标所在的区域进行裁剪，得到目标图像；
将目标图像的尺寸处理为预设尺寸；
基于指定分类模型和处理后的目标图像，确定运动目标所属的类别。
由于指定分类模型一般可以对预设尺寸的图像进行处理，以确定该图像包括的对象所属的类别，因此，服务器对运动目标所在的区域进行裁剪，得到目标图像，并将目标图像的尺寸处理为预设尺寸，可以便于指定分类模型基于处理后目标图像确定运动目标所属的类别，提高类别确定的效率。
在本公开的另一实施例中，指定对象包括宠物。
其中，指定对象包括宠物时，可以避免宠物运动时引起的误报警，降低了误报警率，提高了报警精度。
上述所有可选技术方案，均可按照任意结合形成本公开的可选实施例，本公开实施例对此不再一一赘述。
图3是根据一示例性实施例示出的一种报警方法的流程图，如图3所示，该方法包括以下步骤。
在步骤301中，服务器获取监控视频。
需要说明的是，服务器可以从智能摄像设备中获取该监控视频，当然，该智能摄像设备也可以将该监控视频发送到其它设备中，以使服务器可以从该其它设备中获取该监控视频，本公开实施例对此不做具体限定。
其中，智能摄像设备配置为采集监控区域内的监控视频，且智能摄像设备采集监控区域内的监控视频的过程可以参考相关技术，本公开实施例在此不进行详细阐述。
另外，智能摄像设备可以通过有线网络或者无线网络和服务器或者其它设备进行通信，而当智能摄像设备通过无线网络和服务器或者其它设备进行通信时，智能摄像设备可以通过内置的无线保真(英文：Wireless-Fidelity，简称：WIFI)、蓝牙或者其它无线通信芯片来和服务器或者其它设备进行通信，本公开实施例对此不做具体限定。
在步骤302中，服务器检测该监控视频中是否存在运动目标。
由于智能摄像设备一般是固定的，即该智能摄像设备是对固定监控区 域内的监控视频进行采集，则此时为了检测监控视频中是否存在运动目标，可以对该固定监控区域中的背景建立背景模型，从而可以将监控视频中的每帧视频图像和该背景模型进行比较，来确定该固定监控区域中的前景图像，前景图像是指在假设背景为静止的情况下的任何有意义的运动物体的图像。
因此，服务器检测监控视频中是否存在运动目标的操作可以为：对于监控视频中的每帧视频图像，服务器获取该视频图像中每个像素点的像素值；基于每个像素点的像素值和指定背景模型，判断该视频图像中是否存在前景像素点；当该视频图像中存在前景像素点时，确定该监控视频中存在运动目标，否则，确定该监控视频中不存在运动目标。
其中，指定背景模型用于表征视频图像中每个背景像素点的像素值在时域上的分布特征，且指定背景模型可以为混合高斯模型，当然，指定背景模型也可以为其他背景模型，本公开实施例对此不做具体限定。
另外，指定背景模型可以预先建立，如可以预先根据监控视频的指定视频图像中每个像素点的像素值在时域上的分布情况，建立指定背景模型，当然，也可以以其它方式建立指定背景模型，本公开实施例同样对此不做具体限定。
由于颜色特征是图像的本质特征之一，颜色特征可以表现为图像的像素点的像素值，像素值是指图像的像素点的位置、颜色、亮度等数值，因此，服务器可以基于视频图像中每个像素点的像素值和指定背景模型，判断该视频图像中是否存在前景像素点。而当该视频图像中存在前景像素点时，则表明该视频图像中存在有意义的运动物体，也即是监控视频中存在运动目标。
其中，服务器基于每个像素点的像素值和指定背景模型，判断该视频图像中是否存在前景像素点时，服务器可以将每个像素点的像素值与指定 背景模型进行匹配，当每个像素点的像素值均与指定背景模型匹配成功时，确定该视频图像中不存在前景像素点，否则，确定该视频图像中存在前景像素点，且该前景像素点为与指定背景模型匹配不成功的像素值对应的像素点。
另外，服务器基于每个像素点的像素值与指定背景模型进行匹配的过程可以参考相关技术，本公开实施例对此不进行详细阐述。
