WO2018223955A1 - 目标监控方法、目标监控装置、摄像机及计算机可读介质 - Google Patents

目标监控方法、目标监控装置、摄像机及计算机可读介质 Download PDF

Info

Publication number
WO2018223955A1
WO2018223955A1 PCT/CN2018/089945 CN2018089945W WO2018223955A1 WO 2018223955 A1 WO2018223955 A1 WO 2018223955A1 CN 2018089945 W CN2018089945 W CN 2018089945W WO 2018223955 A1 WO2018223955 A1 WO 2018223955A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
monitored
camera
current
monitoring
Prior art date
Application number
PCT/CN2018/089945
Other languages
English (en)
French (fr)
Inventor
陆卫国
肖可伟
李哲
张赟龙
陈瑞军
Original Assignee
北京深瞐科技有限公司
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 北京深瞐科技有限公司 filed Critical 北京深瞐科技有限公司
Publication of WO2018223955A1 publication Critical patent/WO2018223955A1/zh

Links

Images

Classifications

    • 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; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • 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

Definitions

  • the present application relates to the field of security monitoring technologies, and in particular, to a target monitoring method, a target monitoring device, a video camera, and a computer readable medium.
  • PTZ Pan Tilt Zoom, pan tilt zoom
  • the PTZ camera has a pan/tilt function, which can realize pan/tilt full-scale (left/right/up and down) movement, lens zoom and zoom control.
  • the control of the PTZ camera is usually manually performed by the back-end operator according to the video file transmitted by the PTZ camera through the console to achieve close-up monitoring and tracking of specific monitoring targets, and the operator finds the monitoring tracking.
  • the goal is to take action when alerting the target, such as an alarm.
  • people will be fatigued, and there will be leakage prevention problems due to lack of concentration.
  • there is a problem that the manual control of the PTZ camera for target monitoring has poor real-time performance.
  • the existing target monitoring methods have technical problems that waste labor costs, poor real-time performance, and are prone to leakage prevention.
  • the purpose of the present application is to provide a target monitoring method, a target monitoring device, and a camera to reduce the cost input in the target monitoring process and improve the timeliness and reliability of the target monitoring.
  • an embodiment of the present application provides a target monitoring method, where the method is applied to a camera, including:
  • Controlling by the camera, tracking and shooting the target to be monitored, and obtaining a close-up image of the target to be monitored;
  • the target to be monitored is determined as a monitoring target, and the monitoring target is tracked and monitored.
  • the embodiment of the present application provides the first possible implementation manner of the first aspect, wherein the object in the current frame image is classified, and the suspect object corresponding to the current category of the target object is obtained, including :
  • Sorting objects in the current frame image according to a preset classification rule including classifying objects according to characters, vehicles, and animals;
  • the embodiment of the present application provides a second possible implementation manner of the first aspect, wherein the determining, in the current frame image, the acquired target object according to the current deployment target Targets to be monitored, including:
  • the suspected object with the highest similarity is determined as the target to be monitored.
  • the embodiment of the present application provides a third possible implementation manner of the first aspect, wherein the controlling the camera to perform tracking and shooting on the target to be monitored, and obtaining a close-up image of the target to be monitored ,include:
  • the embodiment of the present application provides a fourth possible implementation manner of the first aspect, wherein the feature of the close-up image is identified according to a feature of the current deployed target, and the target to be monitored is determined. Whether it belongs to the current deployment target, including:
  • the embodiment of the present application provides a fifth possible implementation manner of the first aspect, wherein the method further includes:
  • the alarm information is sent to the monitoring terminal.
  • the embodiment of the present application provides a sixth possible implementation manner of the first aspect, wherein the method further includes:
  • the object in the alert target information is determined as the current deployed target.
  • the embodiment of the present application provides a seventh possible implementation manner of the first aspect, wherein the method further includes:
  • the object in the alert target information is determined as the current deployment target.
  • the embodiment of the present application further provides a target monitoring apparatus, including:
  • An image acquisition module configured to acquire a current frame image acquired by a camera of the camera
  • An object obtaining module configured to classify objects in the current frame image, and obtain a suspect object corresponding to a current category of the control target;
  • a target to be monitored module configured to determine, according to the current deployment target, the target to be monitored in the current frame image from the acquired suspect objects
  • a tracking shooting module configured to control the camera to perform tracking shooting on the target to be monitored, to obtain a close-up image of the target to be monitored
  • a determining module configured to perform feature recognition on the close-up image according to a feature of the current deployed target, and determine whether the target to be monitored belongs to the current deployed target;
  • the monitoring module is configured to determine the target to be monitored as a monitoring target when the determination result of the determining module is YES, and track and monitor the monitoring target.
  • the embodiment of the present application provides a first possible implementation manner of the second aspect, where the object obtaining module is specifically configured to:
  • Sorting objects in the current frame image according to a preset classification rule including classifying objects according to characters, vehicles, and animals;
  • the embodiment of the present application provides the second possible implementation manner of the second aspect, where the target to be monitored module is specifically configured as:
  • the suspected object with the highest similarity is determined as the target to be monitored.
  • the embodiment of the present application provides a third possible implementation manner of the second aspect, where the tracking shooting module is specifically configured to:
  • the embodiment of the present application provides a fourth possible implementation manner of the second aspect, where the determining module is specifically configured to:
  • the embodiment of the present application provides a fifth possible implementation manner of the second aspect, where the apparatus further includes:
  • the alarm information sending module is configured to send an alarm message to the monitoring terminal when determining that the target to be monitored is the current deployed target.
  • the embodiment of the present application provides a sixth possible implementation manner of the second aspect, where the apparatus further includes:
  • the deployment control target determining module is configured to determine, as the current deployment target, the object in the alert target information when receiving the alert target information sent by the monitoring terminal.
  • the embodiment of the present application further provides a camera, including a cloud platform, a PTZ motor connected to the PTZ, a camera disposed on the PTZ, a memory, a processor, and the storage on the memory. And a computer program operable on the processor;
  • the pan/tilt motor and the camera are respectively electrically connected to the processor, and when the processor executes the computer program, the steps of the method described in the first aspect are implemented, wherein the processor controls the The pan-tilt motor drives the camera to rotate.
  • the embodiment of the present application further provides a computer readable medium having a processor-executable non-volatile program code, the program code causing the processor to perform the method of the first aspect.
  • the target monitoring method is applied to the camera, including: acquiring a current frame image collected by the camera of the camera; classifying the objects in the current frame image, and acquiring the suspected object corresponding to the current control target; Determining the target to be monitored in the current frame image in the obtained suspect object; controlling the camera to perform tracking shooting on the target to be monitored, obtaining a close-up image of the target to be monitored; performing feature recognition on the close-up image according to characteristics of the currently deployed target, and determining Whether the monitoring target belongs to the current deployment target; if yes, the target to be monitored is determined as the monitoring target, and the monitoring target is tracked and monitored.
  • the target to be monitored can be automatically screened out from the current frame image collected by the camera of the camera, and the target to be monitored is automatically determined to be the current control target, so the rear end control platform is not relied on, thereby saving labor cost. And it is not easy to have leakage prevention problems.
  • the method since the method is applied to the camera, the video file collected by the camera is directly processed on the camera end, and the video file is no longer needed to be processed to the back end, so the real-time performance is strong.
  • the target monitoring method, the target monitoring device, the camera, and the computer readable medium provided by the embodiments of the present application reduce the cost input in the target monitoring process, and improve the timeliness and reliability of the target monitoring.
  • FIG. 1 is a schematic diagram of a first process of a target monitoring method according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of a second process of a target monitoring method according to an embodiment of the present application.
  • FIG. 3 is a schematic diagram of a third process of a target monitoring method according to an embodiment of the present disclosure.
  • FIG. 4 is a schematic structural diagram of a module of a target monitoring apparatus according to an embodiment of the present disclosure
  • FIG. 5 is a schematic structural diagram of another module of a target monitoring apparatus according to an embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of a camera according to an embodiment of the present application.
  • 10-image acquisition module 20-object acquisition module; 30-to-be-monitored target determination module; 40-tracking capture module; 50-judgment module; 60-monitoring module; 70-alarm information sending module; 80-distribution target determining module; 100-PTZ; 200-PTZ motor; 300-camera; 400-processor; 500-memory; 600-bus; 700-communication interface.
  • the PTZ camera is manually controlled for target monitoring, so there is a technical problem that wastes labor cost, poor real-time performance, and easy leakage prevention.
  • the target monitoring method, the target monitoring device and the camera provided by the embodiments of the present invention save labor cost and are not prone to leakage prevention; the video file collected by the camera is directly processed at the camera end, and the real-time performance is strong.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • the target monitoring method provided by the embodiment of the present application is applied to a camera, and the camera can be installed in various video monitoring places, such as a public place such as a bank, a shopping mall, a station, and a traffic intersection, and a private place such as a personal home.
