Background
In recent years, with the rapid development of computers, networks, image processing and transmission technologies, video surveillance has become an important component of security systems. The video monitoring is widely applied to numerous occasions due to the characteristics of intuition, accuracy, timeliness and rich information content, and has remarkable effect on the aspects of public security, prevention and control, city management, social management and the like. The camera is a data acquisition device for video monitoring and completes video image signal acquisition. However, when a camera in a conventional video monitoring system is installed, a focal length is fixed through manual focusing, so that an area that the camera can shoot is also limited, and for a moving object which appears in a monitoring area and is far away, the camera cannot obtain clear picture information of the moving object. Along with the development of security and protection technology, monitoring equipment is more and more intelligent, and the integration camera can move the camera lens to the moving target that needs the control with the help of the variable ball machine of high-speed rotation, through adjusting the focus of camera lens in order to enlarge or reduce the size of moving target in the control picture.
However, the monitoring action of the integrated camera needs to be executed under the manual operation of an operator, and the requirement of real-time monitoring cannot be met. Therefore, there is a need for a device and a method for automatically controlling the moving lens of some cameras closest to a moving target and adjusting the focal length of the cameras when the moving target appears in a monitored area, so as to obtain a clear monitoring picture of the moving target without affecting the normal monitoring task of other cameras.
Disclosure of Invention
The invention provides a multi-camera intelligent monitoring method and device for monitoring a moving target, which can autonomously judge the moving target and distribute a monitoring task, and realize unmanned intelligent automatic focusing monitoring.
In order to achieve the above object, the present invention is achieved by the following technical solutions.
A multi-camera intelligent monitoring method for monitoring a moving target comprises the following steps:
the monitoring system has m (m is more than 1) cameras. When the camera monitors, whether a moving target exists in a target area is monitored; if so, measuring the distance information between the moving target and the moving target, and calculating the size of the moving target according to information such as video pictures, focal length, distance and the like. The distance information is measured by a distance measuring device of the camera.
The camera head transmits the obtained distance information and the size information of the moving target to other camera heads which are in communication connection with the camera head, the other camera heads determine whether the other camera heads are one of the front n (n < m) camera heads closest to the moving target through a judgment algorithm according to the obtained distance information, and if so, a lens moving instruction and a focusing instruction are generated according to the distance information and the size information of the moving target and executed.
And when the moving target moves out of the monitoring range, the camera executing the lens moving instruction and the focusing instruction restores the initial working state.
And if the difference value of the current monitoring area picture and the historical monitoring area picture is greater than a set threshold value, determining that the moving target exists, and adding a monitoring identifier for the target.
The camera communication connection adopts a switching network topology structure, because in a monitoring system, each camera is limited by factors such as spacing distance, communication power and the like, the monitoring system usually does not have global communication capability, and in addition, the switching network topology structure is more suitable for the complex large-scale working environment in consideration of the conditions such as fault interference of the camera, change of system layout and the like.
The algorithm for determining whether the first n cameras closest to the moving target are determined is to firstly define a piecewise function f (x), wherein f (x) is 0 when x is less than 0, f (x) is 1 when x is greater than 1, and f (x) is x otherwise; defining the communication connection matrix C of the camera to be an m × m matrix, CijIs the element of the ith row and the jth column in the matrix C, which represents the communication connection weight of the ith camera and the first camera, when the two cameras are in communication connection, C ij1, otherwise C ij0; whether the camera is one of the first n cameras closest to the moving target is judged by the following formula:
wherein alpha is
i、β
i、θ
iThe initial value is randomly given for the auxiliary variable, j belongs to M (i) and represents the number of the camera with the weight value of 1 connected with the ith ventilation opening; d
iThe distance from the ith camera to a moving target object is represented, lambda is a sampling interval, b is a design parameter and is required to be as small as possible in practical application; gamma, c
0、c
1Parameters which are all larger than 0, and the superscript k represents the iteration times; d represents a communication delay; y is
iFor switching the network topology compensation function, y when the network topology is not changed
i0 remains unchanged, and if the network topology changes,
representing the difference value of a camera communication connection matrix C in the monitoring system before and after switching when the communication topology network is switched; omega
iIndicating whether the ith camera needs to execute a specific task of monitoring a moving target, namely executing a lens moving instruction and a focusing instruction, and after a certain number of iterations, omega
iHas an output value of 1 or 0,
ω i1 represents that the camera needs to change the working state, omega
iAnd 0 means that the camera maintains the initial working state.