进一步地，服务器确定监控视频中不存在运动目标之后，服务器还可以基于该视频图像中每个像素点的像素值，更新指定背景模型。
由于指定背景模型是服务器预先建立的，且由于光照变化、摄像头抖动等不可测因素的影响，会使背景产生变化，因此，为了避免该不可测因素导致的变化累积，使指定背景模型对运动目标的检测出现误差，当服务器确定监控视频中不存在运动目标时，该服务器可以基于该视频图像中每个像素点的像素值，对该指定背景模型进行实时更新，以使指定背景模型具有自适应性，可以不断贴近当前背景像素点的像素值在时域上的分布特征，进而提高运动目标检测的准确性。
需要说明的是，服务器基于视频图像中每个像素点的像素值，更新指定背景模型的过程可以参考相关技术，本公开实施例在此不进行详细阐述。
在步骤303中，当监控视频中存在运动目标时，服务器判断该运动目标是否为指定对象。
为了避免指定对象在运动时引起的误报警，因此，当服务器确定监控视频中存在运动目标后，服务器可以判断该运动目标是否为指定对象。而服务器判断运动目标是否为指定对象的操作可以为：该服务器可以基于指定分类模型，确定运动目标所属的类别，当运动目标所属的类别为指定类别时，确定运动目标为指定对象，否则，确定运动目标不为指定对象。
需要说明的是，指定对象可以预先设置，且指定对象可以为不会给用 户带来危险的对象，如指定对象可以包括宠物。当然，实际应用中，该指定对象还可以包括其他物品，本公开实施例对此不做具体限定。
另外，指定类别为指定对象所属的类别，且指定类别和指定对象一一对应，如当指定对象包括宠物时，指定类别可以包括猫、狗等，本公开实施例对此不做具体限定。
例如，服务器基于指定分类模型，确定运动目标所属的类别为猫时，假如，指定类别包括猫和狗，因此，运动目标所属的类别为指定类别，则服务器可以确定运动目标为指定对象。
再例如，服务器基于指定分类模型，确定运动目标所属的类别为人时，假如，指定类别包括猫和狗，因此，运动目标所属的类别不是指定类别，则服务器确定运动目标不是指定对象。
其中，服务器基于指定分类模型，确定运动目标所属的类别时，服务器可以在监控视频的视频图像中，对运动目标所在的区域进行裁剪，得到目标图像，并将目标图像的尺寸处理为预设尺寸，之后，基于指定分类模型和处理后的目标图像，确定运动目标所属的类别。
其中，服务器在监控视频的视频图像中，对运动目标所在的区域进行裁剪，得到目标图像时，可以从运动目标所在的视频图像中，截取运动目标的外接矩形，并将该外接矩形确定为运动目标在监控视频中所处的图像区域，即目标图像。或者，服务器还可以从运动目标所在的视频图像中，获取前景像素点，并将获取的前景像素点进行组合，得到运动目标在监控视频中所处的图像区域，即目标图像。又或者，服务器还可以清除运动目标所在视频图像中的背景像素点，得到运动目标在监控视频中所处的图像区域，即目标图像，其中，背景像素点为与指定背景模型匹配成功的像素值对应的像素点。
需要说明的是，指定分类模型用于确定图像所对应的类别，且指定分 类模型可以区分多个类别，该多个类别中包括指定类别，且指定分类模型可以预先建立，而建立指定分类模型时，服务器可以预先获取该多个类别分别对应的样本图片集，其中，每个样本图片集对应一种类别，且每个样本图片集中的每个样本图片均包括该样本图片集所对应类别的物品，服务器可以将该样本图片集中的样本图片的尺寸处理为预设尺寸，并在处理之后，保持预设训练模型的特征层的参数不变，利用处理后样本图片集中的样本图片和预设训练协议对预设训练模型的全连接层中的参数进行部分调整，对预设训练模型的分类器层中的参数进行全部调整，进而得到指定分类模型。当然，服务器还可以以其它方式建立指定分类模型，本公开实施例对此不做具体限定。
由于指定分类模型可以预先建立，且为了提高类别确定的效率，指定分类模型一般可以对预设尺寸的图像进行处理，以确定该图像包括的对象所属的类别，因此，服务器确定运动目标所属的类别之前，需要对运动目标所在的区域进行裁剪，得到目标图像，并将目标图像的尺寸处理为预设尺寸，以便于指定分类模型基于处理后目标图像确定运动目标所属的类别。
