CN108632569A - Video frequency monitoring method and device based on the linkage of rifle ball - Google Patents

Abstract

This application discloses a kind of video frequency monitoring methods and device based on the linkage of rifle ball, belong to video analysis field.The method includes：The characteristic information of the central point in target object region shared in the first video pictures is obtained, and searches the matched pixel of characteristic information of characteristic information and acquisition in the second video pictures, the pixel that then basis is found determines the first rotational angle.Since the first rotational angle is the angle that ball-type IPC need to be rotated when the central point of the second video pictures is turned to the central point in target object region shared in the second video pictures, therefore, the central point of video pictures that ball-type IPC is acquired later according to the rotation of the first rotational angle will be for the central point in target object region shared in the second video pictures, it avoids the occurrence of because of the case where ball-type IPC problem of aging causes target object that may be not at the center for the video pictures that ball-type IPC is acquired, to improve the effect being monitored to the details of target object.

Description

Video surveillance method and device based on gun-ball linkage

technical field

The present application relates to the field of video analysis, in particular to a video monitoring method and device based on gun-ball linkage.

Background technique

Video monitoring refers to sending the collected video images to the client through an Internet Protocol Camera (IPC), so that the client can view the video images. At present, IPCs are mainly divided into gun-type IPCs and ball-type IPCs. Since gun-type IPCs are mainly used to collect large-scale video images, while spherical-type IPCs are mainly used to collect small-scale and high-precision video images, the For the video monitoring of the target object, in order to not only monitor the movement of the target object as a whole, but also monitor the details of the target object, it is usually necessary to adopt the video monitoring method of gun-ball linkage, that is, the server simultaneously collects the gun-type IPC The video screen and the video screen collected by the ball-type IPC configured by the server for the gun-type IPC are sent to the client to realize the video monitoring of the target object based on the gun-ball linkage.

Currently, when performing video surveillance based on gun-ball linkage, the server pre-evenly divides the video images collected by the gun-type IPC into a grid of m rows and n columns. Determine the current grid of the target object in the n-column grid, and control the spherical IPC to collect video images in the grid. When the target object is not located at the center of the grid, the center of the video picture collected by the spherical IPC will not be the target object, resulting in poor video monitoring of the target object through the spherical IPC. Therefore, the server also needs to Determine 4 calibration points from the grid, and determine the offset that the spherical IPC needs to rotate when the center point of the video picture collected by the spherical IPC is moved from the center point of the grid to the 4 calibration points, Get the offset of the 4 calibration points. Afterwards, when the server detects that the target object moves from one position to another within the grid, it selects 3 calibration points that are relatively close to the position after the target object moves from the 4 calibration points, based on the The offset of the 3 calibration points determines the offset that the spherical IPC needs to rotate when the center point of the video picture collected by the spherical IPC is moved from the center point of the grid to the other position, and controls the spherical IPC Rotate according to the offset, so as to ensure that the target object is located at the center of the video image collected by the spherical IPC, so as to realize real-time monitoring of the target object through the spherical IPC after the position of the target object changes.

Since the above method is to determine the offset that the spherical IPC needs to rotate through theoretical calculations, but in practical applications, with the increase in the use time of the spherical IPC, problems such as aging of the spherical IPC’s belt will occur, resulting in spherical IPC When the IPC is offset according to the theoretical offset, there will be an error with the actual offset. Therefore, when performing video monitoring according to the above method, the target object may not be located at the center of the video image collected by the spherical IPC, thereby reducing the effect of monitoring the details of the target object through the spherical IPC.

Contents of the invention

In order to solve the phenomenon in the prior art that the spherical IPC is shifted according to the theoretical offset due to problems such as aging, the target object may not be located in the center of the video picture collected by the spherical IPC, the application provides a method based on A video surveillance method and device for gun-ball linkage. Described technical scheme is as follows:

In the first aspect, a video surveillance method based on gun-ball linkage is provided, the method comprising:

Acquire the feature information of the center point of the area occupied by the target object in the first video picture, the first video picture is a video picture collected by the gun-type network camera IPC, and the feature information includes the pixel values of the center point respectively The pixel difference between the pixel value and the pixel value of a plurality of neighboring pixel points, the plurality of neighboring pixel points are the pixel points in the neighborhood of the central point in the first video picture;

Searching for pixel points whose feature information matches the feature information of the center point of the area occupied by the target object in the first video picture among the pixels included in the second video picture, the second video picture is for the first video picture The video screen currently collected by the spherical IPC configured with the gun-type IPC;

According to the longitude and latitude corresponding to the pixel points found in the spherical coordinate system of the spherical IPC, the first rotation angle of the spherical IPC is determined, and the first rotation angle is the rotation angle of the second video picture The angle at which the spherical IPC needs to be rotated when the center point is rotated to the center point of the area occupied by the target object in the second video frame;

sending a first rotation request to the spherical IPC, the first rotation request carrying the first rotation angle, and the first rotation request is used to instruct the spherical IPC to rotate according to the first rotation angle.

In the embodiment of the present invention, the pixel point whose feature information matches the feature information of the center point of the area occupied by the target object in the first video picture is directly searched in the second video picture, and the spherical IPC is controlled to convert the second video picture The center of the picture is rotated to the center point of the area occupied by the target object in the second video picture, that is, the found pixel point, so as to place the center point of the area occupied by the target object in the second video picture on the second video picture The center of the spherical IPC avoids the situation that the target object may not be located in the center of the video picture collected by the spherical IPC due to the aging problem of the spherical IPC, thereby improving the effect of monitoring the details of the target object through the spherical IPC.

Optionally, the searching pixel points included in the second video frame whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame includes:

In the second video frame, the center point of the second video frame is used as the center to determine the first target area whose area is a preset area, and the preset area is determined according to the aging degree of the spherical IPC ;

Finding, from all the pixels included in the first target area, the pixel points whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame.

Further, in order to improve the efficiency of searching for pixels whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame in the second video frame, the first video frame can be determined in the second video frame. The target area is to search only the pixel points whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame in the first target area.

Optionally, searching for pixels whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame from all the pixels included in the first target area, include:

Obtain feature information of all pixels included in the first target area;

For each pixel in all the pixels included in the first target area, an average operation and a range operation are performed on the pixel difference values included in the feature information of the pixel points to obtain a first average value and a first range difference value, and determine the product between the first average value and the first range value as the feature value of the pixel point;

performing an average operation and a range operation on the pixel difference values included in the feature information of the center point of the area occupied by the target object in the first video frame to obtain a second average value and a second range value, and The product between the second average value and the second range value is determined as a target feature value;

From the eigenvalues of all pixels included in the first target area, select the eigenvalue with the smallest difference with the target eigenvalue, and determine the pixel corresponding to the selected eigenvalue as the feature information and the set The pixel points that match the feature information of the center point of the area occupied by the target object in the first video frame.

In the embodiment of the present invention, to determine the pixel points whose feature information in the first target area matches the feature information of the center point of the area occupied by the target object in the first video frame can be determined by determining all the pixel points in the first target area eigenvalues, and from the eigenvalues of all pixels included in the first target area, find the eigenvalue with the smallest difference with the target eigenvalue.

Optionally, the searching pixel points included in the second video frame whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame includes:

reducing the second video frame multiple times according to preset rules to obtain multiple third video frames;

For each third video frame in the plurality of third video frames, in the third video frame, a second target area whose area is a preset area is determined centering on the center point of the third video frame , the preset area is determined according to the aging degree of the spherical IPC;

Searching for a pixel point whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame from all the pixels included in the obtained plurality of second target areas.

Further, in order to avoid finding no pixel in the current second video frame whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame, the second video frame can be processed multiple times. Zoom out to obtain a plurality of third video frames, and search for pixels whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame in the multiple third video frames.

Optionally, searching for pixels whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame from all the pixels included in the obtained plurality of second target areas points, including:

For each second target area in the plurality of second target areas, acquiring feature information of all pixels included in the second target area;

For each pixel point in all the pixels included in the second target area, the pixel difference value included in the feature information of the pixel point is subjected to an average operation and a range operation to obtain a third average value and a third range difference value, and determine the product between the third average value and the third extreme difference value as the feature value of the pixel point;

performing an average operation and a range operation on the pixel difference values included in the feature information of the center point of the area occupied by the target object in the first video frame to obtain a second average value and a second range value, and The product between the second average value and the second range value is determined as a target feature value;

For each second target area in the plurality of second target areas, from the feature values of all pixels included in the second target area, select the feature with the smallest difference from the target feature value value, and determine the pixel corresponding to the selected feature value as the target pixel;

Select the eigenvalue with the smallest difference between the obtained eigenvalues of the plurality of target pixel points and the target eigenvalue, and determine the target pixel corresponding to the selected eigenvalue as the feature information and the target object in the A pixel point that is matched with feature information of a central point of the area occupied by the first video frame.

In the plurality of third video frames, a pixel point whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame can be determined by determining a plurality of third video frames in the plurality of third video frames. Two target areas, and find the feature value with the smallest difference between the target feature value and the target feature value among the feature values of all the pixel points included in the plurality of second target areas.

Optionally, before searching the pixel points included in the second video frame whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame, further includes :

Determine the coordinates corresponding to the center point of the area occupied by the target object in the first video frame in the plane coordinate system of the gun-type IPC;

According to the determined coordinates and the preset coordinate transformation model, determine the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system, the preset coordinates The transformation model is used to transform the coordinates of the points in the plane coordinate system into the spherical coordinate system;

Determine a second rotation angle of the spherical IPC according to the determined longitude and latitude;

sending a second rotation request to the spherical IPC, the second rotation request carrying the second rotation angle;

receiving a second video frame collected by the spherical IPC, where the second video frame is a video frame collected after the spherical IPC rotates according to the second rotation angle when receiving a second rotation request.