  • a public place such as a bank
  • a shopping mall such as a shopping mall
  • a station such as a station
  • a traffic intersection such as a personal home.
  • a private place such as a personal home.
  • Cameras are installed at general traffic intersections for illegal photographing or for video recording, providing evidence and clues for public security organs to crack down on street crimes.
  • the camera of the camera used in the embodiment of the present application should have a zoom lock function, and the camera can be rotated in various directions to track the target.
  • the camera can be, but is not limited to, a PTZ camera.
  • the camera has pre-stored control targets, and the number of control targets can be multiple, for example, multiple characters, multiple vehicles, etc.
  • a certain type of control target can be used as the current control target according to actual needs.
  • FIG. 1 is a schematic diagram of a first process of a target monitoring method according to an embodiment of the present application. As shown in FIG. 1 , the method includes the following steps:
  • Step S101 Acquire a current frame image acquired by a camera of the camera.
  • the camera of the camera captures a video file of a monitoring area (for example, a traffic intersection) in real time, and the processor acquires the video file and decodes the video file to obtain a current frame image.
  • a monitoring area for example, a traffic intersection
  • Step S102 classify objects in the current frame image, and obtain suspected objects corresponding to the current control target.
  • the objects here generally refer to movable objects.
  • the objects of the same category as the currently deployed target are roughly selected, that is, the suspected objects.
  • the objects in the current frame image are classified according to a preset classification rule, and the preset classification rule includes classifying the objects according to the characters, the vehicle, and the animal, and extracting the suspected objects of the current control target corresponding category from the classified objects. For example, if the current control target is a pedestrian, each pedestrian in the current frame image is regarded as a suspect object; if the current control target is a car, each car in the current frame image is regarded as a suspect object.
  • the specific execution algorithm There is no limitation on the specific execution algorithm here.
  • the step S102 is performed by a depth neural network based video structuring algorithm, which is responsible for analyzing the object in the panoramic video image. Identifying multiple types of objects, including pedestrians, bicycles, motorcycles, cars, tricycles, buses, vans, and trucks, enables accurate detection of objects with categorical attributes.
  • the algorithm can accurately calibrate the categories of objects in the current frame image, providing pre-guarantee for the automatic selection and tracking of the target to be monitored.
  • the video structuring algorithm is stored in an embedded processing chip, and the chip is disposed in the camera.
  • the embedded video structuring algorithm has the characteristics of fast processing speed and high efficiency.
  • Step S103 Determine, from the acquired suspect object, the target to be monitored in the current frame image according to the current deployment target.
  • the processor extracts features of the suspected object and compares with the features of the currently deployed target, and determines the suspected object with the highest similarity as the target to be monitored.
  • the current deployment target may be multiple, and the suspect object may also be multiple.
  • the similarity between each suspect object and each current control target needs to be determined, and the suspect object with the highest similarity is determined as the target to be monitored.
  • Table 1 is a schematic table of similarity results between the suspected object and the current deployed target. Any specific values in Table 1 are merely exemplary and not limiting.
  • the suspected objects include A, B, and C.
  • the current deployment targets include X and Y. The highest similarity between C and X is 50%, and C is determined as the target to be monitored.
  • the video structuring algorithm is used to determine the similarity of the suspect object, and the suspect object with the highest similarity is determined as the target to be monitored.
  • the video structuring algorithm is the same as the algorithm in step S102, and details are not described herein again.
  • Step S104 controlling the camera to perform tracking shooting on the target to be monitored, and obtaining a close-up image of the target to be monitored.
  • the target to be monitored in the panoramic monitoring has less pixels in the current frame image, and it is difficult to determine whether the target to be monitored belongs to the current deployment target according to the current frame image. Therefore, close-up tracking of the target to be monitored is required.
  • the processor adjusts the focal length of the camera, and uses the target tracking algorithm and the image compensation algorithm to determine the position of the target to be monitored in the captured video image, and uses the PTZ control algorithm to control the rotation angle and the rotation speed of the camera according to the position. , get a close-up image of the target to be monitored.
  • the target tracking algorithm is responsible for determining the position of the target to be monitored in the video image; the image compensation algorithm can prevent the motion and jitter of the camera from affecting the target tracking algorithm; the PTZ control algorithm can realize the control of the PTZ motor and make the cloud
  • the station can control the corresponding angle, angular velocity and angular acceleration according to the position given by the target tracking algorithm, and realize real-time tracking of the target to be monitored.
  • Target tracking algorithms include: frame difference method, kinect video tracking algorithm, mineshift target tracking algorithm, kalman filtering algorithm, OAB (Online Adaptive Boosting Tracking) tracking algorithm, IVT (Incremental Learning for Robust Visual Tracing, incremental) Learning) tracking algorithm, MIL (Multiple instance learning) tracking algorithm, CT (Fast Compressive Tracking) tracking algorithm, TLD (Tracking-Learning-Detection), Struck tracking algorithm, and the like.
  • OAB Online Adaptive Boosting Tracking Tracking
  • IVT Intelligent Learning for Robust Visual Tracing, incremental) Learning
  • MIL Multiple instance learning
  • CT Fast Compressive Tracking
  • TLD Track-Learning-Detection
  • Struck tracking algorithm and the like.
  • the camera is a PTZ camera, and the camera is disposed on the cloud platform, and the rotation angle and the rotation speed of the pan/tilt are controlled by the pan/tilt control algorithm to control the rotation angle and the rotation speed of the camera.
  • the PTZ camera has a zoom function, which can close the target to be monitored by zooming in on the lens, so that the target to be monitored occupies a sufficient number of pixels in the video image, which is beneficial to the practical application of the target tracking algorithm and maximizes the target tracking. algorithm.
  • the rotation angle and focal length of the camera are adjusted in real time so that the target to be monitored remains in the center of the video image taken by the camera and takes up most of the image to achieve close-up.
  • Step S105 Perform feature recognition on the close-up image according to the feature of the current deployed target, and determine whether the target to be monitored belongs to the current deployed target.
  • the processor extracts the feature of the target to be monitored in the close-up image, compares the feature of the target to be monitored with the feature of the currently deployed target, and obtains a comparison result, and determines, according to the comparison result, whether the target to be monitored belongs to the current deployed target. .
  • step S105 may be performed by using multiple target recognition algorithms, such as a face recognition algorithm, a vehicle type recognition algorithm, a license plate recognition algorithm, etc., to obtain a matching degree between the target to be monitored and the current deployed target, thereby accurately determining whether the target to be monitored belongs to the target.
  • the current control target The comparison result includes the matching degree, and whether the target to be monitored belongs to the current deployment target according to the relationship between the matching degree and the preset matching threshold, wherein the matching threshold may be set according to actual needs. If the matching degree is greater than the preset matching threshold, it is determined that the target to be monitored belongs to the current deployed target, and vice versa, it is determined that the target to be monitored does not belong to the current deployed target.
  • the preset matching threshold is 0.7
  • the matching degree of the target C to be monitored is greater than 0.7
  • the target C to be monitored belongs to the current deployed target; if the target C to be monitored and the current deployed target X If the matching degree of Y is less than or equal to 0.7, it is determined that the target C to be monitored does not belong to the current deployment target.
  • step S106 is performed; if the determination result is no, the camera is controlled to zoom to the panoramic mode, and step S101 is re-executed.
  • Step S106 if yes, determining the target to be monitored as a monitoring target, and tracking and monitoring the monitoring target.
  • the processor determines that the target to be monitored belongs to the current deployed target, it determines that the target to be monitored is a monitoring target, and continues to track and monitor the monitoring target by using a target tracking algorithm, an image compensation algorithm, and a PTZ control algorithm.
  • the target monitoring method provided by the embodiment of the present application does not depend on the backend control platform, and the processor applying the method has efficient local video processing capability, strong mobility, less bandwidth occupation, and good tracking and recognition effect.
  • the target monitoring method is applied to the camera, including: acquiring a current frame image collected by the camera of the camera; classifying the objects in the current frame image, and acquiring the suspected object corresponding to the current control target; Determining the target to be monitored in the current frame image in the obtained suspect object; controlling the camera to perform tracking shooting on the target to be monitored, obtaining a close-up image of the target to be monitored; performing feature recognition on the close-up image according to characteristics of the currently deployed target, and determining Whether the monitoring target belongs to the current deployment target; if yes, the target to be monitored is determined as the monitoring target, and the monitoring target is tracked and monitored.
  • the target to be monitored can be automatically screened out from the current frame image collected by the camera of the camera, and the target to be monitored is automatically determined to be the current control target, so the rear end control platform is not relied on, thereby saving labor cost. And it is not easy to have leakage prevention problems.
  • the target monitoring method provided by the embodiment of the present application reduces the cost input in the target monitoring process and improves the timeliness and reliability of the target monitoring.
  • the alarm information is sent to the monitoring terminal.
  • the monitoring terminal includes a mobile phone, a computer, a tablet computer, etc., and the monitoring personnel can receive the alarm information in time through the monitoring terminal, thereby facilitating timely arrest of the monitored control target.