另外，预设尺寸可以预先设置，如预设尺寸可以为224*224像素、300*300像素等等，本公开实施例对此不做具体限定。
需要说明的是，服务器基于指定分类模型和处理后的目标图像，确定运动目标所属的类别时，服务器可以根据指定分类模型，从目标图像中抽取特征向量，之后根据该特征向量和指定分类模型，计算目标图像属于该指定分类模型包括的多个类别的概率值，并将概率值最大的类别确定为运动目标所属的类别。
在步骤304中，当运动目标是指定对象时，服务器禁止向终端发送报警信息。
终端可以通过有线网络或者无线网络与服务器进行连接，当运动目标 是指定对象时，也即是，运动目标为不会给用户带来危险的对象时，服务器禁止向终端发送报警信息，从而可以避免指定对象运动时引起的错误报警，降低误报警率，提高报警精度。
需要说明的是，当服务器确定运动目标不是指定对象时，服务器还可以基于指定报警策略，判断是否向终端发送报警信息，以使终端报警。
其中，指定报警策略用于指示当运动目标不是指定对象时是否向终端发送报警信息，且指定报警策略可以预先设置，如指定报警策略可以为判断该运动目标是否为人，当该运动目标为人时，向该终端发送报警信息。当然，实际应用中，指定报警策略还可以包括其他的策略，本公开实施例对此不做具体限定。
另外，报警信息用于提醒用户监控区域内出现异常运动目标。
再者，终端进行报警时，可以通过终端上设置的扬声器直接播放报警信息，当然，终端也可以通过其它方式进行报警，本公开实施例对此不做具体限定。
在本公开实施例中，服务器获取监控视频，并检测监控视频中是否存在运动目标，当监控视频中存在运动目标时，判断运动目标是否为指定对象，当运动目标是指定对象时，禁止向终端发送报警信息，从而避免了指定对象运动时引起的错误报警，降低了误报警率，提高了报警精度。
图4是根据一示例性实施例示出的一种报警装置的框图。参照图4，该装置包括获取模块401，检测模块402，判断模块403，禁止发送模块404。
获取模块401，配置为获取监控视频；
检测模块402，配置为检测监控视频中是否存在运动目标；
判断模块403，配置为当监控视频中存在运动目标时，判断运动目标是否为指定对象；
禁止发送模块404，配置为当运动目标是指定对象时，禁止向终端发送 报警信息。
在本公开的另一实施例中，参照图5，该检测模块402包括获取单元4021，判断单元4022，第一确定单元4023。
获取单元4021，配置为对于监控视频中的每帧视频图像，获取视频图像中每个像素点的像素值；
判断单元4022，配置为基于每个像素点的像素值和指定背景模型，判断视频图像中是否存在前景像素点；
第一确定单元4023，配置为当视频图像中存在前景像素点时，确定监控视频中存在运动目标，否则，确定监控视频中不存在运动目标。
在本公开的另一实施例中，参照图6，该检测模块402还包括更新单元4024。
更新单元4024，配置为基于视频图像中每个像素点的像素值，更新指定背景模型。
在本公开的另一实施例中，参照图7，该判断模块403包括第二确定单元4031，第三确定单元4032。
第二确定单元4031，配置为当监控视频中存在运动目标时，基于指定分类模型，确定运动目标所属的类别；
第三确定单元4032，配置为当类别为指定类别时，确定运动目标为指定对象。
在本公开的另一实施例中，参照图8，该第二确定单元4031包括裁剪子单元40311，处理子单元40312，确定子单元40313。
裁剪子单元40311，配置为当监控视频中存在运动目标时，在监控视频的视频图像中，对运动目标所在的区域进行裁剪，得到目标图像；
处理子单元40312，配置为将目标图像的尺寸处理为预设尺寸；
确定子单元40313，配置为基于指定分类模型和处理后的目标图像，确 定运动目标所属的类别。