In the embodiment of the present invention, when the target object to be monitored is detected in the first video frame, the corresponding coordinates and Preset the coordinate conversion model, determine the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system, and control the spherical IPC to rotate according to the determined longitude and latitude, so as to The center point of the area occupied by the target object in the second video frame is located at the center of the second video frame.

Optionally, according to the determined coordinates and the preset coordinate transformation model, determine the corresponding longitude and latitude of the center point of the area occupied by the target object in the first video frame in the spherical coordinate system ,include:

According to the determined coordinates, determine the distance x ₁ between the center point of the area occupied by the target object in the first video frame and the vertical intersection point, and the area occupied by the target object in the first video frame The distance x ₂ between the center point of the region and the origin of the spherical coordinate system, and the center point of the region occupied by the target object in the first video frame and any one of at least four calibration points The distance x ₃ between the calibration points, the vertical intersection point is the intersection point of a straight line passing through the origin of the spherical coordinate system and perpendicular to the plane coordinate system intersects the plane coordinate system, and the at least four calibration points are Marking points randomly selected from the first video frame;

According to the x ₁ , x ₂ , x ₃ , according to the following preset coordinate transformation model, determine the longitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system and latitude;

in, θ are respectively the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system, and _θ1 are respectively the longitude and latitude of the said perpendicular intersection point in said spherical coordinate system, and _θ3 are respectively the longitude and latitude of any one of the at least four calibration points in the spherical coordinate system.

In the embodiment of the present invention, according to the coordinates corresponding to the center point of the area occupied by the target object in the first video frame in the plane coordinate system and the preset coordinate transformation model, the occupied area of the target object in the first video frame is determined. The longitude and latitude corresponding to the center point of the region in the spherical coordinate system need to determine the undetermined parameters x ₁ , x ₂ , and x ₃ in the preset coordinate transformation model.

Optionally, according to the determined coordinates and the preset coordinate transformation model, determine the corresponding longitude and latitude of the center point of the area occupied by the target object in the first video frame in the spherical coordinate system Previously, also included:

Randomly select at least four calibration points in the first video frame, and any three calibration points in the at least four calibration points are not collinear;

determining the coordinates of each of the at least four calibration points in the planar coordinate system and the longitude and latitude in the spherical coordinate system;

According to the coordinates of each of the at least four calibration points in the plane coordinate system and the longitude and latitude in the spherical coordinate system, determine the position parameter of the vertical intersection point in the spherical coordinate system, The location parameters include the distance between the vertical intersection point and the origin of the spherical coordinate system, and the longitude and latitude of the vertical intersection point in the spherical coordinate system;

The preset coordinate transformation model is established according to the position parameter of the vertical intersection point in the spherical coordinate system and the longitude and latitude of any one of the at least four calibration points in the spherical coordinate system.

In the embodiment of the present invention, the establishment of the preset coordinate transformation model may be based on the coordinates of each of the at least four marking points in the first video frame in the plane coordinate system and the coordinates of each of the at least four marking points in the first video frame. The longitude and latitude of the fixed point in the spherical coordinate system are implemented.

Optionally, the determining the longitude and latitude of each of the at least four calibration points in the spherical coordinate system includes:

For each calibration point in the at least four calibration points, control the center point of the video picture collected by the spherical IPC to rotate to the calibration point from a position where the longitude and latitude of the spherical coordinate system are zero;

Determine the rotation angle of the spherical IPC in the horizontal direction and the rotation angle of the spherical IPC in the vertical direction;

The rotation angle of the spherical IPC in the horizontal direction is determined as the longitude of the calibration point in the spherical coordinate system, and the rotation angle of the spherical IPC in the vertical direction is determined as the calibration point in the Latitude in the spherical coordinate system described above.

In an embodiment of the present invention, determine the longitude and latitude of each of the at least four calibration points in the spherical coordinate system, that is, determine the longitude and latitude of the center point of the video picture collected by the spherical IPC from the spherical coordinate system When the latitude is zero and the position is rotated to the calibration point, the rotation angle of the spherical IPC in the horizontal direction and the rotation angle of the spherical IPC in the vertical direction.

In the second aspect, a video monitoring device based on the gun-ball linkage is provided, and the video monitoring device based on the gun-ball linkage has the function of realizing the behavior of the video monitoring method based on the gun-ball linkage in the first aspect. The video monitoring device based on gun-ball linkage includes at least one module, and the at least one module is used to realize the video monitoring method based on gun-ball linkage provided in the first aspect.

In the third aspect, a video surveillance device based on gunball linkage is provided. The structure of the video surveillance device based on gunball linkage includes a processor and a memory, and the memory is used to store and support video surveillance based on gunball linkage. The device executes the program of the video surveillance method based on the gun-ball linkage provided in the first aspect, and stores the data involved in realizing the video surveillance method based on the gun-ball linkage provided in the first aspect. The processor is configured to execute programs stored in the memory.

In a fourth aspect, a computer storage medium is provided, and instructions are stored in the computer-readable storage medium, and when it is run on a computer, the computer executes the gun-ball-based video surveillance method described in the first aspect.

In the fifth aspect, a computer program product containing instructions is provided, and when it is run on a computer, it causes the computer to execute the gun-ball linkage-based video surveillance method described in the first aspect above.

The technical effects obtained by the above-mentioned second aspect, third aspect, fourth aspect and fifth aspect are similar to those obtained by the corresponding technical means in the first aspect, and will not be repeated here.

The beneficial effect brought by the technical solution provided by the embodiment of the present invention is: to obtain the feature information of the center point of the area occupied by the target object in the first video frame, since the feature information of each pixel point is unique, it can In the second video frame collected by the spherical IPC, search for a pixel point where the feature information matches the acquired feature information, and the pixel point found can represent the center point of the area occupied by the target object in the second video frame; then according to The longitude and latitude corresponding to the pixel points found in the spherical coordinate system of the spherical IPC determine the first rotation angle of the spherical IPC, because the first rotation angle is to rotate the center point of the second video picture to the target object At the center point of the area occupied by the second video picture, the spherical IPC needs to rotate the angle. Therefore, when the spherical IPC receives the first rotation request and rotates according to the first rotation angle, the video picture collected by the spherical IPC The center point will be the center point of the area occupied by the target object in the second video frame, avoiding the situation that the target object may not be located in the center of the video frame collected by the spherical IPC due to the aging problem of the spherical IPC, thereby improving the speed of passing through the spherical IPC. The effect of the type IPC on monitoring the details of the target object.

Description of drawings

Fig. 1 is a schematic diagram of a video monitoring system based on gun-ball linkage provided by an embodiment of the present invention;

Fig. 2 is a schematic structural diagram of an IPC provided by an embodiment of the present invention;

Fig. 3A is a flow chart of a video monitoring method based on gun-ball linkage provided by an embodiment of the present invention;

Fig. 3B is a schematic diagram of pixel point distribution in the neighborhood of a central point provided by an embodiment of the present invention;

Fig. 3C is a schematic diagram of a plane coordinate system and a spherical coordinate system provided by an embodiment of the present invention;

Fig. 3D is a schematic diagram of a geometric model provided by an embodiment of the present invention;

Fig. 3E is a schematic diagram of another geometric model provided by an embodiment of the present invention;

Fig. 4A is a block diagram of a video surveillance device based on gun-ball linkage provided by an embodiment of the present invention;

Fig. 4B is a block diagram of a search module provided by an embodiment of the present invention;

FIG. 4C is a block diagram of another search module provided by an embodiment of the present invention;

Fig. 4D is a block diagram of another video surveillance device based on gun-ball linkage provided by an embodiment of the present invention;

Fig. 4E is a block diagram of a third determination module provided by an embodiment of the present invention;

Fig. 4F is a block diagram of another video monitoring device based on gun-ball linkage provided by an embodiment of the present invention;

Fig. 4G is a block diagram of a fifth determining module provided by an embodiment of the present invention.

Detailed ways

The embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings.

Before explaining and describing the embodiments of the present invention in detail, the application scenarios of the embodiments of the present invention are firstly introduced. For video surveillance based on a single gun-type IPC, that is, video surveillance realized by only one gun-type IPC, when there is a target object to be monitored within the monitoring range, the target object can be detected by the gun-type IPC. In order to determine the details of the target object, it is necessary to zoom in on the video picture collected by the gun-type IPC, and determine the details of the target object in the zoomed-in video picture. However, because the gun-type IPC is used to collect video images in a large area, the resolution of the video images collected by the gun-type IPC is not high, so the effect of determining the details of the target object in the video image after zooming in is not very effective. ideal. Therefore, for the video surveillance of a moving target object in a wide range, in order to monitor the movement of the target object as a whole, and to monitor the details of the target object, a video surveillance method based on gun-ball linkage is usually used, and also That is, the gun-type IPC is used to detect whether there is a target object to be monitored. When the gun-type IPC detects the target object to be monitored, the details of the target object are monitored through the dynamic rotation of the spherical IPC. For example, for the scene of community monitoring, when there is a stranger intruding in the monitoring area, the stranger can be detected by the gun-type IPC. including the stranger. Since the video picture collected by the spherical IPC is zoomed in, the details of the stranger, such as a close-up of the face, can be determined through the video picture collected by the spherical IPC, so as to improve the security of community monitoring. However, the video monitoring method based on the gun-ball linkage provided by the embodiment of the present invention is applied to the scene where the details of the target object are monitored through the dynamic rotation of the ball-type IPC when the target object to be monitored is detected by the gun-type IPC.

FIG. 1 is a video surveillance system 100 based on gunball linkage provided by an embodiment of the present invention. As shown in FIG. 1 , the video surveillance system 100 based on gunball linkage includes a server 101 , a client 102 , and an IPC 103 . The IPC 103 is used to collect video images to be monitored and send the collected video images to the server 101 . When the server 101 receives the video frame collected by the IPC 103 , the server 101 sends the received video frame to the client 102 . When the client 102 receives the video image sent by the server 101 , it displays the received video image, so that the user can monitor the video according to the video image displayed by the client 102 . In addition, when the server 101 receives the video frame collected by the IPC 103 and determines that the video frame collected by the IPC 103 needs to be adjusted, the server 101 can also control the IPC 103 to dynamically rotate and receive the video frame collected by the IPC 103 after the dynamic rotation.