  • FIG. 2 is a second schematic flowchart of a target monitoring method according to an embodiment of the present application. As shown in FIG. 2, the target monitoring method includes the following steps:
  • step S201 when the alert target information sent by the monitoring terminal is received, the object in the alert target information is determined as the current deployment target.
  • the monitoring terminal sends the alert target information to the camera within the specified range, and based on this, the processor will alert the target information when receiving the alert target information sent by the monitoring terminal.
  • the object in the middle is determined as the current control target, so that the monitoring efficiency of the control target can be improved.
  • Step S202 Acquire a current frame image acquired by a camera of the camera.
  • Step S203 classify objects in the current frame image, and obtain suspected objects corresponding to the current control target.
  • Step S204 Determine, from the acquired suspect object, the target to be monitored in the current frame image according to the current deployment target.
  • Step S205 controlling the camera to perform tracking shooting on the target to be monitored, and obtaining a close-up image of the target to be monitored.
  • Step S206 Perform feature recognition on the close-up image according to the feature of the current deployed target, and determine whether the target to be monitored belongs to the current deployed target.
  • Step S207 if yes, determining the target to be monitored as a monitoring target, and tracking and monitoring the monitoring target.
  • Steps S202 to S207 are the same as steps S101 to S106, respectively, and are not described herein again.
  • FIG. 3 is a third schematic flowchart of a target monitoring method according to an embodiment of the present disclosure. As shown in FIG. 3, in an embodiment, the method includes the following steps:
  • Step S301 acquiring a current frame image.
  • Step S302 using a video structuring algorithm for classification and recognition.
  • step S303 the target to be monitored is determined.
  • Step S304 tracking the target to be monitored, adjusting the focal length and rotation of the camera, and acquiring a close-up image of the target to be monitored.
  • Step S305 determining whether the target to be monitored matches the deployment target.
  • step S306 If yes, go to step S306; if no, zoom to the panoramic image and re-execute step S301.
  • Step S306 sending an alarm message to the monitoring terminal.
  • step S307 the tracking is continued.
  • any specific values should be construed as merely exemplary, and not as a limitation, and thus, other examples of the exemplary embodiments may have different values.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • FIG. 4 is a schematic structural diagram of a module of a target monitoring apparatus according to an embodiment of the present disclosure. As shown in FIG. 4, the target monitoring apparatus includes:
  • the image acquisition module 10 is configured to acquire a current frame image acquired by a camera of the camera.
  • the object obtaining module 20 is configured to classify the objects in the current frame image, and obtain the suspected objects corresponding to the current control target.
  • the object obtaining module 20 is specifically configured to:
  • Sorting objects in the current frame image according to a preset classification rule including classifying objects according to characters, vehicles, and animals;
  • the to-be-monitored target determining module 30 is configured to determine, from the acquired suspected objects, the target to be monitored in the current frame image according to the current deployed target.
  • the target to be monitored module 30 is specifically configured as:
  • the suspected object with the highest similarity is determined as the target to be monitored.
  • the tracking shooting module 40 is configured to control the camera to perform tracking shooting on the target to be monitored, and obtain a close-up image of the target to be monitored.
  • the tracking and shooting module 40 is specifically configured to:
  • the determining module 50 is configured to perform feature recognition on the close-up image according to the feature of the current deployed target, and determine whether the target to be monitored belongs to the current deployed target.
  • the determining module 50 is specifically configured to:
  • the monitoring module 60 is configured to determine, as the monitoring target, the target to be monitored when the determination result of the determining module 50 is YES, and track and monitor the monitoring target.
  • the target monitoring device may further include an alarm information sending module 70, and the alarm information sending module 70 is configured to send the alarm information to the monitoring when determining that the target to be monitored is the current deployed target. terminal.
  • the target monitoring device may further include a deployment target determination module 80, and the deployment target determination module 80 is configured to determine, as the current deployment target, the object in the alert target information when receiving the alert target information sent by the monitoring terminal.
  • the image acquiring module 10 acquires the current frame image collected by the camera of the camera; the object acquiring module 20 classifies the objects in the current frame image, and obtains the suspected object of the corresponding category of the current control target; the target to be monitored determining module 30 Determining a target to be monitored in the current frame image from the acquired suspect object according to the current control target; the tracking shooting module 40 controls the camera to perform tracking shooting on the target to be monitored, and obtains a close-up image of the target to be monitored; and the determining module 50 according to the current control target
  • the feature is characterized by the feature image to determine whether the target to be monitored belongs to the current control target; if yes, the monitoring module 60 determines the target to be monitored as the monitoring target, and tracks and monitors the monitoring target.
  • the target monitoring device can automatically filter out the target to be monitored from the current frame image collected by the camera of the camera, and automatically determine whether the target to be monitored belongs to the current control target, and therefore does not rely on the backend control platform, thereby saving labor cost, and not It is prone to leakage prevention.
  • the target monitoring device since the device is applied to the camera, the video file collected by the camera is directly processed on the camera end, and the video file is no longer needed to be processed to the back end, so the real-time performance is strong.
  • the target monitoring device provided by the embodiment of the present application reduces the cost input in the target monitoring process and improves the timeliness and reliability of the target monitoring.
  • each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that includes one or more of the Executable instructions.
  • the functions noted in the blocks may also occur in a different order than that illustrated in the drawings. For example, two consecutive blocks may be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or function. Or it can be implemented by a combination of dedicated hardware and computer instructions.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • FIG. 6 is a schematic structural diagram of a camera according to an embodiment of the present disclosure.
  • the embodiment of the present application further provides a camera, including: a cloud platform 100, a pan/tilt motor 200 connected to the cloud platform 100, and a cloud platform 100.
  • the pan-tilt motor 200 and the camera 300 are respectively electrically connected to the processor 400.
  • the processor 400 executes the computer program, the steps of the method of the first embodiment are implemented.
  • the processor 400 controls the pan-tilt motor 200 to rotate the camera 300.
  • the memory 500 and the processor 400 are integrated in an embedded processing chip.
  • the above camera also includes a bus 600 and a communication interface 700, and the processor 400, the communication interface 700, and the memory 500 are connected by a bus 600.
  • the memory 500 may include a high speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.
  • RAM random access memory
  • non-volatile memory such as at least one disk memory.
  • the communication connection between the system network element and at least one other network element is implemented by at least one communication interface 700 (which may be wired or wireless), and may use an Internet, a wide area network, a local network, a metropolitan area network, or the like.
  • the bus 600 can be an ISA bus, a PCI bus, or an EISA bus.
  • the bus 600 can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one double-headed arrow is shown in Figure 5, but it does not mean that there is only one bus or one type of bus.
  • the memory 500 is used to store a computer program, and the processor 400 executes the computer program after receiving the execution instruction.
  • the method executed by the device defined by the flow process disclosed in any embodiment of the present application may be applied.
  • processor 400 or implemented by processor 400.
  • Processor 400 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method may be completed by an integrated logic circuit of hardware in the processor 400 or an instruction in a form of software.
  • the processor 400 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP processor, etc.), or a digital signal processor (DSP). ), Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component.
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present application may be directly implemented by the hardware decoding processor, or may be performed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
  • the storage medium is located in the memory 500, and the processor 400 reads the information in the memory 500 and performs the steps of the above method in combination with its hardware.
  • the target monitoring device and the camera provided by the embodiment of the present application have the same technical features as the target monitoring method provided by the foregoing embodiment, so that the same technical problem can be solved and the same technical effect can be achieved.
  • Embodiment 4 is a diagrammatic representation of Embodiment 4:
  • the embodiment of the present application further provides a computer readable medium having non-volatile program code executable by a processor, where the program code includes instructions for executing the method described in the foregoing method embodiments, and the specific implementation may be implemented. See the method embodiment, and details are not described herein again.
  • the functions, if implemented in the form of software functional units and sold or used as separate products, may be stored in a non-transitory computer readable storage medium executable by a processor.
  • the technical solution of the present application which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .
  • the disclosed method, apparatus, and video camera may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some communication interface, device or unit, and may be electrical, mechanical or otherwise.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the target monitoring method, the target monitoring device, the camera and the computer readable medium provided by the embodiments of the present application reduce the cost input in the target monitoring process, and improve the timeliness and reliability of the target monitoring.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Signal Processing (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Studio Devices (AREA)
  • Image Analysis (AREA)

