CN-122002135-A - Camera linkage method and system based on joint calibration and track prediction

CN122002135ACN 122002135 ACN122002135 ACN 122002135ACN-122002135-A

Abstract

The invention discloses a camera linkage method and a system based on joint calibration and track prediction, wherein the method comprises the steps of carrying out event identification on a video stream acquired by a first camera; the method comprises the steps of extracting position coordinates of a target event in an image coordinate system of a first camera, converting the position coordinates into PTZ parameters for controlling a second camera according to a joint calibration relation, aligning a shooting center of the second camera to the target event, calculating a motion track and a motion speed of the target event according to position coordinate changes of the target event in a video stream, predicting future position coordinates of the target event, inputting the future position coordinates into the joint calibration relation, generating updated PTZ parameters, controlling the second camera, and tracking and shooting the target event. The invention solves the problems of response lag and limited resolution caused by the existing manual specified coordinate switching visual angle and lack of dynamic focal length adjustment, and realizes automatic linkage and fine detail capture of a moving target under wide area monitoring.

Inventors

WU KEWEI
GOU XIAOGANG
HE XIAOGANG
JIN HAO

Assignees

成都卓视智通科技有限公司
北京卓视智通科技有限责任公司

Dates

Publication Date: 20260508
Application Date: 20251226

Claims (10)

1. A camera linkage method based on joint calibration and track prediction is characterized by comprising the following steps: Acquiring a video stream acquired by a first camera, and carrying out event identification on the video stream; when a target event is identified, extracting the position coordinates of the target event in the image coordinate system of the first camera; converting the position coordinates into PTZ parameters for controlling a second camera according to a pre-established joint calibration relation, wherein the joint calibration relation comprises a mapping relation between an image coordinate system of the first camera and a physical space and a PTZ parameter set of the second camera corresponding to different physical space positions; Controlling the cradle head steering and lens zooming of the second camera by utilizing the PTZ parameters, and aligning the shooting center of the second camera to the target event; Calculating a motion trail and a motion speed of the target event according to the position coordinate change of the target event in the video stream, and predicting future position coordinates of the target event according to the motion trail and the motion speed; And inputting the future position coordinates into the joint calibration relation, generating updated PTZ parameters, controlling the second camera according to the updated PTZ parameters, and tracking and shooting the target event.
2. The camera linkage method based on joint calibration and trajectory prediction according to claim 1, wherein the step of acquiring a video stream acquired by a first camera and performing event recognition on the video stream comprises: receiving a video stream continuously collected and transmitted by the first camera; Decoding the video stream to obtain continuous image frames; Analyzing the continuous image frames frame by frame through a preset event identification algorithm; When a target object meeting a preset condition is identified in the continuous image frames by using the event identification algorithm, the target event is judged to be identified.
3. The camera linkage method based on joint calibration and trajectory prediction according to claim 2, wherein the step of extracting the position coordinates of the target event in the image coordinate system of the first camera comprises: Acquiring a current image frame corresponding to the identified target event; Determining an image area occupied by the target object in the current image frame; And calculating the position coordinates of the target event in the image coordinate system of the first camera according to the boundary information of the image area.
4. A camera linkage method based on joint calibration and trajectory prediction according to claim 3, wherein the step of converting the position coordinates into PTZ parameters for controlling the second camera according to a pre-established joint calibration relationship comprises: Calibrating a mapping relation between an image coordinate system of the first camera and the physical space, wherein the mapping relation maps coordinate points in the image coordinate system to specific positions in the physical space; Calibrating a plurality of groups of PTZ parameters of the second camera in advance aiming at a plurality of preset physical position points in the physical space to form a PTZ parameter set, wherein each group of PTZ parameters in the PTZ parameter set is used for indicating the second camera to aim the shooting center at the corresponding preset physical position point; Determining a corresponding target physical position of the target event in the physical space according to the position coordinates and the mapping relation; And determining the PTZ parameters for controlling the second camera according to the target physical position and the PTZ parameter set.
5. The camera linkage method based on joint calibration and trajectory prediction according to claim 1, wherein the step of controlling pan-tilt steering and lens zooming of the second camera by using the PTZ parameter to align a shooting center of the second camera with the target event comprises: Analyzing corresponding horizontal rotation angle, vertical rotation angle and focal length value from the PTZ parameters; And driving the cradle head of the second camera to execute corresponding rotation according to the horizontal rotation angle and the vertical rotation angle, and adjusting the focal length of the lens of the second camera according to the focal length value so as to enable the shooting center of the second camera to be aligned with the physical position of the target event.
6. The camera linkage method based on joint calibration and trajectory prediction according to any one of claims 1 to 5, wherein the step of calculating a motion trajectory and a motion velocity of the target event from a change in position coordinates of the target event in the video stream, and predicting future position coordinates of the target event from the motion trajectory and the motion velocity comprises: acquiring position coordinates of the target event in a plurality of continuous image frames of the video stream to form a position coordinate sequence; Calculating the motion speed of the target event according to the difference of adjacent position coordinates in the position coordinate sequence and the corresponding image frame time interval; Generating a motion trail of the target event in the video stream from the identified target event according to the position coordinate sequence; and calculating future position coordinates of the target event at a subsequent moment by combining the motion trail and the motion speed.
7. The camera linkage method based on joint calibration and trajectory prediction according to claim 6, wherein the step of inputting the future position coordinates into the joint calibration relation, generating updated PTZ parameters, controlling the second camera according to the updated PTZ parameters, and tracking and photographing the target event comprises: inputting the future position coordinates into the joint calibration relation, and determining a future physical position corresponding to the target event at a future moment; determining the updated PTZ parameters for controlling the second camera based on the future physical location and the set of PTZ parameters; driving a cradle head and a lens of the second camera to correspondingly adjust according to the updated PTZ parameters; And repeatedly executing the steps of predicting the future position coordinates, generating the updated PTZ parameters and driving and adjusting the second camera in the continuous movement process of the target event, and completing tracking shooting of the target event.
8. A camera linked system based on joint calibration and track prediction is characterized by comprising: The identification module is used for acquiring the video stream acquired by the first camera and carrying out event identification on the video stream; The extraction module is used for extracting the position coordinates of the target event in the image coordinate system of the first camera when the target event is identified; The conversion module is used for converting the position coordinates into PTZ parameters for controlling the second camera according to the pre-established joint calibration relation; the joint calibration relation comprises a mapping relation between an image coordinate system of the first camera and a physical space, and PTZ parameter sets of the second camera corresponding to different physical space positions; the control module is used for controlling the rotation of the cradle head and the zooming of the lens of the second camera by utilizing the PTZ parameters and aligning the shooting center of the second camera with the target event; the prediction module is used for calculating the motion trail and the motion speed of the target event according to the position coordinate change of the target event in the video stream, and predicting the future position coordinate of the target event according to the motion trail and the motion speed; and the tracking module is used for inputting the future position coordinates into the joint calibration relation, generating updated PTZ parameters, controlling the second camera according to the updated PTZ parameters, and tracking and shooting the target event.
9. An electronic device comprising a processor coupled to a memory, the memory having stored therein at least one computer program loaded and executed by the processor to cause the electronic device to implement the joint calibration and trajectory prediction based camera linkage method of any one of claims 1 to 7.
10. A computer readable storage medium, wherein at least one computer program is stored in the computer readable storage medium, and when the at least one computer program is executed by a processor, the camera linkage method based on joint calibration and trajectory prediction as claimed in any one of claims 1 to 7 is implemented.