Wherein, the communication between the server 101 and the IPC1 03 may be performed through a wireless network or a wired network, and the communication between the server 101 and the client 102 may also be performed through a wireless network or a wired network. In addition, the video surveillance system based on gun-ball linkage may include multiple IPCs 103 , that is, the video surveillance system based on gun-ball linkage may deploy multiple IPCs. In FIG. 1 , only three IPCs 103 are taken as an example for illustration.

It should be noted that, as shown in FIG. 1 , the IPC 103 includes a gun-type IPC 1031 and a ball-type IPC 1032 to implement the video surveillance method based on gun-ball linkage provided by the embodiment of the present invention. That is to say, the server configures a corresponding spherical IPC for each gun-type IPC. Specifically, when the user determines that the gun-type IPC 1031 and the spherical IPC 1032 in the installed IPC 103 are installed, the client 102 sends a configuration request to the server 101, The configuration request carries the identifier of the gun-type IPC 1031 and the identifier of the ball-type IPC 1032 in the IPC 103 . When the server 101 receives the configuration request, the identifier of the gun-type IPC 1031 in the IPC 103 and the identifier of the ball-type IPC 1032 are stored in the corresponding relationship between the gun-type IPC and the ball-type IPC, so as to realize the gun-type IPC in the IPC 103 1031 configures spherical IPC 1032.

Wherein, the identifier of the gun-type IPC 1031 is used to uniquely identify the gun-type IPC 1031 , and the identifier of the ball-type IPC 1032 is used to uniquely identify the ball-type IPC 1032 . It should be noted that the IPC 103 may include a gun-type IPC and a ball-type IPC, or may include a gun-type IPC and multiple ball-type IPCs, which is not specifically limited in this embodiment of the present invention.

Optionally, the IPC 103 shown in FIG. 1 can also directly communicate with the client 102 through a wireless network or a wired network, that is, the IPC 103 directly sends the captured video images to the client 102 without going through the server 101. The collected video images are sent to the client 102. At this time, the IPC 103 is also used to perform the operation of automatically dynamically rotating and re-collecting the video images when the video images collected by the IPC 103 need to be adjusted. When communicating directly with the client 102 in a wired network, the method provided by the embodiment of the present invention can also be applied to the IPC 103 . In particular, in the embodiment of the present invention, the video surveillance system based on gun-ball linkage shown in FIG. 1 is taken as an example for illustration.

FIG. 2 is a schematic structural diagram of an IPC provided by an embodiment of the present invention. The IPC may be a gun-type IPC 1031 or a ball-type IPC 1032 included in the IPC 103 shown in FIG. 1 . Referring to FIG. 2 , the IPC includes: a video collector 201 , a transmitter 202 and a receiver 203 .

Wherein, the video collector 201 is used to collect video images, and the transmitter 202 can be used to send data and/or signaling. The receiver 203 may be used to receive data and/or signaling and the like.

Optionally, the IPC also includes a memory and a processor. Among them, the memory can be used to store one or more software programs and/or modules. The memory can be read-only memory (Read-only Memory, ROM), random access memory (Random Access Memory, RAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), CD-ROM ( Compact Disc Read-Only Memory, CD-ROM), disk storage medium, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by integrated circuits, but is not limited thereto . The processor can be a general-purpose central processing unit (Central Processing Unit, CPU), a microprocessor, a specific application integrated circuit (Application-Specific Integrated Circuit, ASIC), or one or more integrated circuits used to control the program execution of the application program. circuit. In particular, when the IPC directly communicates with the client through a wireless network or a wired network, the processor included in the IPC can be implemented by running software programs and/or modules stored in the memory, and calling data stored in the memory. The embodiment of the present invention provides a video monitoring method based on gun-ball linkage.

FIG. 3A is a video surveillance method based on gun-ball linkage provided by an embodiment of the present invention, which is applied to the video surveillance system based on gun-ball linkage shown in FIG. 1 above. As shown in Fig. 3A, the video monitoring method based on gun-ball linkage includes the following steps.

Step 301: The gun-type IPC collects a first video frame, and sends the first video frame to the server, and the ball-type IPC collects a second video frame, and sends the second video frame to the server.

As shown in Figure 1, after the server configures the ball IPC for the gun IPC, the gun IPC and the ball IPC start to collect video images and send the collected video images to the server. For the convenience of subsequent description, the video frame collected by the gun-type IPC is called the first video frame, and the video frame collected by the spherical IPC is called the second video frame, that is, the first video frame is the video frame collected by the gun-type IPC , the second video frame is the video frame currently collected by the server on the ball-type IPC configured with the gun-type IPC. When the server receives the first video frame collected by the gun-type IPC and the second video frame collected by the spherical IPC, it sends the first video frame and the second video frame to the client. The client receives the first video frame and the second video frame sent by the server, so that the user can view the first video frame collected by the gun-type IPC and the second video frame collected by the ball-type IPC through the client.

It should be noted that the second video frame collected by the spherical IPC is only a part of the video frame collected by the gun-type IPC, so that the target object may not be included in the current second video frame or the target object may not be in the second video frame. At the center of the picture, in order to improve the effect of video surveillance on the target object, the server needs to first determine the target object to be monitored in the first video picture, and then control the spherical IPC to recapture the video picture through steps 302 to 305. so that the target object is at the center of the second video frame. Wherein, the server determines the target object to be monitored in the first video frame may be: the server detects in real time whether there is a target object to be monitored in the first video frame, and when it is detected that there is a target object to be monitored in the first video frame, A target object in the first video frame is determined. Of course, the server may also determine the target object to be monitored in the first video picture: when the user views the target object to be monitored on the first video picture through the client, the client sends a monitoring request to the server, and the monitoring The request carries the position of the target object in the first video frame, and when the server receives the monitoring request, it determines the target object in the first video frame according to the position of the target object in the first video frame.

Step 302: Obtain the feature information of the center point of the area occupied by the target object in the first video frame, the feature information includes the pixel difference between the pixel value of the center point and the pixel values of multiple neighboring pixel points , the plurality of neighborhood pixels are pixels in the neighborhood of the center point in the first video frame.

When the server detects the target object to be monitored in the first video frame, in order to monitor the details of the target object through the spherical IPC, the server needs to control the spherical IPC to capture the area occupied by the target object in the second video frame The center point of the second video frame is rotated to the center point of the second video frame, that is, the server needs to first search for the center point of the area occupied by the target object in the second video frame in the second video frame. Since the feature information of the center point of the area occupied by the target object in the second video frame is the same as the feature information of the center point of the area occupied by the target object in the first video frame, the server detects in the first video frame When the target object is selected, the feature information of the center point of the area occupied by the target object in the first video frame needs to be obtained first.

Wherein, the feature information includes the pixel difference between the pixel value of the center point of the area occupied by the target object in the first video frame and the pixel values of multiple neighboring pixel points respectively. That is, after determining the pixel value of the center point of the area occupied by the target object in the first video frame, the server obtains the pixel value of each pixel point in the neighborhood of the center point, and for the For each pixel point, determine the difference between the pixel value of the pixel point and the pixel value of the center point to obtain multiple pixel difference values, and the multiple pixel difference values are the feature information of the center point, that is, The feature information of the central point is a set of data. In addition, the server may acquire the pixel value of the central point and the pixel value of each pixel in the neighborhood of the central point through optical flow, which will not be described in detail in this embodiment of the present invention. As shown in Figure 3B, point P is the center point of the area occupied by the target object in the first video frame, P ₁ to P ₈ are the pixels in the neighborhood of point P, G ₁ is the pixel value of point P and P ₁ The difference between the pixel values of point, G ₂ is the difference between the pixel value of point P and the pixel value of point P ₂ , ..., and so on, G ₈ is the pixel value of point P and the pixel value of point P ₈ The difference between pixel values, so the feature information of point P can be expressed as an array (G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ).

In addition, in the embodiment of the present invention, in order to facilitate the server to compare the difference between the feature information of two pixels, for the feature information of a certain pixel, the feature information of the pixel is statistically processed to obtain a The value representing the feature information of the pixel point is referred to as the feature value of the pixel point for convenience of description. In a possible implementation, since the average value can indicate the average size of a set of data, and the range value can indicate the degree of dispersion of a set of data, the average value and range value of the feature information of the pixel can be The product between is determined as the feature value of the pixel. For example, for the feature information of point P above (G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ), the feature value of point P can be expressed as Among them, p _x is the eigenvalue of point P, G _max is the maximum value among the feature information (G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ) of point P, and G _min is the minimum value among the feature information (G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ) of point P, It is to add eight pieces of data in the feature information (G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ) of point P.

It should be noted that when the server determines the target object to be monitored in the first video frame, the video frame captured by the spherical IPC may or may not include the target object. When the video frame collected by the spherical IPC includes the target object, the server directly controls the spherical IPC according to steps 303 to 305 to rotate the target object from the center point of the area occupied by the second video frame to the center point of the second video frame . When the video frame collected by the spherical IPC does not include the target object, the server needs to first control the spherical IPC to collect the video frame that includes the target object, and then control the spherical IPC to place the target object in the second frame through steps 303 to 305. The center point of the area occupied by the video frame is rotated to the center point of the second video frame.

Specifically, the way in which the server controls the spherical IPC to collect the video frame that includes the target object may be as follows: the server determines that the center point of the area occupied by the target object in the first video frame corresponds to Coordinates; according to the determined coordinates and the preset coordinate conversion model, the server determines the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system; according to the determined longitude and latitude, Determine the second rotation angle of the spherical IPC; the server sends a second rotation request to the spherical IPC, and the second rotation request carries a second rotation angle; when the spherical IPC receives the second rotation request, according to the second rotation angle Carry out rotation, and collect the video picture after the rotation, send the video picture that gathers after the rotation to server; The video picture is a video picture collected after the spherical IPC rotates according to the second rotation angle when receiving the second rotation request.