Abstract

本申请提供了一种目标监控方法、目标监控装置及摄像机，该方法应用于摄像机，包括：获取摄像机的摄像头采集的当前帧图像；对当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；根据当前布控目标从获取到的疑似对象中确定当前帧图像中的待监控目标；控制摄像头对待监控目标进行跟踪拍摄，得到待监控目标的特写图像；根据当前布控目标的特征对特写图像进行特征识别，判断待监控目标是否属于当前布控目标；如果是，则将待监控目标确定为监控目标，并跟踪监控该监控目标。本申请提供的目标监控方法、目标监控装置及摄像机，可以降低目标监控过程中的成本投入，提升目标监控的时效性和可靠性。

Description

目标监控方法、目标监控装置、摄像机及计算机可读介质
本申请要求于2017年06月09日提交中国专利局的申请号为201710435936.0名称为“目标监控方法、目标监控装置及摄像机”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及安防监控技术领域，尤其是涉及一种目标监控方法、目标监控装置、摄像机及计算机可读介质。
背景技术
随着“平安城市”系统建设的快速发展，对视频监控摄像机的需求急剧增加。传统安防监控领域通常采用PTZ(Pan Tilt Zoom，平移倾斜变焦)摄像机进行视频监控，PTZ摄像机带有云台功能，可以实现云台全方位(左右/上下)移动及镜头变倍、变焦控制。
在各个安防应用场景中，PTZ摄像机的控制通常是由后端操作人员根据PTZ摄像机传送的视频文件通过控制台来手动完成，以实现对特定监控目标的特写监控及跟踪，当操作人员发现监控跟踪的目标确实为警戒目标时会采取相应行动，例如报警等。但是人是会疲劳的，会因为注意力不集中而出现漏防问题。另外由于传送视频文件时存在一定的时间差，因此人工控制PTZ摄像机进行目标监控存在实时性差的问题。
综上可知，现有的目标监控方法存在浪费人力成本、实时性差、容易出现漏防的技术问题。
发明内容
有鉴于此，本申请的目的在于提供一种目标监控方法、目标监控装置及摄像机，以降低目标监控过程中的成本投入，提升目标监控的时效性和可靠性。
第一方面，本申请实施例提供了一种目标监控方法，所述方法应用于摄像机，包括：
获取所述摄像机的摄像头采集的当前帧图像；
对所述当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；
根据所述当前布控目标从获取到的所述疑似对象中确定所述当前帧图像中的待监控目标；
控制所述摄像头对所述待监控目标进行跟踪拍摄，得到所述待监控目 标的特写图像；
根据所述当前布控目标的特征对所述特写图像进行特征识别，判断所述待监控目标是否属于所述当前布控目标；
如果是，则将所述待监控目标确定为监控目标，并跟踪监控所述监控目标。
结合第一方面，本申请实施例提供了第一方面的第一种可能的实施方式，其中，所述对所述当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象，包括：
根据预设分类规则对所述当前帧图像内的对象进行分类，所述预设分类规则包括将对象按照人物、车辆和动物分类；
从分类后的所述对象中提取所述当前布控目标对应类别的疑似对象。
结合第一方面，本申请实施例提供了第一方面的第二种可能的实施方式，其中，所述根据所述当前布控目标从获取到的所述疑似对象中确定所述当前帧图像中的待监控目标，包括：
根据所述当前布控目标，采用视频结构化算法确定所述疑似对象的相似度；
将相似度最高的所述疑似对象确定为所述待监控目标。
结合第一方面，本申请实施例提供了第一方面的第三种可能的实施方式，其中，所述控制所述摄像头对所述待监控目标进行跟踪拍摄，得到所述待监控目标的特写图像，包括：
调整所述摄像头的焦距，并采用目标跟踪算法和图像补偿算法确定所述待监控目标在跟踪拍摄的视频图像中的位置，根据所述位置采用云台控制算法控制所述摄像头的转动角度和转动速度，得到所述待监控目标的特写图像。
结合第一方面，本申请实施例提供了第一方面的第四种可能的实施方式，其中，所述根据所述当前布控目标的特征对所述特写图像进行特征识别，判断所述待监控目标是否属于所述当前布控目标，包括：
提取所述特写图像中的所述待监控目标的特征；
将所述待监控目标的特征与所述当前布控目标的特征进行对比，得到对比结果；
根据所述对比结果，判断所述待监控目标是否属于所述当前布控目标。
结合第一方面，本申请实施例提供了第一方面的第五种可能的实施方 式，其中，所述方法还包括：
当判断所述待监控目标为所述当前布控目标时，发送报警信息至监控终端。
结合第一方面，本申请实施例提供了第一方面的第六种可能的实施方式，其中，所述方法还包括：
当接收到监控终端发送的警戒目标信息时，将所述警戒目标信息中的对象确定为所述当前布控目标。
结合第一方面，本申请实施例提供了第一方面的第七种可能的实施方式，其中，所述方法还包括：
当接收到监控终端发送的警戒目标信息时，将警戒目标信息中的对象确定为当前布控目标。
第二方面，本申请实施例还提供一种目标监控装置，包括：
图像获取模块，用于获取所述摄像机的摄像头采集的当前帧图像；
对象获取模块，用于对所述当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；
待监控目标确定模块，用于根据所述当前布控目标从获取到的所述疑似对象中确定所述当前帧图像中的待监控目标；
跟踪拍摄模块，用于控制所述摄像头对所述待监控目标进行跟踪拍摄，得到所述待监控目标的特写图像；
判断模块，用于根据所述当前布控目标的特征对所述特写图像进行特征识别，判断所述待监控目标是否属于所述当前布控目标；
监控模块，用于当所述判断模块的判断结果为是时，将所述待监控目标确定为监控目标，并跟踪监控所述监控目标。
结合第二方面，本申请实施例提供了第二方面的第一种可能实施方式，对象获取模块具体配置为：
根据预设分类规则对所述当前帧图像内的对象进行分类，所述预设分类规则包括将对象按照人物、车辆和动物分类；
从分类后的所述对象中提取所述当前布控目标对应类别的疑似对象。
结合第二方面，本申请实施例提供了第二方面的第二种可能实施方式，待监控目标确定模块具体配置为：
根据所述当前布控目标，采用视频结构化算法确定所述疑似对象的相似度；
将相似度最高的所述疑似对象确定为所述待监控目标。
结合第二方面，本申请实施例提供了第二方面的第三种可能实施方式，跟踪拍摄模块具体配置成：
调整所述摄像头的焦距，并采用目标跟踪算法和图像补偿算法确定所述待监控目标在跟踪拍摄的视频图像中的位置，根据所述位置采用云台控制算法控制所述摄像头的转动角度和转动速度，得到所述待监控目标的特写图像。
结合第二方面，本申请实施例提供了第二方面的第四种可能实施方式，判断模块具体配置成：
提取所述特写图像中的所述待监控目标的特征；
将所述待监控目标的特征与所述当前布控目标的特征进行对比，得到对比结果；
根据所述对比结果，判断所述待监控目标是否属于所述当前布控目标。
结合第二方面，本申请实施例提供了第二方面的第五种可能实施方式，装置还包括：
报警信息发送模块，配置成当判断所述待监控目标为所述当前布控目标时，发送报警信息至监控终端。
结合第二方面，本申请实施例提供了第二方面的第六种可能实施方式，装置还包括：
布控目标确定模块，配置成当接收到监控终端发送的警戒目标信息时，将警戒目标信息中的对象确定为当前布控目标。
第三方面，本申请实施例还提供一种摄像机，包括云台、与所述云台连接的云台电机、设置在所述云台上的摄像头、存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序；
所述云台电机、所述摄像头分别与所述处理器电连接，所述处理器执行所述计算机程序时实现上述第一方面所述的方法的步骤，其中，所述处理器通过控制所述云台电机带动所述摄像头转动。
第四方面，本申请实施例还提供一种具有处理器可执行的非易失的程序代码的计算机可读介质，所述程序代码使所述处理器执行第一方面所述方法。
本申请实施例带来了以下有益效果：
本申请实施例中，目标监控方法应用于摄像机，包括：获取摄像机的 摄像头采集的当前帧图像；对当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；根据当前布控目标从获取到的疑似对象中确定当前帧图像中的待监控目标；控制该摄像头对待监控目标进行跟踪拍摄，得到待监控目标的特写图像；根据当前布控目标的特征对该特写图像进行特征识别，判断待监控目标是否属于当前布控目标；如果是，则将待监控目标确定为监控目标，并跟踪监控该监控目标。应用该目标监控方法时，可以从摄像机的摄像头采集的当前帧图像中自动筛选出待监控目标，并自动判断该待监控目标是否属于当前布控目标，因此不依赖后端控制平台，节约了人力成本，且不容易出现漏防问题。同时，由于该方法应用于摄像机，直接在摄像机端处理该摄像机采集的视频文件，不再需要传送视频文件至后端处理，因此实时性强。综上可知，本申请实施例提供的目标监控方法、目标监控装置、摄像机及计算机可读介质降低了目标监控过程中的成本投入，并提升了目标监控的时效性和可靠性。
本申请的其他特征和优点将在随后的说明书中阐述，并且，部分地从说明书中变得显而易见，或者通过实施本申请而了解。本申请的目的和其他优点在说明书、权利要求书以及附图中所特别指出的结构来实现和获得。
为使本申请的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。
附图说明
为了更清楚地说明本申请具体实施方式或现有技术中的技术方案，下面将对具体实施方式或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图是本申请的一些实施方式，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为本申请实施例提供的目标监控方法的第一种流程示意图；
图2为本申请实施例提供的目标监控方法的第二种流程示意图；
图3为本申请实施例提供的目标监控方法的第三种流程示意图；