Description

Camera linkage method and system based on joint calibration and track prediction Technical Field The invention relates to the technical field of intelligent monitoring and target tracking, in particular to a camera linkage method and system based on joint calibration and track prediction. Background In the technical field of intelligent monitoring and target tracking, multi-camera collaborative linkage has become an important means for realizing wide area coverage and detail capture. However, the prior art scheme still has significant limitations that firstly, part of the system lacks dynamic tracking capability of focal length, real-time self-adaptive adjustment of zoom parameters cannot be realized, so that image recording resolution of a moving target is limited, refined detail features are difficult to continuously acquire, particularly, information loss easily occurs when the target scale is changed greatly, secondly, the scheme relies on manually appointed space coordinate points to switch camera view angles, an automatic trigger mechanism of an event recognition algorithm is not established, linkage response lag is caused, and real-time processing requirements on sudden events in complex scenes cannot be met. The problems restrict the development of the monitoring system in the intelligent and automatic directions, and a novel linkage method capable of integrating space calibration and track prediction is needed to realize global perception of a target, intelligent focal length adjustment and real-time cooperative control of event driving. Accordingly, there is a need to provide a solution to the above-mentioned problems. Disclosure of Invention In order to solve the technical problems, the invention provides a camera linkage method and a camera linkage system based on joint calibration and track prediction. In a first aspect, the invention provides a camera linkage method based on joint calibration and track prediction, which has the following technical scheme: Acquiring a video stream acquired by a first camera, and carrying out event identification on the video stream; when a target event is identified, extracting the position coordinates of the target event in the image coordinate system of the first camera; converting the position coordinates into PTZ parameters for controlling a second camera according to a pre-established joint calibration relation, wherein the joint calibration relation comprises a mapping relation between an image coordinate system of the first camera and a physical space and a PTZ parameter set of the second camera corresponding to different physical space positions; Controlling the cradle head steering and lens zooming of the second camera by utilizing the PTZ parameters, and aligning the shooting center of the second camera to the target event; Calculating a motion trail and a motion speed of the target event according to the position coordinate change of the target event in the video stream, and predicting future position coordinates of the target event according to the motion trail and the motion speed; And inputting the future position coordinates into the joint calibration relation, generating updated PTZ parameters, controlling the second camera according to the updated PTZ parameters, and tracking and shooting the target event. The camera linkage method based on the joint calibration and track prediction has the following beneficial effects: According to the method, after the first camera recognizes a target event, the image coordinates are automatically converted into PTZ parameters of the second camera by utilizing the joint calibration relation to control the target alignment, and the PTZ parameters are updated in real time by combining track prediction to realize zoom tracking, so that the problems of response lag and limited resolution caused by the fact that the existing manual specified coordinates switch the visual angle and lack of dynamic focal length adjustment are solved, and the automatic linkage and fine detail capture of the moving target under wide area monitoring are realized. Based on the scheme, the camera linkage method based on the joint calibration and track prediction can be improved as follows. In an optional manner, the step of acquiring the video stream acquired by the first camera and performing event recognition on the video stream includes: receiving a video stream continuously collected and transmitted by the first camera; Decoding the video stream to obtain continuous image frames; Analyzing the continuous image frames frame by frame through a preset event identification algorithm; When a target object meeting a preset condition is identified in the continuous image frames by using the event identification algorithm, the target event is judged to be identified. The method has the beneficial effects that the method further utilizes the preset algorithm to analyze the target object frame by continuously receiving the video stream and dec