Wherein, the plane coordinate system of the gun-type IPC is the plane coordinate system determined by the server in advance according to the first video picture collected by the gun-type IPC, and the origin of the plane coordinate system is the preset origin, that is, for any first video picture A point has a corresponding coordinate in the plane coordinate system. The spherical coordinate system is the spherical coordinate system determined by the server in advance according to all video images collected by the spherical IPC within the full viewing angle. The spherical coordinate system is preset with a position where both longitude and latitude are 0, that is, for spherical IPC acquisition Any point in the second video frame has a corresponding longitude and latitude in the spherical coordinate system. Since the spherical coordinate system is the spherical coordinate system determined by the server in advance according to all video frames collected by the spherical IPC within the full viewing angle, any point in the first video frame has a corresponding point in the spherical coordinate system, that is, the first Any point in a video frame has a corresponding longitude and latitude in the spherical coordinate system.

Therefore, when the server determines the center point of the area occupied by the target object in the first video frame, it can determine the corresponding longitude and latitude of the center point in the spherical coordinate system, and according to the longitude of the center point of the current second video frame and latitude, determine the difference between the longitude of the center point of the current second video frame and the corresponding longitude of the center point in the spherical coordinate system, obtain the first angle, and determine the first angle as the second rotation angle in The angle in the horizontal direction. At the same time, the server determines the difference between the latitude of the center point of the current second video frame and the latitude corresponding to the center point in the spherical coordinate system to obtain the second angle, and determines the second angle as the second rotation angle in the vertical The angle in the direction, when the server determines the angle of the second rotation angle in the horizontal direction and the angle of the second rotation angle in the vertical direction, that is, determines the second rotation angle.

It is worth noting that after the spherical IPC rotates according to the second rotation angle, the center point of the area occupied by the target object in the second video frame may not be in the center point of the second video frame due to the aging problem of the spherical IPC. In this case, at this time, the server still needs to control the spherical IPC to rotate the center point of the target object in the area occupied by the second video frame to the center point of the second video frame through steps 303 to 305.

In addition, the preset coordinate transformation model is used to transform the coordinates of points in the plane coordinate system into the spherical coordinate system, that is, the server can determine that any point in the first video frame is in the spherical coordinate system according to the preset coordinate transformation model corresponding longitude and latitude. And the preset coordinate transformation model is a coordinate transformation model pre-established by the server. Specifically, the establishment of the preset coordinate transformation model by the server can be realized through the following steps:

(1) The server determines the coordinates of each of the at least four calibration points in the plane coordinate system and the longitude and latitude in the spherical coordinate system.

Wherein, the at least four marking points are at least four marking points randomly selected by the server in the first video frame. It is worth noting that since the at least four calibration points are used to indicate the plane coordinate system, any three calibration points in the at least four calibration points are not collinear, and at the same time in order to improve the accuracy of the preset coordinate transformation model , the relative positions of the at least four marking points in the first video frame are scattered. Optionally, the at least four marking points may also be marking points selected by the user in the first video frame through the client, that is, when the client determines the four marking points selected by the user in the first video frame, Send a calibration request to the server, the calibration request carries the coordinates of the at least four calibration points in the plane coordinate system, and when the server receives the calibration request sent by the client, determine the coordinates of the at least four calibration points in the plane coordinate system .

In particular, the implementation of the server determining the longitude and latitude of the at least four calibration points in the spherical coordinate system may be: for each calibration point in the at least four calibration points, the server controls the video frame to be collected by the spherical IPC The center point rotates from the position where the longitude and latitude of the spherical coordinate system are zero to the calibration point; the server determines the rotation angle of the spherical IPC in the horizontal direction and the rotation angle of the spherical IPC in the vertical direction; then the spherical IPC is in the The rotation angle in the horizontal direction is determined as the longitude of the calibration point in the spherical coordinate system, and the rotation angle of the spherical IPC in the vertical direction is determined as the latitude of the calibration point in the spherical coordinate system.

As shown in Figure 3C, the plane T is the plane represented by the above-mentioned plane coordinate system, the ball O represents the space of the above-mentioned spherical coordinate system, and O' is a straight line passing through the origin O of the spherical coordinate system and perpendicular to the plane coordinate system T that intersects the plane coordinate system A, B, C and D are the four calibration points randomly selected by the server in the plane coordinate system T. Since A, B, C and D are four calibration points randomly selected by the server in the plane coordinate system T, the server can directly determine the coordinates of A, B, C and D in the plane coordinate system as (x _A , y _A ), (x _B , y _B ), (x _C , y _C ) and (x _D , y _D ), and at the same time, the server determines the longitude and latitude of the at least four calibration points in the spherical coordinate system according to the above method, The longitude and latitude of A, B, C and D in the spherical surface coordinate system are respectively and

(2), the server determines the position parameter of the vertical intersection point in the spherical coordinate system according to the coordinates of each of the at least four calibration points in the plane coordinate system and the longitude and latitude in the spherical coordinate system, the position The parameters include the distance between the vertical intersection point and the origin of the spherical coordinate system, and the longitude and latitude of the vertical intersection point in the spherical coordinate system. The intersection point of the coordinate system.

Specifically, as shown in Figure 3C, three calibration points A, B, and C are selected from the four calibration points, and the three calibration points and the vertical intersection O' and the origin of the spherical coordinate system form a graph as shown in Figure 3D geometry model. According to the coordinates of A, B and C in the plane coordinate system, the distance AB between A and B, the distance BC between B and C, and the distance AC between A and C can be determined according to the following formula (1).

In the spherical coordinate system, when the longitude and latitude of A, B and C in the spherical coordinate system are determined as and , the spatial coordinates of the three points A, B and C in the spherical coordinate system can be determined according to the following formula (2).

Among them, θ, are the longitude and latitude of a point in the spherical coordinate system, (x, y, z) are the spatial coordinates of the point in the spherical coordinate system, and r is the distance from the point to the origin. In particular, when r is 1, (x, y, z) is the unit vector of the vector formed by the point and the origin.

Determine the vector according to the above formula (2) the unit vector of for vector the unit vector of for vector the unit vector of for and the vector and vector The angle between ∠AOB is also the vector and vector Angle between, vector and vector The angle between ∠BOC is also the vector and vector Angle between, vector and vector The angle between ∠AOC is also the vector and vector angle between.

Meanwhile, the server may respectively determine the lengths of A'B', B'C' and A'C' according to the following formula (3).

From the trigonometric cosine theorem, we know that Since the lengths of OA' and OB' are 1, the value of A'B' determined according to formula (3) is substituted to determine ∠AOB. In the same way, ∠BOC and ∠AOC can be determined.

Suppose AB=m, AC=n, BC=l, ∠AOB=α, ∠BOC=β, ∠AOC=γ, OA=x, OB=y, OC=z, then the following formula can be determined according to the trigonometric cosine theorem (4).

Since m, n, l, cosα, cosβ, and cosγ in the above formula (4) are determined data, the lengths of OA, OB, and OC can be determined through the above formula (4). It should be noted that in the process of solving the lengths of OA, OB and OC according to the above formula (4), a set of solutions may be obtained. At this time, another calibration point D is used to determine the lengths of OA, OB and OD according to the above method. A set of solutions of length, taking the common solution in these two sets of solutions, can determine the unique solution of OA and OB. Similarly, a unique solution to OC can also be determined.

So far, for the three-sided cone OABC in the spherical coordinate system, since the longitude and latitude of A, B and C are known, and the lengths of OA, OB and OC have also been determined, A and B can be determined according to formula (2) The spatial coordinates of the three points and C in the spherical coordinate system, that is, the spatial position of the three-pointed pyramid OABC in the spherical coordinate system is determined, so the spatial coordinates of the vertical intersection point O' in the three-pointed pyramid OABC can be determined according to the space vector geometry, That is to determine the length of OO' and the longitude and latitude of O' in the spherical coordinate system, that is to determine the position parameters of the vertical intersection point.

(3) Establishing a preset coordinate conversion model according to the position parameter of the vertical intersection point in the spherical coordinate system and the longitude and latitude of any one of the at least four calibration points in the spherical coordinate system.

If any one of the at least four calibration points is point A shown in Figure 3C or Figure 3D, for any point X in the plane coordinate system, the any point X, the vertical intersection O' and any one of the calibration points A A geometric model as shown in Figure 3E can be constructed. As shown in Figure 3E, the coordinates of point X in the plane coordinate system are known coordinates, so the lengths of O'X and XA can be directly determined, because the triangle OO'X is For a right triangle, on the premise that OO' and O'X are known, the length of OX can be directly determined. At this time, all side lengths of triangle OO'X and triangle OAX have been determined. Therefore, according to the triangle cosine theorem, the following Equation (5) determines ∠AOX and ∠O'OX.

Assume that the longitude and latitude of X and O' in the spherical coordinate system are and Therefore, it is possible to determine the vector the unit vector of for vector the unit vector of for vector the unit vector of for and the vector and vector The angle between ∠AOX is also the vector and vector Angle between, vector and vector The angle between ∠O'OX is also the vector and vector angle between.

At the same time, the server can determine the length of A'X' and the length of O"X' according to the following formula (6).

The above formula (6) can be abbreviated as

And because, the length of O "X' and the length of A'X' can be obtained according to the trigonometric cosine theorem, as following formula (7):

Substituting formula (5) into formula (7), the transformed formula (7) can be obtained as follows:

In the transformed formula (7), since X is any point in the plane coordinate system, that is, OX, O'X, and AX are undetermined parameters, the transformed formula (7) can be abbreviated as: O"X ' ² =f ₁ (XO',OX), A'X' ² =f ₁ (OX,AX).