图4为本申请实施例提供的一种目标监控装置的模块组成示意图；
图5为本申请实施例提供的另一种目标监控装置的模块组成示意图；
图6为本申请实施例提供的摄像机的结构示意图。
图标：
10-图像获取模块；20-对象获取模块；30-待监控目标确定模块；40-跟 踪拍摄模块；50-判断模块；60-监控模块；70-报警信息发送模块；80-布控目标确定模块；100-云台；200-云台电机；300-摄像头；400-处理器；500-存储器；600-总线；700-通信接口。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合附图对本申请的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
目前通常根据摄像机传送的视频文件，采用人工控制PTZ摄像机进行目标监控，因此存在浪费人力成本、实时性差、容易出现漏防的技术问题。基于此，本申请实施例提供的一种目标监控方法、目标监控装置及摄像机，节约了人力成本，且不容易出现漏防问题；直接在摄像机端处理该摄像机采集的视频文件，实时性强。
为便于对本实施例进行理解，首先对本申请实施例所公开的一种目标监控方法进行详细介绍。
实施例一：
本申请实施例提供的目标监控方法应用于摄像机，该摄像机可以安装在各个视频监控场所，例如银行、商场、车站和交通路口等公共场所，以及个人家庭等私人场所，下面以监控交通路口为例进行说明。
一般交通路口都安装有摄像机，用于违章拍照，或者用于录像，为公安机关打击街面犯罪提供证据和线索。本申请实施例所采用的摄像机的摄像头应具有变焦锁定功能，且该摄像头可以在各个方向上转动以跟踪目标，该摄像机可以但不限制为PTZ摄像机。同时该摄像机内预存储有布控目标，该布控目标可以为多个，例如多个人物、多辆车等，当进行监控时，可以根据实际需要将某一类别的布控目标作为当前布控目标。
图1为本申请实施例提供的目标监控方法的第一种流程示意图，如图1所示，该方法包括以下几个步骤：
步骤S101，获取摄像机的摄像头采集的当前帧图像。
具体地，上述摄像机的摄像头实时拍摄监控区域(比如，交通路口)的视频文件，处理器获取该视频文件，并对该视频文件进行解码，可以得到当前帧图像。
步骤S102，对当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象。
当前帧图像内可能存在一个或多个对象，这里的对象一般指可移动的物体，首先粗选出与当前布控目标相同类别的对象，即疑似对象。具体地，根据预设分类规则对当前帧图像内的对象进行分类，该预设分类规则包括将对象按照人物、车辆和动物分类，从分类后的对象中提取当前布控目标对应类别的疑似对象。例如，若当前布控目标为行人，则将当前帧图像内各个行人作为疑似对象；若当前布控目标为小汽车，则将当前帧图像内各个小汽车作为疑似对象。这里对具体的执行算法不作限制。
通常视频图像中较小且移动缓慢的对象不容易被侦测到，优选地，步骤S102处理器采用基于深度神经网络的视频结构化算法执行，该算法负责全景视频图像中的对象的分析，能够识别出多类对象，包括行人、自行车、摩托车、小汽车、三轮车、巴士、面包车和卡车等，能够实现对具有分类属性的对象的精确检测。该算法可以精确标定当前帧图像中的对象的类别，为待监控目标的自动选定和跟踪提供预先保障。进一步地，为了提升处理效率，该视频结构化算法存储在嵌入式处理芯片中，该芯片设置在摄像机内。采用嵌入式的视频结构化算法，具有处理速度快，效率高特点。
步骤S103，根据当前布控目标从获取到的疑似对象中确定当前帧图像中的待监控目标。
具体地，处理器提取疑似对象的特征，并与当前布控目标的特征进行对比，将相似度最高的疑似对象确定为待监控目标。当前布控目标可以为多个，疑似对象也可以为多个，这里需要确定每个疑似对象与每个当前布控目标的相似度，并将其中相似度最高的疑似对象确定为待监控目标。这里对具体的执行算法不作限制。
表1为疑似对象与当前布控目标的相似度结果示意表格，表1中的任何具体值仅仅是示例性的，而不是作为限制。如表1所示，疑似对象包括A、B、C，当前布控目标包括X、Y，其中C与X的相似度最高，为50％，则将C确定为待监控目标。
表1
Figure PCTCN2018089945-appb-000001
Figure PCTCN2018089945-appb-000002
优选地，根据当前布控目标，采用视频结构化算法确定上述疑似对象的相似度，将相似度最高的疑似对象确定为待监控目标。该视频结构化算法与步骤S102中的算法相同，这里不再赘述。
步骤S104，控制上述摄像头对待监控目标进行跟踪拍摄，得到该待监控目标的特写图像。
由于布控场景中，在全景监控时待监控目标在当前帧图像中所占像素都较少，难以根据当前帧图像来判定待监控目标是否属于当前布控目标。因此，需要对待监控目标进行特写跟踪。
具体地，处理器调整摄像头的焦距，并采用目标跟踪算法和图像补偿算法确定待监控目标在跟踪拍摄的视频图像中的位置，根据该位置采用云台控制算法控制该摄像头的转动角度和转动速度，得到待监控目标的特写图像。其中，目标跟踪算法负责确定待监控目标在视频图像中的位置；图像补偿算法可以防止摄像机的运动和抖动给目标跟踪算法带来影响；云台控制算法可以实现对云台电机的控制，使云台能够根据目标跟踪算法给出的位置进行相应角度、角速度、角加速度的控制，对待监控目标实现实时跟踪。
目标跟踪算法包括：帧差法、kinect视频追踪算法、mineshift目标跟踪算法、kalman滤波算法、OAB(Online Adaptive Boosting Tracking，在线自适应增强)跟踪算法、IVT(Incremental Learning for Robust Visual Tracing，增量式学习)跟踪算法、MIL(Multiple instance learning，多示例学习)跟踪算法、CT(Fast Compressive Tracking，高速压缩)跟踪算法、TLD(Tracking-Learning-Detection)、Struck跟踪算法等。这里对具体的目标跟踪算法、图像补偿算法和云台控制算法不作限制。
在一个实施例中，上述摄像机为PTZ摄像机，其摄像头设置在云台上，通过云台控制算法控制云台的转动角度和转动速度来控制摄像头的转动角度和转动速度。PTZ摄像机具有变焦功能，可以通过拉近镜头实现对待监控目标进行特写跟踪，使待监控目标在视频图像中占据有足够的像素数量，有利于目标跟踪算法的实际应用，能最大限度地优化目标跟踪算法。在跟踪过程中，实时调节摄像头的转动角度和焦距，以使待监控目标保持在摄 像头拍摄的视频图像的中央并占据大部分图像而实现特写。
步骤S105，根据当前布控目标的特征对上述特写图像进行特征识别，判断待监控目标是否属于该当前布控目标。
具体地，处理器提取特写图像中的待监控目标的特征，将待监控目标的特征与当前布控目标的特征进行对比，得到对比结果，根据该对比结果，判断待监控目标是否属于该当前布控目标。
进一步地，可以采用多种目标识别算法执行步骤S105，例如人脸识别算法、车型识别算法、车牌识别算法等，得到待监控目标与当前布控目标的匹配度，从而精确判断待监控目标是否属于该当前布控目标。上述对比结果包括该匹配度，根据该匹配度与预设的匹配阈值之间的大小关系来判断待监控目标是否属于该当前布控目标，其中，匹配阈值可以根据实际需要设置。若该匹配度大于预设的匹配阈值，则判断待监控目标属于该当前布控目标，反之则判断待监控目标不属于该当前布控目标。
例如，预设的匹配阈值为0.7，若待监控目标C与当前布控目标X或Y的匹配度大于0.7，则判断待监控目标C属于该当前布控目标；若待监控目标C与当前布控目标X、Y的匹配度均小于等于0.7，则判断待监控目标C不属于该当前布控目标。
若判断结果为是，则执行步骤S106；若判断结果为否，则控制摄像头变焦至全景模式，重新执行上述步骤S101。
步骤S106，如果是，则将上述待监控目标确定为监控目标，并跟踪监控该监控目标。
当处理器判断待监控目标属于该当前布控目标时，确定该待监控目标为监控目标，并采用目标跟踪算法、图像补偿算法和云台控制算法继续跟踪监控该监控目标。
本申请实施例提供的目标监控方法不依赖后端控制平台，应用该方法的处理器具有高效的本地视频处理能力，机动性强，带宽占据量少，跟踪识别效果好。
本申请实施例中，目标监控方法应用于摄像机，包括：获取摄像机的摄像头采集的当前帧图像；对当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；根据当前布控目标从获取到的疑似对象中确定当前帧图像中的待监控目标；控制该摄像头对待监控目标进行跟踪拍摄，得到待监控目标的特写图像；根据当前布控目标的特征对该特写图像进行 特征识别，判断待监控目标是否属于当前布控目标；如果是，则将待监控目标确定为监控目标，并跟踪监控该监控目标。应用该目标监控方法时，可以从摄像机的摄像头采集的当前帧图像中自动筛选出待监控目标，并自动判断该待监控目标是否属于当前布控目标，因此不依赖后端控制平台，节约了人力成本，且不容易出现漏防问题。同时，由于该方法应用于摄像机，直接在摄像机端处理该摄像机采集的视频文件，不再需要传送视频文件至后端处理，因此实时性强。综上可知，本申请实施例提供的目标监控方法降低了目标监控过程中的成本投入，并提升了目标监控的时效性和可靠性。
为了加快监控人员对布控目标的处理速度，提高处理效率，当判断待监控目标为当前布控目标时，发送报警信息至监控终端。监控终端包括手机、电脑、平板电脑等，监控人员可以通过监控终端及时接收到报警信息，从而便于及时抓捕监控到的布控目标。
图2为本申请实施例提供的目标监控方法的第二种流程示意图，如图2所示，该目标监控方法包括以下几个步骤：
步骤S201，当接收到监控终端发送的警戒目标信息时，将警戒目标信息中的对象确定为当前布控目标。
考虑到当发现某一布控目标已至指定范围时，监控终端会向该指定范围内的摄像机发送警戒目标信息，基于此，处理器当接收到监控终端发送的警戒目标信息时，将警戒目标信息中的对象确定为当前布控目标，从而可以提高对该布控目标的监控效率。