在本公开的另一实施例中，指定对象包括宠物。
在本公开实施例中，服务器获取监控视频，并检测监控视频中是否存在运动目标，当监控视频中存在运动目标时，判断运动目标是否为指定对象，当运动目标是指定对象时，禁止向终端发送报警信息，从而避免了指定对象运动时引起的错误报警，降低了误报警率，提高了报警精度。
关于上述实施例中的装置，其中各个模块执行操作的具体方式已经在有关该方法的实施例中进行了详细描述，此处将不做详细阐述说明。
图9是根据一示例性实施例示出的一种用于报警的装置900的框图。例如，装置900可以被提供为一服务器。参照图9，装置900包括处理组件922，其进一步包括一个或多个处理器，以及由存储器932所代表的存储器资源，配置为存储可由处理组件922的执行的指令，例如应用程序。存储器932中存储的应用程序可以包括一个或一个以上的每一个对应于一组指令的模块。
装置900还可以包括一个电源组件926被配置为执行装置900的电源管理，一个有线或无线网络接口950被配置为将装置900连接到网络，和一个输入输出(I/O)接口958。装置900可以操作基于存储在存储器932的操作系统，例如Windows ServerTM，Mac OS XTM，UnixTM,LinuxTM，FreeBSDTM或类似。
此外，处理组件922被配置为执行指令，以执行下述报警方法，所述方法包括：
获取监控视频。
检测监控视频中是否存在运动目标。
当监控视频中存在运动目标时，判断运动目标是否为指定对象。
当运动目标是指定对象时，禁止向终端发送报警信息。
在本公开的另一实施例中，检测监控视频中是否存在运动目标，包括：
对于监控视频中的每帧视频图像，获取视频图像中每个像素点的像素值；
基于每个像素点的像素值和指定背景模型，判断视频图像中是否存在前景像素点；
当视频图像中存在前景像素点时，确定监控视频中存在运动目标，否则，确定监控视频中不存在运动目标。
在本公开的另一实施例中，确定监控视频中不存在运动目标之后，还包括：
基于视频图像中每个像素点的像素值，更新指定背景模型。
在本公开的另一实施例中，判断运动目标是否为指定对象，包括：
基于指定分类模型，确定运动目标所属的类别；
当类别为指定类别时，确定运动目标为指定对象。
在本公开的另一实施例中，基于指定分类模型，确定运动目标所属的类别，包括：
在监控视频的视频图像中，对运动目标所在的区域进行裁剪，得到目标图像；
将目标图像的尺寸处理为预设尺寸；
基于指定分类模型和处理后的目标图像，确定运动目标所属的类别。
在本公开的另一实施例中，指定对象包括宠物。
在本公开实施例中，服务器获取监控视频，并检测监控视频中是否存在运动目标，当监控视频中存在运动目标时，判断运动目标是否为指定对象，当运动目标是指定对象时，禁止向终端发送报警信息，从而避免了指定对象运动时引起的错误报警，降低了误报警率，提高了报警精度。
本领域技术人员在考虑说明书及实践这里公开的发明后，将容易想到 本发明的其它实施方案。本申请旨在涵盖本发明的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本发明的一般性原理并包括本公开实施例未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本发明的真正范围和精神由下面的权利要求指出。
应当理解的是，本发明并不局限于上面已经描述并在附图中示出的精确结构，并且可以在不脱离其范围进行各种修改和改变。本发明的范围仅由所附的权利要求来限制。
工业实用性
在本公开实施例中，服务器获取监控视频，并检测监控视频中是否存在运动目标，当监控视频中存在运动目标时，判断运动目标是否为指定对象，当运动目标是指定对象时，禁止向终端发送报警信息，从而避免了指定对象运动时引起的错误报警，降低了误报警率，提高了报警精度。