From the abbreviated formula (6) and the abbreviated formula (7), the following formula (8) can be obtained:

In formula (8), for any point X in the plane coordinate system, when the coordinates of X in the plane coordinate system are determined, the three undetermined parameters OX, O'X, AX in formula (8) can be obtained, because Longitude and latitude of the perpendicular intersection point O' in formula (8) And the longitude and latitude of any one of the four calibration points A is a determined parameter, so the longitude and latitude of any point X in the spherical coordinate system can be determined according to formula (8)

Therefore, in order to determine the corresponding longitude and latitude in the spherical coordinate system of the center point of the area occupied by the target object in the first video frame, the server can establish the following preset coordinate conversion model according to formula (8)

Wherein, x1 is the distance between the center point of the area occupied by the target object in the _first video frame and the vertical intersection point, and _x2 is the center point of the area occupied by the target object in the first video frame and the spherical coordinate system The distance between the origins, x ₃ is the distance between the center point of the area occupied by the target object in the first video frame and any one of the at least four marking points, θ are respectively the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system, and _θ1 are the longitude and latitude of the vertical intersection point in the spherical coordinate system, and θ ₃ are respectively the longitude and latitude of any one of the at least four calibration points in the spherical coordinate system.

It can be known from the above-mentioned preset coordinate conversion model that according to the determined coordinates and the preset coordinate conversion model, the implementation method of determining the corresponding longitude and latitude of the center point of the area occupied by the target object in the first video frame in the spherical coordinate system It may be: according to the determined coordinates, determine the distance x ₁ between the center point of the area occupied by the target object in the first video frame and the vertical intersection point, and the center point of the area occupied by the target object in the first video frame The distance x ₂ between the origin of the spherical coordinate system and the distance x ₃ between the center point of the area occupied by the target object in the first video frame and any one of the at least four marked points, the vertical intersection point is the intersection of a straight line passing through the origin of the spherical coordinate system and perpendicular to the plane coordinate system and intersecting the plane coordinate system, at least four calibration points are randomly selected calibration points from the first video frame; according to x ₁ , x ₂ , x ₃ , according to the above preset coordinate conversion model, determine the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system.

Step 303: The server searches the pixels included in the second video frame for pixels whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame.

Specifically, step 303 may be implemented through the following two possible implementation manners:

In the first possible implementation, in the second video frame, the center point of the second video frame is used as the center to determine the first target area with a preset area, and the preset area is determined according to the aging degree of the spherical IPC , from all the pixels included in the first target area, search for the pixel whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame.

In order to improve the efficiency of searching for pixels whose feature information matches the feature information of the center point of the area occupied by the target object in the first video picture in the second video picture, in the second video picture, only the first video picture can be A pixel point whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame is searched for among all the pixels included in the target area. Wherein, the preset area is the preset area determined by the server according to the aging degree of the spherical IPC, that is, the corresponding relationship between the aging degree of the spherical IPC and the preset area is stored in the server, when the server determines that the spherical IPC The predetermined area of the first target area can be determined according to the corresponding relationship between the aging degree of the spherical IPC and the predetermined area. In addition, the aging degree is determined by the server according to the delivery time of the spherical IPC, that is, for each spherical IPC, the server stores the delivery time of the spherical IPC, and according to the delivery time of the spherical IPC, it can be determined. The factory use time of the spherical IPC; the server can determine the aging degree of the spherical IPC according to the correspondence between the factory use time and the aging degree. It should be noted that, in this embodiment of the present invention, the shape of the first target area may be a rectangle or a circle, which is not specifically limited in this embodiment of the present invention.

In particular, from all the pixels included in the first target area, the implementation method of finding the pixel whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame may be: obtain the first The feature information of all pixels included in the target area; for each pixel in all the pixels included in the first target area, the pixel difference included in the feature information of the pixel is averaged and ranged to obtain the first An average value and the first extreme difference value, and the product between the first average value and the first extreme difference value is determined as the feature value of the pixel point; To the center point of the area occupied by the target object in the first video picture The pixel difference value included in the feature information is averaged and ranged to obtain the second average value and the second range value, and the product between the second average value and the second range value is determined as the target feature value; From the eigenvalues of all pixels included in the first target area, select the eigenvalue with the smallest difference with the target eigenvalue, and determine the pixel corresponding to the selected eigenvalue as the feature information and the target object in the first A pixel point matched with feature information of the center point of the area occupied by the video frame.

For example, the first target area includes 50 pixels, and the feature information of the 50 pixels acquired by the server through optical flow is P1(G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ), P2 (G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ),…, P50 (G ₁ , G ₂ , G ₃ , G ₄ , G ₅ , G ₆ , G ₇ , G ₈ ), for the feature information of each pixel in these 50 pixels, according to the formula Determine the feature value of the pixel, and obtain the feature values p ₁ , p ₂ , ..., p ₅₀ of 50 pixel points; for the feature information of the center point of the area occupied by the target object in the first video frame, also according to the formula Determine the feature value p of the center point of the area occupied by the target object in the first video frame, and determine p as the target feature value; select from the feature values p ₁ , p ₂ , ..., p ₅₀ of 50 pixel points The feature value closest to the target feature value p, that is, the feature value with the smallest difference. If the selected feature value is p ₃₀ , the pixel corresponding to the feature value p ₃₀ is the area occupied by the feature information and the target object in the first video frame The feature information of the center point matches the pixel point.

Optionally, the center point of the area occupied by the target object in the second video frame may not be located in the first target area, that is, at this time, the feature information searched in the first target area is the same as that of the target object in the first video frame. The pixel point that matches the feature information of the center point of the area occupied by the target object may not be the center point of the area occupied by the target object in the second video frame. Therefore, the server also needs to send the current second video The picture is reduced, and the pixel points whose feature information matches the feature information of the center point of the area occupied by the target object in the first video picture are searched again in the reduced video picture. Specifically, the server may shrink the second video frame multiple times according to preset rules to obtain multiple third video frames; for each third video frame in the multiple third video frames, in the third video frame, Taking the center point of the third video frame as the center to determine the second target area with the area as the preset area, thereby obtaining a plurality of second target areas; searching for feature information and the target from all pixels included in the plurality of second target areas A pixel point that matches the feature information of the center point of the area occupied by the object in the first video frame.

Wherein, the implementation process of the server searching for the pixel point whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame from all the pixels included in the plurality of second target areas may be: for For each second target area in the plurality of second target areas, the feature information of all the pixels included in the second target area is obtained; for each pixel in all the pixels included in the second target area, for the pixel The pixel difference value included in the feature information is averaged and ranged to obtain the third average value and the third range value, and the product between the third average value and the third range value is determined as the pixel value Feature value; the pixel difference value included in the feature information of the center point of the area occupied by the target object in the first video frame is averaged and ranged to obtain the second average value and the second range value, and the second The product between the two mean values and the second extreme difference value is determined as the target feature value; for each second target area in a plurality of second target areas, from the feature values of all pixels included in the second target area, Select the eigenvalue with the smallest difference with the target eigenvalue, and determine the pixel point corresponding to the selected eigenvalue as the target pixel point, so as to obtain multiple target pixel points; select from the eigenvalues of multiple target pixel points The feature value with the minimum difference between the target feature value, and the target pixel point corresponding to the selected feature value is determined as the feature information matching the feature information of the center point of the area occupied by the target object in the first video frame pixel.

It should be noted that the area of the second target area is also a preset area, which is determined according to the aging degree of the spherical IPC. Wherein, the server determines the preset area of the second target area according to the aging degree of the spherical IPC in the same manner as the server determines the preset area of the first target area according to the aging degree of the spherical IPC, and will not be described in detail here. In addition, the shape of the second target area may be a rectangle or a circle, which is not specifically limited in this embodiment of the present invention. In particular, the area and shape of the second target area may be the same as those of the first target area.

In addition, the implementation process of reducing the second video frame multiple times according to preset rules to obtain multiple third video frames may be as follows: reducing the current second video frame according to preset rules to obtain the first third video frame ;Continue to shrink the first third video frame according to the preset rules to obtain the second third video frame; continue to shrink the second third video frame according to the preset rules to obtain the third third video frame; Video frames..., and so on, to obtain a plurality of third video frames. That is, each time the second video frame is reduced according to the preset rule, a third video frame can be obtained. Wherein, the preset rule may be to reduce one pixel point in the pixel row and column of the current video frame each time, that is, reduce one pixel point in each row of pixels and each column pixel point of the current video frame each time.

In addition, it should be noted that when the server reduces the second video frame multiple times according to preset rules, the number of times the server reduces the video frame may be a preset number of times, and the preset number of times may be 5, 10, or 15. In particular, the server may reduce the second video frame multiple times until the scaling ratio of the reduced video frame is consistent with the scaling ratio of the first video frame.

For example, the server reduces the rows and columns of pixels in the current second video frame by one pixel respectively to obtain the first third video frame; the server continues to reduce the rows and columns of pixels in the first third video frame by one pixel respectively to obtain the first third video frame Two third video frames; ...; and so on, until the reduced video frame is consistent with the scaling ratio of the first video frame, at this moment, 20 third video frames are obtained; in the 20 third video frames, according to the above-mentioned method Determine 20 second target areas; for each second target area in the 20 second target areas, among all the pixels included in the second target area, the server determines the target pixels, and the features of the target pixels The difference between the value and the target feature value is the smallest; when the server has performed the above operations on the 20 second target areas, 20 target pixels are obtained; then from the 20 target pixels, select the feature value and Target the target pixel with the smallest feature value difference, and determine the feature information of the selected target pixel as the pixel that matches the feature information of the center point of the area occupied by the target object in the first video frame.

In a second possible implementation manner, all pixels included in the second video frame are searched for a pixel whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame.

Of course, the server may not determine the first target area in the second video frame, and directly search for the relationship between the feature information and the center point of the area occupied by the target object in the first video frame in all the pixels included in the second video frame. Pixels matching feature information, wherein the server searches all pixels included in the second video frame for a pixel whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame The method is basically the same as the realization method that the server searches all the pixels included in the first target area for the pixel whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame, and is not described here. Do elaborate.