步骤S202，获取摄像机的摄像头采集的当前帧图像。
步骤S203，对当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象。
步骤S204，根据当前布控目标从获取到的疑似对象中确定当前帧图像中的待监控目标。
步骤S205，控制上述摄像头对待监控目标进行跟踪拍摄，得到该待监控目标的特写图像。
步骤S206，根据当前布控目标的特征对上述特写图像进行特征识别，判断待监控目标是否属于该当前布控目标。
步骤S207，如果是，则将上述待监控目标确定为监控目标，并跟踪监控该监控目标。
步骤S202至步骤S207分别与步骤S101至步骤S106相同，这里不再赘述。
图3为本申请实施例提供的目标监控方法的第三种流程示意图，如图3所示，在一个实施例中，该方法包括以下几个步骤：
步骤S301，获取当前帧图像。
步骤S302，采用视频结构化算法进行分类识别。
步骤S303，确定待监控目标。
步骤S304，对待监控目标进行跟踪，调整摄像头的焦距和转动，获取该待监控目标的特写图像。
步骤S305，判断待监控目标与布控目标是否匹配。
若是，则执行步骤S306；若否，则变焦至全景图像，重新执行步骤S301。
步骤S306，发送报警信息至监控终端。
步骤S307，继续跟踪。
图3中的各个步骤与前述方法的内容相同，这里不再赘述。
在实施例一种示出和描述的所有示例中，任何具体值应被解释为仅仅是示例性的，而不是作为限制，因此，示例性实施例的其他示例可以具有不同的值。
实施例二：
图4为本申请实施例提供的目标监控装置的模块组成示意图，如图4所示，该目标监控装置包括：
图像获取模块10，用于获取摄像机的摄像头采集的当前帧图像。
对象获取模块20，用于对当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象。
在本实施例中，对象获取模块20具体用于：
根据预设分类规则对所述当前帧图像内的对象进行分类，所述预设分类规则包括将对象按照人物、车辆和动物分类；
从分类后的所述对象中提取所述当前布控目标对应类别的疑似对象。
待监控目标确定模块30，用于根据当前布控目标从获取到的疑似对象中确定当前帧图像中的待监控目标。
在本实施例中，待监控目标确定模块30具体配置为：
根据所述当前布控目标，采用视频结构化算法确定所述疑似对象的相似度；
将相似度最高的所述疑似对象确定为所述待监控目标。
跟踪拍摄模块40，用于控制摄像头对待监控目标进行跟踪拍摄，得到该待监控目标的特写图像。
在本实施例中，跟踪拍摄模块40具体用于：
调整所述摄像头的焦距，并采用目标跟踪算法和图像补偿算法确定所述待监控目标在跟踪拍摄的视频图像中的位置，根据所述位置采用云台控制算法控制所述摄像头的转动角度和转动速度，得到所述待监控目标的特写图像。
判断模块50，用于根据当前布控目标的特征对特写图像进行特征识别，判断待监控目标是否属于当前布控目标。
在本实施例中，判断模块50具体用于：
提取所述特写图像中的所述待监控目标的特征；
将所述待监控目标的特征与所述当前布控目标的特征进行对比，得到对比结果；
根据所述对比结果，判断所述待监控目标是否属于所述当前布控目标。
监控模块60，用于当判断模块50的判断结果为是时，将待监控目标确定为监控目标，并跟踪监控该监控目标。
请参照图5，在本实施例中，目标监控装置还可以包括报警信息发送模块70，报警信息发送模块70用于当判断所述待监控目标为所述当前布控目标时，发送报警信息至监控终端。
在本实施例中，目标监控装置还可以包括布控目标确定模块80，布控目标确定模块80用于当接收到监控终端发送的警戒目标信息时，将警戒目标信息中的对象确定为当前布控目标。
本申请实施例中，图像获取模块10获取摄像机的摄像头采集的当前帧图像；对象获取模块20对当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；待监控目标确定模块30根据当前布控目标从获取到的疑似对象中确定当前帧图像中的待监控目标；跟踪拍摄模块40控制该摄像头对待监控目标进行跟踪拍摄，得到待监控目标的特写图像；判断模块50根据当前布控目标的特征对该特写图像进行特征识别，判断待监控目标是否属于当前布控目标；如果是，则监控模块60将待监控目标确定为监控目标，并跟踪监控该监控目标。该目标监控装置可以从摄像机的摄像头 采集的当前帧图像中自动筛选出待监控目标，并自动判断该待监控目标是否属于当前布控目标，因此不依赖后端控制平台，节约了人力成本，且不容易出现漏防问题。同时，由于该装置应用于摄像机，直接在摄像机端处理该摄像机采集的视频文件，不再需要传送视频文件至后端处理，因此实时性强。综上可知，本申请实施例提供的目标监控装置降低了目标监控过程中的成本投入，并提升了目标监控的时效性和可靠性。
本申请实施例所提供的装置，其实现原理及产生的技术效果和前述方法实施例相同，为简要描述，装置实施例部分未提及之处，可参考前述方法实施例中相应内容。
附图中的流程图和框图显示了根据本申请的多个实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段或代码的一部分，所述模块、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个连续的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或动作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
实施例三：
图6为本申请实施例提供的摄像机的结构示意图，参见图6，本申请实施例还提供一种摄像机，包括：云台100、与云台100连接的云台电机200、设置在云台100上的摄像头300、存储器500、处理器400及存储在存储器500上并可在处理器400上运行的计算机程序。云台电机200、摄像头300分别与处理器400电连接，处理器400执行该计算机程序时实现上述实施例一的方法的步骤，其中，处理器400通过控制云台电机200带动摄像头300转动。
优选地，存储器500、处理器400集成在嵌入式处理芯片内。
上述摄像机还包括总线600和通信接口700，处理器400、通信接口700和存储器500通过总线600连接。
其中，存储器500可能包含高速随机存取存储器(RAM，Random Access Memory)，也可能还包括非不稳定的存储器(non-volatile memory)，例 如至少一个磁盘存储器。通过至少一个通信接口700(可以是有线或者无线)实现该系统网元与至少一个其他网元之间的通信连接，可以使用互联网，广域网，本地网，城域网等。
总线600可以是ISA总线、PCI总线或EISA总线等。总线600可以分为地址总线、数据总线、控制总线等。为便于表示，图5中仅用一个双向箭头表示，但并不表示仅有一根总线或一种类型的总线。
其中，存储器500用于存储计算机程序，所述处理器400在接收到执行指令后，执行所述计算机程序，前述本申请实施例任一实施例揭示的流过程定义的装置所执行的方法可以应用于处理器400中，或者由处理器400实现。
处理器400可能是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过处理器400中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器400可以是通用处理器，包括中央处理器(Central Processing Unit，简称CPU)、网络处理器(Network Processor，简称NP)等；还可以是数字信号处理器(Digital Signal Processing，简称DSP)、专用集成电路(Application Specific Integrated Circuit，简称ASIC)、现成可编程门阵列(Field-Programmable Gate Array，简称FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器500，处理器400读取存储器500中的信息，结合其硬件完成上述方法的步骤。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的摄像机的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
本申请实施例提供的目标监控装置及摄像机，与上述实施例提供的目标监控方法具有相同的技术特征，所以也能解决相同的技术问题，达到相同的技术效果。
实施例四：
本申请实施例还提供了一种具有处理器可执行的非易失的程序代码的计算机可读介质，所述程序代码包括的指令可用于执行前面方法实施例中所述的方法，具体实现可参见方法实施例，在此不再赘述。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
在本申请所提供的几个实施例中，应该理解到，所揭露的方法、装置和摄像机，可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，又例如，多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
最后应说明的是：以上所述实施例，仅为本申请的具体实施方式，用以说明本申请的技术方案，而非对其限制，本申请的保护范围并不局限于此，尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变 化，或者对其中部分技术特征进行等同替换；而这些修改、变化或者替换，并不使相应技术方案的本质脱离本申请实施例技术方案的精神和范围，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应所述以权利要求的保护范围为准。
工业实用性
本申请实施例提供的目标监控方法、目标监控装置、摄像机及计算机可读介质降低了目标监控过程中的成本投入，并提升了目标监控的时效性和可靠性。