Claims (13)

  1. 一种报警方法，所述方法包括：
    获取监控视频；
    检测所述监控视频中是否存在运动目标；
    当所述监控视频中存在运动目标时，判断所述运动目标是否为指定对象；
    当所述运动目标是指定对象时，禁止向终端发送报警信息。
  2. 如权利要求1所述的方法，其中，所述检测所述监控视频中是否存在运动目标，包括：
    对于所述监控视频中的每帧视频图像，获取所述视频图像中每个像素点的像素值；
    基于所述每个像素点的像素值和指定背景模型，判断所述视频图像中是否存在前景像素点；
    当所述视频图像中存在前景像素点时，确定所述监控视频中存在运动目标，否则，确定所述监控视频中不存在运动目标。
  3. 如权利要求2所述的方法，其中，所述确定所述监控视频中不存在运动目标之后，还包括：
    基于所述视频图像中每个像素点的像素值，更新所述指定背景模型。
  4. 如权利要求1所述的方法，其中，所述判断所述运动目标是否为指定对象，包括：
    基于指定分类模型，确定所述运动目标所属的类别；
    当所述类别为指定类别时，确定所述运动目标为指定对象。
  5. 如权利要求4所述的方法，其中，所述基于指定分类模型，确定所述运动目标所属的类别，包括：
    在所述监控视频的视频图像中，对所述运动目标所在的区域进行裁 剪，得到目标图像；
    将所述目标图像的尺寸处理为预设尺寸；
    基于指定分类模型和处理后的目标图像，确定所述运动目标所属的类别。
  6. 如权利要求1-5任一权利要求所述的方法，其中，所述指定对象包括宠物。
  7. 一种报警装置，所述装置包括：
    获取模块，配置为获取监控视频；
    检测模块，配置为检测所述监控视频中是否存在运动目标；
    判断模块，配置为当所述监控视频中存在运动目标时，判断所述运动目标是否为指定对象；
    禁止发送模块，配置为当所述运动目标是指定对象时，禁止向终端发送报警信息。
  8. 如权利要求7所述的装置，其中，所述检测模块包括：
    获取单元，配置为对于所述监控视频中的每帧视频图像，获取所述视频图像中每个像素点的像素值；
    判断单元，配置为基于所述每个像素点的像素值和指定背景模型，判断所述视频图像中是否存在前景像素点；
    第一确定单元，配置为当所述视频图像中存在前景像素点时，确定所述监控视频中存在运动目标，否则，确定所述监控视频中不存在运动目标。
  9. 如权利要求8所述的装置，其中，所述检测模块还包括：
    更新单元，配置为基于所述视频图像中每个像素点的像素值，更新所述指定背景模型。
  10. 如权利要求7所述的装置，其中，所述判断模块包括：
    第二确定单元，配置为当所述监控视频中存在运动目标时，基于指定分类模型，确定所述运动目标所属的类别；
    第三确定单元，配置为当所述类别为指定类别时，确定所述运动目标为指定对象。
  11. 如权利要求10所述的装置，其中，所述第二确定单元包括：
    裁剪子单元，配置为当所述监控视频中存在运动目标时，在所述监控视频的视频图像中，对所述运动目标所在的区域进行裁剪，得到目标图像；
    处理子单元，配置为将所述目标图像的尺寸处理为预设尺寸；
    确定子单元，配置为基于指定分类模型和处理后的目标图像，确定所述运动目标所属的类别。
  12. 如权利要求7-11任一权利要求所述的装置，其中，所述指定对象包括宠物。
  13. 一种报警装置，所述装置包括：
    处理器；
    配置为存储处理器可执行指令的存储器；
    其中，所述处理器被配置为：
    获取监控视频；
    检测所述监控视频中是否存在运动目标；
    当所述监控视频中存在运动目标时，判断所述运动目标是否为指定对象；
    当所述运动目标是指定对象时，禁止向终端发送报警信息。
PCT/CN2015/099562 2015-10-28 2015-12-29 报警方法及装置 Ceased WO2017071084A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
KR1020167014501A KR101825045B1 (ko) 2015-10-28 2015-12-29 경보 방법 및 장치
MX2016004500A MX361526B (es) 2015-10-28 2015-12-29 Método y dispositivo de alarma.
JP2016535034A JP2017537357A (ja) 2015-10-28 2015-12-29 警報方法および装置
RU2016117335A RU2629469C1 (ru) 2015-10-28 2015-12-29 Способ и устройство тревожной сигнализации