Step 304: The server determines the first rotation angle of the spherical IPC according to the longitude and latitude corresponding to the searched pixel points in the spherical coordinate system of the spherical IPC, and the first rotation angle is to rotate the center point of the second video picture The angle at which the spherical IPC needs to rotate to the center point of the area occupied by the target object in the second video frame.

It should be noted that since the searched pixel point is the pixel point in the video frame collected by the spherical IPC, and the searched pixel point is the center point of the feature information and the area occupied by the target object in the first video frame The pixel points matched by the feature information of the target object, therefore, the found pixel points can represent the center point of the area occupied by the target object in the second video frame, so the first rotation angle is also to rotate the center point of the second video frame The angle at which the spherical IPC needs to rotate when reaching the found pixel.

Therefore, the server determines that the first rotation angle of the spherical IPC can be: the server determines the corresponding longitude and latitude of the pixel point found in the spherical coordinate system; according to the longitude and latitude of the center point of the video picture collected by the current spherical IPC , determine the difference between the longitude of the center point of the video frame collected by the current spherical IPC and the corresponding longitude of the found pixel point in the spherical coordinate system, obtain the third angle, and determine the third angle as the first The rotation angle is the angle in the horizontal direction; at the same time, the server determines the difference between the latitude of the center point of the video picture collected by the current spherical IPC and the corresponding latitude of the found pixel point in the spherical coordinate system to obtain the fourth angle, And the fourth angle is determined as the angle of the first rotation angle in the vertical direction. When the server determines the angle of the first rotation angle in the horizontal direction and the angle of the first rotation angle in the vertical direction, it also determines the first rotation angle. A rotation angle.

Step 305: the server sends a first rotation request to the spherical IPC, the first rotation request carries a first rotation angle, and the first rotation request is used to instruct the spherical IPC to rotate according to the first rotation angle.

When the spherical IPC receives the first rotation request sent by the server, it rotates according to the first rotation angle carried in the first rotation request, and recaptures the video picture after the rotation, and at the same time sends to the server The collected video picture; when the server receives the video picture collected after the spherical IPC rotates according to the first rotation angle, it sends the received video picture to the client, so that the user can view the spherical IPC through the client. A video frame collected after the first rotation angle is rotated. It should be noted that since the first rotation angle is the angle that the spherical IPC needs to rotate when the center point of the second video frame is rotated to the center point of the area occupied by the target object in the second video frame, the spherical IPC In the video frame collected after being rotated according to the first rotation angle, the center point of the area occupied by the target object in the video frame is the center point of the video frame.

In the embodiment of the present invention, when the server determines the target object to be monitored in the first video frame collected by the gun-type IPC, the feature information of the center point of the area occupied by the target object in the first video frame is acquired, because The feature information of each pixel point is unique, therefore, the pixel point whose feature information matches the acquired feature information can be found in the second video frame collected by the spherical IPC, and the found pixel point can represent the target object at the second The center point of the area occupied by the second video frame. Then according to the longitude and latitude corresponding to the pixel point found in the spherical IPC in the spherical coordinate system of the spherical IPC, determine the first rotation angle of the spherical IPC, because the first rotation angle is to rotate the center point of the second video picture to When the center point of the area occupied by the target object in the second video frame, the spherical IPC needs to rotate the angle, therefore, when the spherical IPC receives the first rotation request and rotates according to the first rotation angle, the spherical IPC collected The center point of the video picture will be the center point of the area occupied by the target object in the second video picture, avoiding the situation that the target object may not be located in the center of the video picture collected by the spherical IPC due to the aging problem of the spherical IPC, thereby improving The effect of monitoring the details of the target object through the spherical IPC. In addition, the server also pre-establishes a preset coordinate transformation model, which is used to indicate that the point in the plane coordinate system of the video picture collected by the gun-type IPC is in the spherical coordinate system of the video picture collected by the spherical IPC. Position, that is, for any point in the video screen collected by the gun-type IPC, the server can determine the longitude and latitude of any point in the spherical coordinate system through the preset coordinate transformation model, and according to the longitude and latitude of any point, control The spherical IPC rotates to enable video monitoring of any point.

Fig. 4A is a video monitoring device 400 based on gun-ball linkage provided by an embodiment of the present invention. The video monitoring device based on gun-ball linkage can be implemented as part or all of a server by software, hardware or a combination of the two. It should be noted that when the IPC shown in Figure 1 communicates directly with the client through a wireless network or a wired network, the video monitoring device based on gun-ball linkage can also be realized by software, hardware or a combination of both. Part or all of the IPC, the IPC may be the IPC shown in FIG. 2 . Referring to FIG. 4A , the device includes an acquiring module 401 , a searching module 402 , a first determining module 403 and a first sending module 404 .

An acquisition module 401, configured to execute step 302 in the embodiment of FIG. 3A;

The search module 402 is configured to execute step 303 in the embodiment of FIG. 3A , wherein the second video frame is a video frame currently collected by a ball-type IPC configured for a gun-type IPC;

The first determining module 403 is configured to execute step 304 in the embodiment of FIG. 3A;

The first sending module 404 is configured to execute step 305 in the embodiment of FIG. 3A .

Optionally, referring to FIG. 4B, the search module 402 includes a first determination unit 4021 and a first search unit 4022:

The first determining unit 4021 is used to determine, in the second video frame, a first target area with a preset area centered on the center point of the second video frame, and the preset area is determined according to the aging degree of the spherical IPC of;

The first searching unit 4022 is configured to search, from all the pixels included in the first target area, the pixel points whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame.

Optionally, the first search unit 4021 is specifically configured to:

Obtain feature information of all pixels included in the first target area;

For each pixel point in all the pixels included in the first target area, the pixel difference value included in the feature information of the pixel point is averaged and the range operation is performed to obtain the first average value and the first range value, and Determine the product between the first average value and the first range value as the feature value of the pixel point;

Perform an average operation and a range operation on the pixel difference value included in the feature information of the center point of the area occupied by the target object in the first video frame to obtain a second average value and a second range value, and divide the second average value The product between the value and the second range value is determined as the target feature value;

From the eigenvalues of all pixels included in the first target area, select the eigenvalue with the smallest difference with the target eigenvalue, and determine the pixel corresponding to the selected eigenvalue as the feature information and the target object in the first A pixel point matched with feature information of the center point of the area occupied by the video frame.

Optionally, referring to FIG. 4C, the search module 402 includes a reduction unit 4023, a second determination unit 4024, and a second search unit 4025:

A reduction unit 4023, configured to reduce the second video frame multiple times according to preset rules to obtain multiple third video frames;

The second determination unit 4024 is configured to, for each third video frame in the plurality of third video frames, in the third video frame, determine an area centered on the center point of the third video frame as the preset area The second target area, the preset area is determined according to the aging degree of the spherical IPC;

The second search unit 4025 is configured to search for a pixel point whose feature information matches the feature information of the center point of the area occupied by the target object in the first video frame from all the pixels included in the obtained plurality of second target areas.

Optionally, the second search unit 4025 is specifically configured to:

For each second target area in the plurality of second target areas, acquiring feature information of all pixels included in the second target area;

For each pixel in all the pixels included in the second target area, an average operation and a range operation are performed on the pixel difference included in the feature information of the pixel to obtain a third average value and a third range value, and determining the product between the third average value and the third extreme difference as the feature value of the pixel point;

Perform an average operation and a range operation on the pixel difference value included in the feature information of the center point of the area occupied by the target object in the first video frame to obtain a second average value and a second range value, and divide the second average value The product between the value and the second range value is determined as the target feature value;

For each of the plurality of second target areas, from the feature values of all pixels included in the second target area, select the feature value with the smallest difference between the target feature value and the selected The pixel point corresponding to the feature value is determined as the target pixel point;

Select the eigenvalue with the minimum difference between the target eigenvalues from the eigenvalues of the multiple target pixels obtained, and determine the target pixel corresponding to the selected eigenvalues as the feature information and the target object in the first video frame The pixel points that match the feature information of the center point of the area occupied by .

Optionally, referring to FIG. 4D, the apparatus 400 further includes a second determination module, a second determination module 405, a third determination module 406, a fourth determination module 407, a second sending module 408, and a receiving module 409:

The second determination module 405 is used to determine the coordinates corresponding to the center point of the area occupied by the target object in the first video frame in the plane coordinate system of the gun-type IPC;

The third determining module 406 is configured to determine the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system according to the determined coordinates and the preset coordinate transformation model, the preset The coordinate transformation model is used to convert the coordinates of points in the plane coordinate system to the spherical coordinate system;

The fourth determination module 407 is used to determine the second rotation angle of the spherical IPC according to the determined longitude and latitude;

The second sending module 408 is configured to send a second rotation request to the spherical IPC, where the second rotation request carries a second rotation angle;

The receiving module 409 is configured to receive a second video frame collected by the spherical IPC, where the second video frame is a video frame collected after the spherical IPC rotates according to the second rotation angle when receiving the second rotation request.

Optionally, referring to FIG. 4E, the third determining module 406 includes a third determining unit 4061 and a fourth determining unit 4062:

The third determining unit 4061 is configured to determine the distance x ₁ between the center point of the area occupied by the target object in the first video frame and the vertical intersection point, and the area occupied by the target object in the first video frame according to the determined coordinates The distance x ₂ between the center point of the area and the origin of the spherical coordinate system, and the distance between the center point of the area occupied by the target object in the first video frame and any one of the at least four marking points x ₃ , the vertical intersection point is the intersection point where a straight line passing through the origin of the spherical coordinate system and perpendicular to the plane coordinate system intersects the plane coordinate system, and at least four calibration points are randomly selected calibration points from the first video frame;

The fourth determining unit 4062 is configured to determine the longitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system according to x ₁ , x ₂ , and x ₃ according to the following preset coordinate transformation model and latitude;

in, θ are respectively the longitude and latitude corresponding to the center point of the area occupied by the target object in the first video frame in the spherical coordinate system, and _θ1 are the longitude and latitude of the vertical intersection point in the spherical coordinate system, respectively, and θ ₃ are respectively the longitude and latitude of any one of the at least four calibration points in the spherical coordinate system.