Claims (17)

  1. 一种目标监控方法，其特征在于，所述方法应用于摄像机，包括：
    获取所述摄像机的摄像头采集的当前帧图像；
    对所述当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；
    根据所述当前布控目标从获取到的所述疑似对象中确定所述当前帧图像中的待监控目标；
    控制所述摄像头对所述待监控目标进行跟踪拍摄，得到所述待监控目标的特写图像；
    根据所述当前布控目标的特征对所述特写图像进行特征识别，判断所述待监控目标是否属于所述当前布控目标；
    如果是，则将所述待监控目标确定为监控目标，并跟踪监控所述监控目标。
  2. 根据权利要求1所述的方法，其特征在于，所述对所述当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象，包括：
    根据预设分类规则对所述当前帧图像内的对象进行分类，所述预设分类规则包括将对象按照人物、车辆和动物分类；
    从分类后的所述对象中提取所述当前布控目标对应类别的疑似对象。
  3. 根据权利要求1所述的方法，其特征在于，所述根据所述当前布控目标从获取到的所述疑似对象中确定所述当前帧图像中的待监控目标，包括：
    根据所述当前布控目标，采用视频结构化算法确定所述疑似对象的相似度；
    将相似度最高的所述疑似对象确定为所述待监控目标。
  4. 根据权利要求1所述的方法，其特征在于，所述控制所述摄像头对所述待监控目标进行跟踪拍摄，得到所述待监控目标的特写图像，包括：
    调整所述摄像头的焦距，并采用目标跟踪算法和图像补偿算法确定所述待监控目标在跟踪拍摄的视频图像中的位置，根据所述位置采用云台控制算法控制所述摄像头的转动角度和转动速度，得到所述待监控目标的特写图像。
  5. 根据权利要求1所述的方法，其特征在于，所述根据所述当前布控 目标的特征对所述特写图像进行特征识别，判断所述待监控目标是否属于所述当前布控目标，包括：
    提取所述特写图像中的所述待监控目标的特征；
    将所述待监控目标的特征与所述当前布控目标的特征进行对比，得到对比结果；
    根据所述对比结果，判断所述待监控目标是否属于所述当前布控目标。
  6. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    当判断所述待监控目标为所述当前布控目标时，发送报警信息至监控终端。
  7. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    当接收到监控终端发送的警戒目标信息时，将所述警戒目标信息中的对象确定为所述当前布控目标。
  8. 根据权利要求1-7中任意一项所述的方法，其特征在于，在获取所述摄像机的摄像头采集的当前帧图像的步骤之前，所述方法还包括：
    当接收到监控终端发送的警戒目标信息时，将警戒目标信息中的对象确定为当前布控目标。
  9. 一种目标监控装置，其特征在于，应用于摄像机，包括：
    图像获取模块，配置成获取所述摄像机的摄像头采集的当前帧图像；
    对象获取模块，配置成对所述当前帧图像内的对象进行分类，获取当前布控目标对应类别的疑似对象；
    待监控目标确定模块，配置成根据所述当前布控目标从获取到的所述疑似对象中确定所述当前帧图像中的待监控目标；
    跟踪拍摄模块，配置成控制所述摄像头对所述待监控目标进行跟踪拍摄，得到所述待监控目标的特写图像；
    判断模块，配置成根据所述当前布控目标的特征对所述特写图像进行特征识别，判断所述待监控目标是否属于所述当前布控目标；
    监控模块，配置成当所述判断模块的判断结果为是时，将所述待监控目标确定为监控目标，并跟踪监控所述监控目标。
  10. 如权利要求9所述的目标监控装置，其特征在于，所述对象获取模块具体配置为：
    根据预设分类规则对所述当前帧图像内的对象进行分类，所述预设分类规则包括将对象按照人物、车辆和动物分类；
    从分类后的所述对象中提取所述当前布控目标对应类别的疑似对象。
  11. 如权利要求9所述的目标监控装置，其特征在于，所述待监控目标确定模块具体配置为：
    根据所述当前布控目标，采用视频结构化算法确定所述疑似对象的相似度；
    将相似度最高的所述疑似对象确定为所述待监控目标。
  12. 如权利要求9所述的目标监控装置，其特征在于，所述跟踪拍摄模块具体配置成：
    调整所述摄像头的焦距，并采用目标跟踪算法和图像补偿算法确定所述待监控目标在跟踪拍摄的视频图像中的位置，根据所述位置采用云台控制算法控制所述摄像头的转动角度和转动速度，得到所述待监控目标的特写图像。
  13. 如权利要求9所述的目标监控装置，其特征在于，所述判断模块具体配置成：
    提取所述特写图像中的所述待监控目标的特征；
    将所述待监控目标的特征与所述当前布控目标的特征进行对比，得到对比结果；
    根据所述对比结果，判断所述待监控目标是否属于所述当前布控目标。
  14. 如权利要求9所述的目标监控装置，其特征在于，所述装置还包括：
    报警信息发送模块，配置成当判断所述待监控目标为所述当前布控目标时，发送报警信息至监控终端。
  15. 如权利要求9-14中任意一项所述的目标监控装置，其特征在于，所述装置还包括：
    布控目标确定模块，配置成当接收到监控终端发送的警戒目标信息时，将警戒目标信息中的对象确定为当前布控目标。
  16. 一种摄像机，其特征在于，包括云台、与所述云台连接的云台电机、设置在所述云台上的摄像头、存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序；
    所述云台电机、所述摄像头分别与所述处理器电连接，所述处理器执行所述计算机程序时实现上述权利要求1-8任一项所述的方法的步骤，其中，所述处理器通过控制所述云台电机带动所述摄像头转动。
  17. 一种具有处理器可执行的非易失的程序代码的计算机可读介质，其特征在于，所述程序代码使所述处理器执行所述权利要求1-8任一所述方法。
PCT/CN2018/089945 2017-06-09 2018-06-05 目标监控方法、目标监控装置、摄像机及计算机可读介质 WO2018223955A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710435936.0 2017-06-09
CN201710435936.0A CN107105207A (zh) 2017-06-09 2017-06-09 目标监控方法、目标监控装置及摄像机

Publications (1)

Publication Number Publication Date
WO2018223955A1 true WO2018223955A1 (zh) 2018-12-13

Family

ID=59660134

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/089945 WO2018223955A1 (zh) 2017-06-09 2018-06-05 目标监控方法、目标监控装置、摄像机及计算机可读介质

Country Status (2)

Country Link
CN (1) CN107105207A (zh)
WO (1) WO2018223955A1 (zh)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111460977A (zh) * 2020-03-30 2020-07-28 广东电网有限责任公司电力科学研究院 一种跨视域人员再识别方法、装置、终端及存储介质
CN111626165A (zh) * 2020-05-15 2020-09-04 安徽江淮汽车集团股份有限公司 行人识别方法、装置、电子设备及存储介质
CN111860063A (zh) * 2019-04-30 2020-10-30 杭州海康威视数字技术股份有限公司 步态数据构建系统、方法及装置
CN112241649A (zh) * 2019-07-16 2021-01-19 浙江宇视科技有限公司 一种目标识别方法及装置
CN112580469A (zh) * 2020-12-10 2021-03-30 上海慈航建设发展有限公司 违法视频的审核、管理系统、方法、设备及存储介质
CN112883815A (zh) * 2021-01-26 2021-06-01 首都师范大学 一种农机具自动监管方法、系统及服务器
CN113111685A (zh) * 2020-01-10 2021-07-13 杭州海康威视数字技术股份有限公司 追踪系统、追踪数据的采集/处理方法和装置
CN113421431A (zh) * 2021-06-17 2021-09-21 京东方科技集团股份有限公司 应急通道监测方法及装置
CN113487727A (zh) * 2021-07-14 2021-10-08 广西民族大学 一种三维建模系统、装置及方法
CN114549876A (zh) * 2022-01-10 2022-05-27 上海明胜品智人工智能科技有限公司 图像处理方法、设备以及系统
CN115082836A (zh) * 2022-07-23 2022-09-20 深圳神目信息技术有限公司 一种行为识别辅助的目标物体检测方法及装置
CN115457449A (zh) * 2022-11-11 2022-12-09 深圳市马博士网络科技有限公司 一种基于ai视频分析和监控安防的预警系统
CN117255180A (zh) * 2023-11-20 2023-12-19 山东通广电子股份有限公司 一种智能安全监控设备及监控方法