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510711332.5A CN105336074A (zh) 2015-10-28 2015-10-28 报警方法及装置
CN201510711332.5 2015-10-28

Publications (1)

Publication Number Publication Date
WO2017071084A1 true WO2017071084A1 (zh) 2017-05-04

Family

ID=55286571

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/099562 Ceased WO2017071084A1 (zh) 2015-10-28 2015-12-29 报警方法及装置

Country Status (8)

Country Link
US (1) US10147288B2 (zh)
EP (1) EP3163543B1 (zh)
JP (1) JP2017537357A (zh)
KR (1) KR101825045B1 (zh)
CN (1) CN105336074A (zh)
MX (1) MX361526B (zh)
RU (1) RU2629469C1 (zh)
WO (1) WO2017071084A1 (zh)

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9544636B2 (en) * 2014-07-07 2017-01-10 Google Inc. Method and system for editing event categories
US10140827B2 (en) 2014-07-07 2018-11-27 Google Llc Method and system for processing motion event notifications
US9361011B1 (en) 2015-06-14 2016-06-07 Google Inc. Methods and systems for presenting multiple live video feeds in a user interface
US10506237B1 (en) 2016-05-27 2019-12-10 Google Llc Methods and devices for dynamic adaptation of encoding bitrate for video streaming
US10957171B2 (en) 2016-07-11 2021-03-23 Google Llc Methods and systems for providing event alerts
CN106503666A (zh) * 2016-10-26 2017-03-15 珠海格力电器股份有限公司 一种安全监控方法、装置及电子设备
CN107040768A (zh) * 2017-05-18 2017-08-11 深圳市海和高新技术有限公司 智能家居的监控实现方法及系统
CN107845103B (zh) * 2017-10-23 2022-03-11 广东美的制冷设备有限公司 基于纹理信息的报警方法、装置以及计算机可读存储介质
CN107977638B (zh) * 2017-12-11 2020-05-26 智美达（江苏）数字技术有限公司 视频监控报警方法、装置、计算机设备和存储介质
CN108198203B (zh) * 2018-01-30 2022-02-08 广东美的制冷设备有限公司 运动报警方法、装置以及计算机可读存储介质
CN110110111B (zh) * 2018-02-02 2021-12-31 兴业数字金融服务（上海）股份有限公司 用于监控屏幕的方法和装置
CN108920995A (zh) * 2018-04-08 2018-11-30 华中科技大学 智能安防视频监控方法及其系统及监控终端
CN109359620A (zh) * 2018-10-31 2019-02-19 银河水滴科技（北京）有限公司 一种识别可疑对象的方法及装置
CN110147752A (zh) * 2019-05-15 2019-08-20 浙江大华技术股份有限公司 运动检测处理方法、装置、电子设备和存储介质
CN111753609B (zh) * 2019-08-02 2023-12-26 杭州海康威视数字技术股份有限公司 一种目标识别的方法、装置及摄像机
CN110927731B (zh) * 2019-11-15 2021-12-17 深圳市镭神智能系统有限公司 一种立体防护方法、三维检测装置和计算机可读存储介质
CN111263114B (zh) * 2020-02-14 2022-06-17 北京百度网讯科技有限公司 异常事件报警方法和装置
CN112257569B (zh) * 2020-10-21 2021-11-19 青海城市云大数据技术有限公司 一种基于实时视频流的目标检测和识别方法
CN113920339A (zh) * 2021-12-15 2022-01-11 智洋创新科技股份有限公司 一种基于深度学习的输电线路通道隐患中吊车误告警方法
CN114513608A (zh) * 2022-02-21 2022-05-17 深圳市美科星通信技术有限公司 移动侦测方法、装置及电子设备
CN115641528B (zh) * 2022-09-16 2025-12-09 杭州华橙软件技术有限公司 一种视频检测方法、设备及系统
CN116612414A (zh) * 2023-05-25 2023-08-18 中国农业银行股份有限公司 视频检测方法、系统、电子设备及存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0993665A (ja) * 1995-09-26 1997-04-04 Meidensha Corp 監視装置
CN101572803A (zh) * 2009-06-18 2009-11-04 中国科学技术大学 基于视频监控的可定制式自动跟踪系统
CN102479416A (zh) * 2010-11-29 2012-05-30 上海银晨智能识别科技有限公司 消除监控系统中误报警的方法、系统及装置
WO2012124852A1 (ko) * 2011-03-14 2012-09-20 (주)아이티엑스시큐리티 감시구역 상의 객체의 경로를 추적할 수 있는 스테레오 카메라 장치, 그를 이용한 감시시스템 및 방법
CN103516955A (zh) * 2012-06-26 2014-01-15 郑州大学 视频监控中的入侵检测方法
CN103581620A (zh) * 2012-07-26 2014-02-12 索尼公司 图像处理设备、图像处理方法及程序
CN104392464A (zh) * 2014-09-30 2015-03-04 天津艾思科尔科技有限公司 一种基于彩色视频图像的人为入侵检测方法
CN104700532A (zh) * 2013-12-11 2015-06-10 杭州海康威视数字技术股份有限公司 一种视频报警方法和装置