And the lens moving instruction is used for adjusting the rotation angle of the high-speed dome camera to move the lens to the moving target.
The focusing instruction is to adjust the focal length of the camera lens to zoom in or zoom out the moving object.
The invention also provides a multi-camera intelligent monitoring device for monitoring the moving target, which comprises:
the data acquisition module is used for capturing the current monitoring area picture; the distance measuring device is used for measuring the distance between the camera and the moving target;
the communication module is used for being in wireless connection with other cameras and sharing the size information of the moving target and the distance information between the moving target and the cameras;
the processor module comprises a computing unit and a storage unit; calculating a difference value between a current monitoring picture and a historical monitoring picture, comparing the difference value with a set threshold value, calculating the size of a moving target according to information such as a video picture, a focal length, a distance and the like, determining whether the moving target needs to change the working state by using a judgment algorithm, and generating a lens moving instruction and a focusing instruction according to the target size and the distance information; the storage unit is used for storing an algorithm program and related monitoring parameters;
and the execution module is used for executing the lens moving instruction and the focusing instruction.
The method specifically comprises the following steps: the monitoring system is provided with m (m is more than 1) cameras, the processor module computing unit compares the difference value between the current monitoring picture and the historical monitoring area picture with a set threshold value, if the difference value is larger than the set threshold value, a moving target is determined to appear in the monitoring area, the data acquisition module measures the distance between the cameras and the moving target, the computing unit determines the size of the moving target through the information, the communication module shares the distance information between the cameras and the moving target with other cameras in communication connection with the communication module, and each camera computing unit determines whether the camera computing unit needs to change the working state by using a determination algorithm. The camera needing to change the working state is executed by the execution module by utilizing the lens moving instruction and the focusing instruction determined by the calculation unit. And after the moving target moves out of the monitoring area, the camera executing the instruction restores the initial working state.
The invention provides the multi-camera intelligent monitoring method and the device for monitoring the moving target, so that the camera can automatically complete the monitoring task after the moving target is found, a clear monitoring video is obtained, meanwhile, a switching network topological structure is adopted, the problem of communication connection among the cameras due to factors such as environmental interference, single communication fault and sensor limitation is effectively solved, the robustness of a monitoring system is enhanced, the scale extensibility of the camera system is improved, the limitation of the communication distance is effectively overcome, and the automatic distribution of the camera monitoring task in the monitoring system is realized.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings and examples.
The invention provides a multi-camera intelligent monitoring method and device for monitoring a moving target, which can autonomously judge the moving target and distribute a monitoring task, and realize unmanned intelligent automatic focusing monitoring.
In the present embodiment, the camera is an integrated camera with a high-speed dome camera.
A multi-camera intelligent monitoring method for moving object monitoring is shown in fig. 1:
s1: a certain monitoring system has m (m is more than 1) cameras, when the cameras monitor, whether a moving target exists in a monitoring area is monitored, if so, the target is marked, the distance between the target and the moving target is measured, and the size of the moving target is calculated;
s2: transmitting the distance information and the size information of the moving target to other cameras in communication connection with the moving target, and determining whether the camera is one of the first n (n < m) cameras closest to the moving target according to a judgment algorithm;
s3: if so, generating a lens moving instruction and a focusing instruction according to the distance information and the moving target size information and executing;
s4: when the moving object moves out of the monitoring range, the camera that executed the lens movement instruction and the focusing instruction in S3 resumes the initial operating state.