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107105207A (zh) * 2017-06-09 2017-08-29 北京深瞐科技有限公司 目标监控方法、目标监控装置及摄像机
CN107569848B (zh) * 2017-08-30 2020-08-04 武汉斗鱼网络科技有限公司 一种游戏分类方法、装置及电子设备
CN107742422A (zh) * 2017-10-11 2018-02-27 公安部第三研究所 道路车辆违章监测系统及其方法
CN108038408A (zh) * 2017-10-13 2018-05-15 西安天和防务技术股份有限公司 智能跟踪方法及装置、存储介质及电子终端
CN108174102A (zh) * 2018-01-05 2018-06-15 北京红云融通技术有限公司 一种获取动态对象的图像的方法以及系统
CN108921001B (zh) * 2018-04-18 2019-07-02 特斯联（北京）科技有限公司 一种采用人工智能预测追踪的视频监视云台及其追踪方法
CN110599721A (zh) * 2018-06-13 2019-12-20 杭州海康威视数字技术股份有限公司 一种监控方法、装置、系统及监控设备
CN110706247B (zh) * 2018-06-22 2023-03-07 杭州海康威视数字技术股份有限公司 一种目标跟踪方法、装置及系统
CN109102531A (zh) * 2018-08-21 2018-12-28 北京深瞐科技有限公司 一种目标轨迹追踪方法及装置
CN109024370A (zh) * 2018-10-16 2018-12-18 青岛特锐德电气股份有限公司 一种车位锁控制方法和系统
CN110378189A (zh) * 2019-04-22 2019-10-25 北京旷视科技有限公司 一种监控布置方法、装置、终端及存储介质
CN112085927B (zh) * 2019-06-15 2022-09-27 湖北讯泰通信电子有限公司 可靠式目标监控系统
CN110266984B (zh) * 2019-07-01 2020-12-18 浙江大学 一种云台摄像智能分析教学录播一体机
CN112180818A (zh) * 2019-07-05 2021-01-05 深圳市华创技术有限公司 一种智能ai装置及控制方法
CN111402297A (zh) * 2020-03-25 2020-07-10 杭州博雅鸿图视频技术有限公司 目标追踪检测方法、系统、电子设备及存储介质
CN112132315A (zh) * 2020-08-18 2020-12-25 华为技术有限公司 目标对象的逃逸路线预测方法和布控平台
CN112861853A (zh) * 2021-01-21 2021-05-28 济南浪潮高新科技投资发展有限公司 基于知识图谱的车辆识别方法和装置
CN113507577A (zh) * 2021-07-07 2021-10-15 杭州海康威视系统技术有限公司 目标对象检测方法、装置、设备及存储介质
CN113724324B (zh) * 2021-08-30 2023-12-19 杭州华橙软件技术有限公司 云台的控制方法、装置、存储介质及电子装置
CN113665495A (zh) * 2021-09-23 2021-11-19 温州大学大数据与信息技术研究院 一种便于安装的基于计算机视觉的车载目标跟踪器
CN114040115A (zh) * 2021-11-29 2022-02-11 海南哦课教育科技有限公司 目标对象异常动作的捕获方法、装置、介质和电子设备
CN114419555A (zh) * 2022-01-19 2022-04-29 广东飞达交通工程有限公司 一种基于深度卷积神经网络的道路交通目标跟踪方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005050972A2 (en) * 2003-11-13 2005-06-02 Objectvideo, Inc. Active camera video-based surveillance systems and methods
CN101626489A (zh) * 2008-07-10 2010-01-13 苏国政 无人值守目标智能识别与自动跟踪的方法及其系统
WO2010090621A1 (en) * 2009-02-09 2010-08-12 Vitamin D, Inc. Systems and methods for video monitoring
CN102968802A (zh) * 2012-11-28 2013-03-13 无锡港湾网络科技有限公司 一种基于视频监控的运动目标分析跟踪方法及系统
CN106600977A (zh) * 2017-02-13 2017-04-26 深圳英飞拓科技股份有限公司 基于多特征识别的违停检测方法及系统
CN107105207A (zh) * 2017-06-09 2017-08-29 北京深瞐科技有限公司 目标监控方法、目标监控装置及摄像机

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201278180Y (zh) * 2008-05-28 2009-07-22 丁国锋 一种自动追踪识别系统
CN101635834A (zh) * 2008-07-21 2010-01-27 侯荣琴 类神经控制自动追踪识别系统
CN103679156A (zh) * 2013-12-31 2014-03-26 武汉理工大学 一种多类运动目标自动识别与跟踪的方法
CN104506819B (zh) * 2015-01-06 2018-06-15 成都新舟锐视科技有限公司 一种多摄像头实时联动的互反馈跟踪系统及方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005050972A2 (en) * 2003-11-13 2005-06-02 Objectvideo, Inc. Active camera video-based surveillance systems and methods
CN101626489A (zh) * 2008-07-10 2010-01-13 苏国政 无人值守目标智能识别与自动跟踪的方法及其系统
WO2010090621A1 (en) * 2009-02-09 2010-08-12 Vitamin D, Inc. Systems and methods for video monitoring
CN102968802A (zh) * 2012-11-28 2013-03-13 无锡港湾网络科技有限公司 一种基于视频监控的运动目标分析跟踪方法及系统
CN106600977A (zh) * 2017-02-13 2017-04-26 深圳英飞拓科技股份有限公司 基于多特征识别的违停检测方法及系统
CN107105207A (zh) * 2017-06-09 2017-08-29 北京深瞐科技有限公司 目标监控方法、目标监控装置及摄像机

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111860063B (zh) * 2019-04-30 2023-08-11 杭州海康威视数字技术股份有限公司 步态数据构建系统、方法及装置
CN111860063A (zh) * 2019-04-30 2020-10-30 杭州海康威视数字技术股份有限公司 步态数据构建系统、方法及装置
CN112241649A (zh) * 2019-07-16 2021-01-19 浙江宇视科技有限公司 一种目标识别方法及装置
CN113111685A (zh) * 2020-01-10 2021-07-13 杭州海康威视数字技术股份有限公司 追踪系统、追踪数据的采集/处理方法和装置
CN111460977B (zh) * 2020-03-30 2024-02-20 广东电网有限责任公司电力科学研究院 一种跨视域人员再识别方法、装置、终端及存储介质
CN111460977A (zh) * 2020-03-30 2020-07-28 广东电网有限责任公司电力科学研究院 一种跨视域人员再识别方法、装置、终端及存储介质
CN111626165B (zh) * 2020-05-15 2024-02-02 安徽江淮汽车集团股份有限公司 行人识别方法、装置、电子设备及存储介质
CN111626165A (zh) * 2020-05-15 2020-09-04 安徽江淮汽车集团股份有限公司 行人识别方法、装置、电子设备及存储介质
CN112580469A (zh) * 2020-12-10 2021-03-30 上海慈航建设发展有限公司 违法视频的审核、管理系统、方法、设备及存储介质
CN112883815A (zh) * 2021-01-26 2021-06-01 首都师范大学 一种农机具自动监管方法、系统及服务器
CN112883815B (zh) * 2021-01-26 2022-11-29 首都师范大学 一种农机具自动监管方法、系统及服务器
CN113421431A (zh) * 2021-06-17 2021-09-21 京东方科技集团股份有限公司 应急通道监测方法及装置
CN113421431B (zh) * 2021-06-17 2022-12-02 京东方科技集团股份有限公司 应急通道监测方法及装置
CN113487727B (zh) * 2021-07-14 2022-09-02 广西民族大学 一种三维建模系统、装置及方法
CN113487727A (zh) * 2021-07-14 2021-10-08 广西民族大学 一种三维建模系统、装置及方法
CN114549876A (zh) * 2022-01-10 2022-05-27 上海明胜品智人工智能科技有限公司 图像处理方法、设备以及系统
CN115082836B (zh) * 2022-07-23 2022-11-11 深圳神目信息技术有限公司 一种行为识别辅助的目标物体检测方法及装置
CN115082836A (zh) * 2022-07-23 2022-09-20 深圳神目信息技术有限公司 一种行为识别辅助的目标物体检测方法及装置
CN115457449A (zh) * 2022-11-11 2022-12-09 深圳市马博士网络科技有限公司 一种基于ai视频分析和监控安防的预警系统
CN117255180A (zh) * 2023-11-20 2023-12-19 山东通广电子股份有限公司 一种智能安全监控设备及监控方法
CN117255180B (zh) * 2023-11-20 2024-02-09 山东通广电子股份有限公司 一种智能安全监控设备及监控方法

Also Published As

Publication number Publication date
CN107105207A (zh) 2017-08-29

Similar Documents

Publication Publication Date Title
WO2018223955A1 (zh) 目标监控方法、目标监控装置、摄像机及计算机可读介质
CN109166261B (zh) 基于图像识别的图像处理方法、装置、设备及存储介质
US10599929B2 (en) Event monitoring with object detection systems
WO2022105243A1 (zh) 事件检测方法、装置、电子设备及存储介质
US9843611B2 (en) Incident data collection for public protection agencies
GB2542686A (en) Video data analyzing method and apparatus and parking lot monitoring system
Zabłocki et al. Intelligent video surveillance systems for public spaces–a survey
CN110580808B (zh) 一种信息处理方法、装置、电子设备及智能交通系统
US11990035B2 (en) Infringement detection method, device and system
US20160210759A1 (en) System and method of detecting moving objects
US11416542B1 (en) System and method for uploading still images of matching plates in response to an alert hit list using distributed LPR acquisition
CN109815839B (zh) 微服务架构下的徘徊人员识别方法及相关产品
WO2022198897A1 (zh) 一种路侧停车的管理方法及装置
CN113112813B (zh) 违章停车检测方法及装置
Carrell et al. Identification of driver phone usage violations via state-of-the-art object detection with tracking
CN105702046A (zh) 一种智能化违章停车自动监测系统
CN112601049A (zh) 视频的监控方法、装置、计算机设备及存储介质
CN114913470B (zh) 一种事件检测方法及装置
CN112435479B (zh) 目标物体违章检测方法、装置、计算机设备及系统
US11917325B2 (en) Automated scope limiting for video analytics
CN115174889A (zh) 摄像机的位置偏移检测方法、电子设备和存储介质
WO2021149274A1 (ja) 監視システム、監視装置、監視方法、およびプログラム
KR101327256B1 (ko) Ptz 카메라에 의한 차량 그림자 영역 탐지를 이용한 차량 검지 시스템 및 방법
CN110581979B (zh) 一种图像采集系统、方法及装置
KR20210103210A (ko) 영상처리장치 및 그 장치의 구동방법

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: 18813217

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: 18813217

Country of ref document: EP

Kind code of ref document: A1