Family Cites Families (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6985172B1 (en) * 1995-12-01 2006-01-10 Southwest Research Institute Model-based incident detection system with motion classification
US6097429A (en) * 1997-08-01 2000-08-01 Esco Electronics Corporation Site control unit for video security system
EP0977437A3 (en) * 1998-07-28 2007-11-21 Hitachi Denshi Kabushiki Kaisha Method of distinguishing a moving object and apparatus of tracking and monitoring a moving object
RU2146837C1 (ru) * 1998-11-30 2000-03-20 Таланов Борис Петрович Сигнальное устройство нарушения стабильной обстановки
US20050162515A1 (en) * 2000-10-24 2005-07-28 Objectvideo, Inc. Video surveillance system
JP2004295798A (ja) * 2003-03-28 2004-10-21 Japan Best Rescue System Kk 警備システム
EP2090106A4 (en) * 2006-12-05 2010-10-27 Fujifilm Corp OUTPUT DEVICE, OUTPUT METHOD AND PROGRAM
KR101036947B1 (ko) * 2008-04-30 2011-05-25 현운혁 컴퓨터 영상 분석기술을 이용한 범죄 및 사고예방 자동경비 시스템
CN101635835A (zh) * 2008-07-25 2010-01-27 深圳市信义科技有限公司 智能视频监控方法及系统
JP5047361B2 (ja) * 2008-08-28 2012-10-10 有限会社 ラムロック映像技術研究所 監視システム
CN101686338B (zh) * 2008-09-26 2013-12-25 索尼株式会社 分割视频中的前景和背景的系统和方法
CN201662861U (zh) * 2010-04-23 2010-12-01 泉州市科立信安防电子有限公司 入侵探测设备
CN102142179A (zh) * 2010-12-20 2011-08-03 闫凯锋 基于视觉的智能安防系统
KR101237970B1 (ko) 2011-01-17 2013-02-28 포항공과대학교 산학협력단 영상 감시 시스템, 및 이의 방치 및 도난 검출 방법
CN102811343B (zh) * 2011-06-03 2015-04-29 南京理工大学 一种基于行为识别的智能视频监控系统
US8681223B2 (en) * 2011-06-24 2014-03-25 Honeywell International Inc. Video motion detection, analysis and threat detection device and method
KR101394242B1 (ko) * 2011-09-23 2014-05-27 광주과학기술원 영상 감시 장치 및 영상 감시 방법
JP6046948B2 (ja) * 2012-08-22 2016-12-21 キヤノン株式会社 物体検知装置及びその制御方法、プログラム、並びに記憶媒体
CN103731598B (zh) * 2012-10-12 2017-08-11 中兴通讯股份有限公司 一种智能监控终端及视频监控方法
KR20150043818A (ko) * 2013-10-15 2015-04-23 삼성전자주식회사 영상처리장치 및 그 제어방법
US11120478B2 (en) * 2015-01-12 2021-09-14 Ebay Inc. Joint-based item recognition
CN104658152B (zh) * 2015-02-15 2017-10-20 西安交通大学 一种基于视频的运动物体入侵报警方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0993665A (ja) * 1995-09-26 1997-04-04 Meidensha Corp 監視装置
CN101572803A (zh) * 2009-06-18 2009-11-04 中国科学技术大学 基于视频监控的可定制式自动跟踪系统
CN102479416A (zh) * 2010-11-29 2012-05-30 上海银晨智能识别科技有限公司 消除监控系统中误报警的方法、系统及装置
WO2012124852A1 (ko) * 2011-03-14 2012-09-20 (주)아이티엑스시큐리티 감시구역 상의 객체의 경로를 추적할 수 있는 스테레오 카메라 장치, 그를 이용한 감시시스템 및 방법
CN103516955A (zh) * 2012-06-26 2014-01-15 郑州大学 视频监控中的入侵检测方法
CN103581620A (zh) * 2012-07-26 2014-02-12 索尼公司 图像处理设备、图像处理方法及程序
CN104700532A (zh) * 2013-12-11 2015-06-10 杭州海康威视数字技术股份有限公司 一种视频报警方法和装置
CN104392464A (zh) * 2014-09-30 2015-03-04 天津艾思科尔科技有限公司 一种基于彩色视频图像的人为入侵检测方法