The determination algorithm in step S2 is as follows:
first, a piecewise function f (x) is defined, where f (x) is 0 when x < 0, and f (x) is 0 when x > 1x) 1, otherwise, f (x) x; defining the communication connection matrix C of the camera to be an m × m matrix, CijIs the element of the ith row and the jth column in the matrix C, which represents the communication connection weight of the ith camera and the jth camera, when the two cameras are in communication connection, C ij1, otherwise C ij0; whether the camera is one of the first n cameras closest to the moving target is judged by the following formula:
wherein alpha is
i、β
i、θ
iThe initial value is randomly given as an auxiliary variable, j belongs to M (i) and represents the number of the camera with the weight value of 1 connected with the ith camera; d
iThe distance between the ith camera and the moving target object is represented, λ is a sampling interval, λ is 0.01, and the parameter γ is 10000 and c
0=5、
c 15, b is 0.01, and the superscript k represents the number of iterations; d represents a communication delay; y is
iFor switching the network topology compensation function, y when the network topology is not changed
i0 remains unchanged, and if the network topology changes,
ΔC
ijrepresenting the difference value of a camera communication connection matrix delta C in the monitoring system before and after switching when the communication topology network is switched; omega
iIndicating whether the ith camera needs to execute a specific task of monitoring a moving target, namely executing a lens moving instruction and a focusing instruction, and after a certain number of iterations, omega
iHas an output value of 1 or 0,
ω i1 represents that the camera needs to change the working state, omega
iAnd 0 means that the camera maintains the initial working state.
And if the difference value of the current monitoring area picture and the historical monitoring area picture is greater than a set threshold value, determining that the moving target exists, and adding a monitoring identifier for the target.
The camera communication connection adopts a switching network topology structure, because in a monitoring system, each camera is limited by factors such as spacing distance, communication power and the like, the monitoring system usually does not have global communication capability, and in addition, the switching network topology structure is more suitable for the complex large-scale working environment in consideration of the conditions such as fault interference of the camera, change of system layout and the like.
A many cameras intelligent monitoring device for moving object monitoring includes:
the data acquisition module is used for capturing the current monitoring area picture; the distance measuring device is used for measuring the distance between the camera and the moving target;
the communication module is used for being in wireless connection with other cameras and sharing the size information of the moving target and the distance information between the moving target and the cameras;
the processor module comprises a computing unit and a storage unit; calculating a difference value between a current monitoring picture and a historical monitoring picture, comparing the difference value with a set threshold value, calculating the size of a moving target according to information such as a video picture, a focal length, a distance and the like, determining whether the moving target needs to change the working state by using a judgment algorithm, and generating a lens moving instruction and a focusing instruction according to the target size and the distance information; the storage unit is used for storing an algorithm program and related monitoring parameters;
and the execution module is used for executing the lens moving instruction and the focusing instruction.
As shown in fig. 2, the working diagram of the multi-camera intelligent monitoring device for monitoring a moving object is that 20 cameras are provided in the monitoring system, the processor module computing unit compares the difference value between the current monitoring picture and the historical monitoring area picture with the set threshold value, and finds the moving
object 0 if the difference value is larger than the set threshold value. The data acquisition module measures the distance between the data acquisition module and the moving target, the calculation unit determines the size of the moving target according to the information, the communication module shares the distance information between the data acquisition module and the moving target with other cameras in communication connection with the communication module, and each camera calculation unit determines whether the camera calculation unit needs to change the working state or not by using a judgment algorithm, namely whether the camera calculation unit is one of 2 cameras closest to the moving
target 0 or not. And determining that the working states of the
camera 6 and the
camera 10 need to be changed according to the algorithm, and executing the working states by using the lens moving instruction and the focusing instruction determined by the calculation unit through an execution module. As shown in the working schematic diagram of the multi-camera intelligent monitoring device for monitoring a moving object in fig. 3, in a certain monitoring period, the communication connection between the
camera 5 and the
camera 10 is interrupted, the communication connection between the camera 12 and the
camera 6 is interrupted due to a fault, and the
camera 5 is connected with the
camera 9. At the moment, the communication network topology structure changes, and when the judgment calculation is carried out, a switching network topology structure compensation function y needs to be added
i,

And after the moving target moves out of the monitoring area, the camera executing the instruction restores the initial working state. The invention provides the multi-camera intelligent monitoring method and the device for monitoring the moving target, so that the camera can automatically complete the monitoring task after the moving target is found, a clear monitoring video is obtained, meanwhile, a switching network topological structure is adopted, the communication connection problem caused by factors such as environmental interference, single communication fault, sensor limitation and the like among the cameras is effectively avoided, the robustness of a monitoring system is enhanced, the scale extensibility of the camera system is improved, the limitation of the communication distance is effectively overcome, and the camera intelligent monitoring method and the device for monitoring the moving target realize the camera intelligent monitoring in the monitoring systemAutomatic allocation of head monitoring tasks.