Optionally, referring to FIG. 4F, the device 400 further includes:

The selection module 410 is used to randomly select at least four calibration points in the first video frame, and any three calibration points in the at least four calibration points are not collinear;

The fifth determination module 411 is used to determine the coordinates of each of the at least four calibration points in the plane coordinate system and the longitude and latitude in the spherical coordinate system;

The sixth determination module 412 is used to determine the position parameter of the vertical intersection point in the spherical coordinate system according to the coordinates of each of the at least four calibration points in the plane coordinate system and the longitude and latitude in the spherical coordinate system, The position parameter includes the distance between the vertical intersection point and the origin of the spherical coordinate system, and the longitude and latitude of the vertical intersection point in the spherical coordinate system;

The establishing module 413 is configured to establish the preset coordinate conversion model according to the position parameter of the vertical intersection point in the spherical coordinate system and the longitude and latitude of any one of the at least four calibration points in the spherical coordinate system.

Optionally, referring to FIG. 4G, the fifth determination module 411 includes a control unit 4111, a fifth determination unit 4112, and a sixth determination unit 4113:

The control unit 4111 is used to control the center point of the video picture captured by the spherical IPC to rotate from a position where the longitude and latitude of the spherical coordinate system are zero to the calibration point for each of the at least four calibration points;

The fifth determining unit 4112 is used to determine the rotation angle of the spherical IPC in the horizontal direction and the rotation angle of the spherical IPC in the vertical direction;

The sixth determination unit 4113 is used to determine the rotation angle of the spherical IPC in the horizontal direction as the longitude of the calibration point in the spherical coordinate system, and determine the rotation angle of the spherical IPC in the vertical direction as the longitude of the calibration point on the spherical surface. The latitude in the coordinate system.

In the embodiment of the present invention, when the server determines the target object to be monitored in the first video frame collected by the gun-type IPC, the feature information of the center point of the area occupied by the target object in the first video frame is acquired, because The feature information of each pixel point is unique, therefore, the pixel point whose feature information matches the acquired feature information can be found in the second video frame collected by the spherical IPC, and the found pixel point can represent the target object at the second The center point of the area occupied by the second video frame. Then according to the longitude and latitude corresponding to the pixel point found in the spherical IPC in the spherical coordinate system of the spherical IPC, determine the first rotation angle of the spherical IPC, because the first rotation angle is to rotate the center point of the second video picture to When the center point of the area occupied by the target object in the second video frame, the spherical IPC needs to rotate the angle, therefore, when the spherical IPC receives the first rotation request and rotates according to the first rotation angle, the spherical IPC collected The center point of the video picture will be the center point of the area occupied by the target object in the second video picture, avoiding the situation that the target object may not be located in the center of the video picture collected by the spherical IPC due to the aging problem of the spherical IPC, thereby improving The effect of monitoring the details of the target object through the spherical IPC. In addition, the server also pre-establishes a preset coordinate transformation model, which is used to indicate that the point in the plane coordinate system of the video picture collected by the gun-type IPC is in the spherical coordinate system of the video picture collected by the spherical IPC. Position, that is, for any point in the video screen collected by the gun-type IPC, the server can determine the longitude and latitude of any point in the spherical coordinate system through the preset coordinate transformation model, and according to the longitude and latitude of any point, control The spherical IPC rotates to enable video monitoring of any point.

It should be noted that when the video monitoring device based on gun-ball linkage provided by the above-mentioned embodiment performs video monitoring, it only uses the division of the above-mentioned functional modules as an example. In practical applications, the above-mentioned functions can be assigned by different The functional modules are completed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the video monitoring device based on the gun-ball linkage provided in the above embodiment and the video monitoring method based on the gun-ball linkage provided in the embodiment of FIG. 3A belong to the same concept. The specific implementation process is shown in the embodiment of FIG.

In the above embodiments, all or part may be implemented by software, hardware, firmware or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on the computer, the procedures or functions according to the embodiments of the present invention will be generated in whole or in part. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website, computer, server or data center Transmission to another website site, computer, server or data center via wired (eg coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (eg infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center integrated with one or more available media. The available medium may be a magnetic medium (for example: floppy disk, hard disk, magnetic tape), an optical medium (for example: Digital Versatile Disc (Digital Versatile Disc, DVD)), or a semiconductor medium (for example: Solid State Disk (Solid State Disk, SSD) )Wait.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above embodiments can be completed by hardware, and can also be completed by instructing related hardware through a program. The program can be stored in a computer-readable storage medium. The above-mentioned The storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk, and the like.

The above-mentioned embodiments provided by the application are not intended to limit the application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the application shall be included in the protection scope of the application. Inside.

Claims (18)

1. A video monitoring method based on gun and ball linkage is characterized by comprising the following steps:

acquiring feature information of a central point of an area occupied by a target object in a first video picture, wherein the first video picture is a video picture acquired by a gun-shaped network camera IPC, the feature information comprises pixel difference values between a pixel value of the central point and pixel values of a plurality of neighborhood pixels, and the neighborhood pixels are pixels in the neighborhood of the central point in the first video picture;

searching pixel points with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture in pixel points included in a second video picture, wherein the second video picture is a video picture currently acquired by a spherical IPC configured for the gun-shaped IPC;

determining a first rotation angle of the spherical IPC according to the longitude and latitude corresponding to the searched pixel point in the spherical coordinate system of the spherical IPC, wherein the first rotation angle is an angle which the spherical IPC needs to rotate when the central point of the second video picture is rotated to the central point of the area occupied by the target object in the second video picture;

and sending a first rotation request to the spherical IPC, wherein the first rotation request carries the first rotation angle, and the first rotation request is used for indicating the spherical IPC to rotate according to the first rotation angle.

2. The method according to claim 1, wherein the searching for a pixel point whose feature information matches with the feature information of the center point of the region occupied by the target object in the first video frame among the pixel points included in the second video frame comprises:

in the second video picture, a first target area with the area as a preset area is determined by taking the central point of the second video picture as a center, and the preset area is determined according to the aging degree of the spherical IPC;

and searching pixel points with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture from all the pixel points included in the first target area.

3. The method according to claim 2, wherein the searching for a pixel point whose feature information matches with the feature information of the center point of the region occupied by the target object in the first video frame from all the pixel points included in the first target region comprises:

acquiring characteristic information of all pixel points included in the first target area;

for each pixel point in all pixel points included in the first target area, performing average operation and range operation on pixel difference values included in the feature information of the pixel points to obtain a first average value and a first range value, and determining the product of the first average value and the first range value as the feature value of the pixel point;

carrying out average operation and range operation on pixel difference values included in the feature information of the central point of the area occupied by the target object in the first video picture to obtain a second average value and a second range value, and determining the product of the second average value and the second range value as a target feature value;

and selecting a characteristic value with the minimum difference value with the target characteristic value from the characteristic values of all pixel points included in the first target area, and determining the pixel point corresponding to the selected characteristic value as a pixel point with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture.

4. The method according to claim 1, wherein the searching for a pixel point whose feature information matches with the feature information of the center point of the region occupied by the target object in the first video frame among the pixel points included in the second video frame comprises:

reducing the second video image for multiple times according to a preset rule to obtain multiple third video images;

for each third video picture in the plurality of third video pictures, determining a second target area with the area as a preset area in the third video picture by taking the center point of the third video picture as the center, wherein the preset area is determined according to the aging degree of the spherical IPC;

and searching pixel points with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture from all the pixel points in the obtained second target areas.

5. The method according to claim 4, wherein the searching for a pixel point whose feature information matches with the feature information of the center point of the region occupied by the target object in the first video frame from all the pixel points included in the obtained plurality of second target regions comprises:

for each second target area in the plurality of second target areas, acquiring feature information of all pixel points included in the second target area;

for each pixel point in all pixel points included in the second target region, performing average operation and range operation on pixel difference values included in the feature information of the pixel points to obtain a third average value and a third range value, and determining a product between the third average value and the third range value as a feature value of the pixel point;

carrying out average operation and range operation on pixel difference values included in the feature information of the central point of the area occupied by the target object in the first video picture to obtain a second average value and a second range value, and determining the product of the second average value and the second range value as a target feature value;

for each second target area in the plurality of second target areas, selecting a characteristic value with the minimum difference value with the target characteristic value from the characteristic values of all pixel points included in the second target area, and determining the pixel point corresponding to the selected characteristic value as a target pixel point;

and selecting a characteristic value with the minimum difference value with the target characteristic value from the obtained characteristic values of the plurality of target pixel points, and determining the target pixel point corresponding to the selected characteristic value as a pixel point with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture.

6. The method according to claim 1, wherein before searching for a pixel point whose feature information matches with the feature information of the center point of the region occupied by the target object in the first video frame among the pixel points included in the second video frame, the method further comprises:

determining the corresponding coordinate of the central point of the area occupied by the target object in the first video picture in the plane coordinate system of the gun type IPC;

determining longitude and latitude corresponding to the central point of the area occupied by the target object in the first video picture in the spherical coordinate system according to the determined coordinates and a preset coordinate conversion model, wherein the preset coordinate conversion model is used for converting the coordinates of the point in the plane coordinate system into the spherical coordinate system;

determining a second rotation angle of the spherical IPC according to the determined longitude and latitude;

sending a second rotation request to the spherical IPC, wherein the second rotation request carries the second rotation angle;

and receiving a second video picture acquired by the spherical IPC, wherein the second video picture is the video picture acquired after the spherical IPC rotates according to the second rotation angle when a second rotation request is received.