Also Published As

Publication number Publication date
EP3163543B1 (en) 2018-12-26
RU2629469C1 (ru) 2017-08-29
US20170124821A1 (en) 2017-05-04
MX361526B (es) 2018-12-07
KR101825045B1 (ko) 2018-02-02
CN105336074A (zh) 2016-02-17
US10147288B2 (en) 2018-12-04
MX2016004500A (es) 2017-06-09
EP3163543A1 (en) 2017-05-03
JP2017537357A (ja) 2017-12-14

Similar Documents

Publication Publication Date Title
WO2017071084A1 (zh) 报警方法及装置
US9953506B2 (en) Alarming method and device
US11875467B2 (en) Processing method for combining a real-world environment with virtual information according to a video frame difference value to provide an augmented reality scene, terminal device, system, and computer storage medium
JP6132452B2 (ja) 無線通信装置及び無線通信システム
US20160189501A1 (en) Security monitoring system and corresponding alarm triggering method
JP6704979B1 (ja) 無人飛行装置、無人飛行システム及び無人飛行装置制御システム
WO2017071086A1 (zh) 用于视频播放的方法及装置
TW201516939A (zh) 查詢使用者標識的方法及裝置、獲取使用者標識的方法及裝置與即時通訊中添加好友的方法及裝置
EP3573022A1 (en) Method for tracking pedestrian and electronic device
CN110834327A (zh) 一种机器人的控制方法及设备
CN110400346A (zh) 基于智能家居操作系统的信息处理方法及装置、存储介质
JP2017511983A (ja) 画像処理システム、リモコン撮影モジュール、移動端末および露出情報提示方法
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
CN109960969A (zh) 移动路线生成的方法、装置及系统
US12063451B2 (en) Modification of camera functionality based on orientation
CN111813983A (zh) 一种异常体温预警方法、装置及系统
CN104616408A (zh) 门铃控制方法及装置
CN113011222B (zh) 一种活体检测系统、方法及电子设备
CN105227914A (zh) 基于云服务器的生物特征信息的识别方法、装置及系统
CN114935405A (zh) 一种恒温动物监测方法及恒温动物监测设备
JP7207586B2 (ja) 撮像制御装置及び撮像システム
CN117111887A (zh) 一种跌倒检测方法和相关装置
JP7124411B2 (ja) 撮像制御装置及び撮像システム
JP2017135644A (ja) 無線装置、アクセス規制方法及びプログラム
Shankar et al. Transfer Inference Learning

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: MX/A/2016/004500

Country of ref document: MX

ENP Entry into the national phase

Ref document number: 2016117335

Country of ref document: RU

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2016535034

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20167014501

Country of ref document: KR

Kind code of ref document: A

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

Ref document number: 15907145

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15907145

Country of ref document: EP

Kind code of ref document: A1