7. The method of claim 6, wherein determining the longitude and latitude of the center point of the area occupied by the target object in the first video frame in the spherical coordinate system according to the determined coordinates and a preset coordinate transformation model comprises:

according to the determined coordinates, determining the distance x between the central point of the area occupied by the target object in the first video picture and the vertical intersection point₁And the distance x between the central point of the area occupied by the target object in the first video picture and the origin of the spherical coordinate system₂And the distance x between the central point of the area occupied by the target object in the first video picture and any one of at least four calibration points₃The vertical intersection point passes through the spherical surfaceThe coordinate system has an origin and a straight line perpendicular to the plane coordinate system intersects at the intersection point of the plane coordinate system, and the at least four calibration points are randomly selected from the first video picture;

according to the x₁、x₂、x₃Determining the longitude and the latitude corresponding to the central point of the area occupied by the target object in the first video picture in the spherical coordinate system according to a preset coordinate conversion model;

wherein,respectively corresponding longitude and latitude in the spherical coordinate system of the central point of the area occupied by the target object in the first video picture,and theta₁Respectively the longitude and latitude of the vertical intersection point in the spherical coordinate system,and theta₃Respectively the longitude and the latitude of any one of the at least four calibration points in the spherical coordinate system.

8. The method according to claim 6 or 7, wherein the determining, according to the determined coordinates and a preset coordinate transformation model, before the longitude and the latitude of the center point of the area occupied by the target object in the first video frame in the spherical coordinate system, further comprises:

randomly selecting at least four index points in the first video picture, wherein any three index points in the at least four index points are not collinear;

determining coordinates of each of the at least four calibration points in the planar coordinate system and a longitude and latitude in the spherical coordinate system;

determining a position parameter of a vertical intersection point in the spherical coordinate system according to the coordinates of each of the at least four calibration points in the planar coordinate system and the longitude and latitude in the spherical coordinate system, wherein the position parameter comprises a distance between the vertical intersection point and an origin of the spherical coordinate system and the longitude and latitude of the vertical intersection point in the spherical coordinate system;

and establishing the preset coordinate conversion model according to the position parameter of the vertical intersection point in the spherical coordinate system and the longitude and latitude of any one of the at least four calibration points in the spherical coordinate system.

9. The method of claim 8, wherein said determining the longitude and latitude in the spherical coordinate system of each of the at least four calibration points comprises:

for each of the at least four calibration points, controlling a central point of a video picture acquired by the spherical IPC to rotate to the calibration point from a position where the longitude and the latitude of the spherical coordinate system are both zero;

determining a rotation angle of the spherical IPC in the horizontal direction and a rotation angle of the spherical IPC in the vertical direction;

and determining the rotation angle of the spherical IPC in the horizontal direction as the longitude of the calibration point in the spherical coordinate system, and determining the rotation angle of the spherical IPC in the vertical direction as the latitude of the calibration point in the spherical coordinate system.

10. A video monitoring device based on rifle ball linkage, its characterized in that, the device includes:

the device comprises an acquisition module, a processing module and a display module, wherein the acquisition module is used for acquiring characteristic information of a central point of an area occupied by a target object in a first video picture, the first video picture is a video picture acquired by a gun-shaped network camera IPC, the characteristic information comprises pixel difference values between a pixel value of the central point and pixel values of a plurality of neighborhood pixels, and the neighborhood pixels are pixels in the neighborhood of the central point in the first video picture;

the searching module is used for searching pixel points with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture in pixel points included in a second video picture, wherein the second video picture is a video picture currently acquired by the spherical IPC configured for the gun-shaped IPC;

the first determining module is used for determining a first rotation angle of the spherical IPC according to the longitude and the latitude of the searched pixel point corresponding to the spherical coordinate system of the spherical IPC, wherein the first rotation angle is an angle which is required to rotate the spherical IPC when the central point of the second video picture is rotated to the central point of the area occupied by the target object in the second video picture;

the first sending module is used for sending a first rotation request to the spherical IPC, the first rotation request carries the first rotation angle, and the first rotation request is used for indicating the spherical IPC to rotate according to the first rotation angle.

11. The apparatus of claim 10, wherein the lookup module comprises:

a first determining unit, configured to determine, in the second video picture, a first target area having an area as a preset area with a center point of the second video picture as a center, where the preset area is determined according to an aging degree of the spherical IPC;

and the first searching unit is used for searching pixel points with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture from all the pixel points included in the first target area.

12. The apparatus of claim 11, wherein the first lookup unit is specifically configured to:

acquiring characteristic information of all pixel points included in the first target area;

for each pixel point in all pixel points included in the first target area, performing average operation and range operation on pixel difference values included in the feature information of the pixel points to obtain a first average value and a first range value, and determining the product of the first average value and the first range value as the feature value of the pixel point;

carrying out average operation and range operation on pixel difference values included in the feature information of the central point of the area occupied by the target object in the first video picture to obtain a second average value and a second range value, and determining the product of the second average value and the second range value as a target feature value;

and selecting a characteristic value with the minimum difference value with the target characteristic value from the characteristic values of all pixel points included in the first target area, and determining the pixel point corresponding to the selected characteristic value as a pixel point with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture.

13. The apparatus of claim 10, wherein the lookup module comprises:

the reducing unit is used for reducing the second video picture for multiple times according to a preset rule to obtain multiple third video pictures;

a second determining unit, configured to determine, for each of the plurality of third video pictures, a second target area having an area as a preset area in the third video picture with a center point of the third video picture as a center, where the preset area is determined according to an aging degree of the spherical IPC;

and the second searching unit is used for searching pixel points with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture from all the pixel points included in the obtained second target areas.

14. The apparatus of claim 13, wherein the second lookup unit is specifically configured to:

for each second target area in the plurality of second target areas, acquiring feature information of all pixel points included in the second target area;

for each pixel point in all pixel points included in the second target region, performing average operation and range operation on pixel difference values included in the feature information of the pixel points to obtain a third average value and a third range value, and determining a product between the third average value and the third range value as a feature value of the pixel point;

carrying out average operation and range operation on pixel difference values included in the feature information of the central point of the area occupied by the target object in the first video picture to obtain a second average value and a second range value, and determining the product of the second average value and the second range value as a target feature value;

for each second target area in the plurality of second target areas, selecting a characteristic value with the minimum difference value with the target characteristic value from the characteristic values of all pixel points included in the second target area, and determining the pixel point corresponding to the selected characteristic value as a target pixel point;

and selecting a characteristic value with the minimum difference value with the target characteristic value from the obtained characteristic values of the plurality of target pixel points, and determining the target pixel point corresponding to the selected characteristic value as a pixel point with characteristic information matched with the characteristic information of the central point of the area occupied by the target object in the first video picture.

15. The apparatus of claim 10, wherein the apparatus further comprises:

the second determination module is used for determining the corresponding coordinate of the central point of the area occupied by the target object in the first video picture in the plane coordinate system of the gun type IPC;

a third determining module, configured to determine, according to the determined coordinate and a preset coordinate conversion model, a longitude and a latitude, in the spherical coordinate system, of a center point of an area occupied by the target object in the first video frame, where the preset coordinate conversion model is used to convert coordinates of a point in the planar coordinate system into the spherical coordinate system;

the fourth determining module is used for determining a second rotation angle of the spherical IPC according to the determined longitude and latitude;

the second sending module is used for sending a second rotation request to the spherical IPC, and the second rotation request carries the second rotation angle;

and the receiving module is used for receiving a second video picture acquired by the spherical IPC, and the second video picture is a video picture acquired after the spherical IPC rotates according to the second rotation angle when receiving a second rotation request.

16. The apparatus of claim 15, wherein the third determining module comprises:

a third determining unit configured to determine, according to the determined coordinates, a distance x between a center point of an area occupied by the target object in the first video picture and a vertical intersection point₁And the distance x between the central point of the area occupied by the target object in the first video picture and the origin of the spherical coordinate system₂And the distance x between the central point of the area occupied by the target object in the first video picture and any one of at least four calibration points₃The vertical intersection point is an intersection point of a straight line which passes through the origin of the spherical coordinate system and is vertical to the plane coordinate system and intersects the plane coordinate system, and the at least four calibration points are points along the first video pictureMachine-selected index points;

a fourth determination unit for determining x₁、x₂、x₃Determining the longitude and the latitude corresponding to the central point of the area occupied by the target object in the first video picture in the spherical coordinate system according to a preset coordinate conversion model;

wherein,respectively corresponding longitude and latitude in the spherical coordinate system of the central point of the area occupied by the target object in the first video picture,and theta₁Respectively the longitude and latitude of the vertical intersection point in the spherical coordinate system,and theta₃Respectively the longitude and the latitude of any one of the at least four calibration points in the spherical coordinate system.

17. The apparatus of claim 15 or 16, wherein the apparatus further comprises:

a selection module, configured to randomly select at least four calibration points in the first video picture, where any three calibration points of the at least four calibration points are not collinear;

a fifth determining module for determining coordinates of each of the at least four calibration points in the planar coordinate system and longitude and latitude in the spherical coordinate system;

a sixth determining module, configured to determine, according to coordinates of each of the at least four calibration points in the planar coordinate system and longitude and latitude in the spherical coordinate system, location parameters of a vertical intersection in the spherical coordinate system, where the location parameters include a distance between the vertical intersection and an origin of the spherical coordinate system, and longitude and latitude of the vertical intersection in the spherical coordinate system;

and the establishing module is used for establishing the preset coordinate conversion model according to the position parameter of the vertical intersection point in the spherical coordinate system and the longitude and the latitude of any one of the at least four calibration points in the spherical coordinate system.

18. The apparatus of claim 17, wherein the fifth determining module comprises:

the control unit is used for controlling the central point of a video picture acquired by the spherical IPC to rotate to the calibration point from a position where the longitude and the latitude of the spherical coordinate system are both zero aiming at each of the at least four calibration points;

a fifth determining unit, configured to determine a rotation angle of the spherical IPC in the horizontal direction and a rotation angle of the spherical IPC in the vertical direction;

a sixth determining unit, configured to determine a rotation angle of the spherical IPC in the horizontal direction as a longitude of the calibration point in the spherical coordinate system, and determine a rotation angle of the spherical IPC in the vertical direction as a latitude of the calibration point in the spherical coordinate system.