CN111709978A - Cross-screen target tracking method, system, device and storage medium - Google Patents
Cross-screen target tracking method, system, device and storage medium Download PDFInfo
- Publication number
- CN111709978A CN111709978A CN202010372081.3A CN202010372081A CN111709978A CN 111709978 A CN111709978 A CN 111709978A CN 202010372081 A CN202010372081 A CN 202010372081A CN 111709978 A CN111709978 A CN 111709978A
- Authority
- CN
- China
- Prior art keywords
- target object
- video stream
- image
- current frame
- frame image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/36—Applying a local operator, i.e. means to operate on image points situated in the vicinity of a given point; Non-linear local filtering operations, e.g. median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nonlinear Science (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a cross-screen target tracking method, a system, a device and a storage medium, wherein the method comprises the following steps: acquiring a first video stream; determining a target object and the position thereof in a first current frame image of a first video stream, and displaying a locking identifier at the position of the target object in the first current frame image; determining the position of a target object in a next frame image of a first video stream, and displaying a locking mark at the position of the target object in the next frame image; determining a second video stream according to the region of the target object determined for the last time in the first video stream; the position of the target object is determined in the second video stream and the locking identification of the target object is displayed. According to the method and the device, the second video stream is determined, and the locking identification of the target object is displayed in the second video stream, so that cross-screen target automatic tracking is realized, and the target object is prevented from being lost. The invention can be widely applied to the technical field of target tracking.
Description
Technical Field
The invention relates to the technical field of target tracking, in particular to a cross-screen target tracking method, a cross-screen target tracking system, a cross-screen target tracking device and a storage medium.
Background
The target tracking technology is used as a bottom support technology in computer vision, and has wide application value in the fields of intelligent traffic monitoring, human-computer interaction, robot vision navigation, virtual reality, medical diagnosis and the like. Especially, when the method is used for tracking criminal suspects, the target automatic tracking has higher efficiency, and a large amount of manpower and material resources can be saved.
The moving target tracking is to calculate the position of the target on each frame of image by analyzing the image sequence shot by the video acquisition device. However, in the conventional moving target tracking method, a plurality of fixed video acquisition devices are usually adopted to perform video shooting and track a target independently, and when the target leaves the visual field range of the current video acquisition device, it is difficult to determine another video acquisition device capable of continuously tracking the target in time, so that the target cannot be automatically tracked across screens, and the tracked target is lost. And due to the uncontrollable movement of the target and the changeability of the target characteristics, if the target cannot be continuously tracked, the target is difficult to be tracked again after being lost.
Disclosure of Invention
To solve the above technical problems, the present invention aims to: a cross-screen target tracking method, system, device and storage medium are provided.
The technical scheme adopted by the invention on one hand is as follows:
a cross-screen target tracking method comprises the following steps:
acquiring a first video stream;
determining a target object and the position thereof in a first current frame image of a first video stream, and displaying a locking identifier at the position of the target object in the first current frame image;
determining the position of a target object in a next frame image of a first video stream, and displaying a locking mark at the position of the target object in the next frame image;
determining a second video stream according to the region of the target object determined for the last time in the first video stream;
the position of the target object is determined in the second video stream and the locking identification of the target object is displayed.
Further, the method also comprises the following steps:
and acquiring characteristic information of the target object.
Further, the step of determining the target object and its position in the first current frame image of the first video stream, and displaying the locking mark in the position of the target object in the first current frame image includes:
acquiring a first background image of a first video stream;
acquiring a first current frame image of a first video stream;
acquiring a first current frame foreground image according to the first background image and the first current frame image;
determining a target object and the position thereof in the foreground image of the first current frame according to the characteristic information, and extracting the contour of the target object;
and displaying a locking mark at the position of the target object in the first current frame image according to the outline of the target object.
Further, the step of determining a position of the target object in a next frame image of the first video stream and displaying a lock mark at the position of the target object in the next frame image includes:
determining the moving direction and the moving speed of a target object according to the video frames of the first video stream;
predicting the position of a target object in a next frame image of the first video stream according to the position, the moving direction and the moving speed of the target object in the first current frame image;
displaying a locking mark in the next frame image according to the predicted position of the target object;
or the like, or, alternatively,
acquiring a next frame image of the first video stream;
acquiring a next frame of foreground image according to the first background image and the next frame of image;
determining the position of a target object in the next frame of foreground image according to the characteristic information;
and displaying a locking mark at the position of the target object in the next frame of image.
Further, the step of determining the second video stream according to the last determined region to the target object in the first video stream includes:
acquiring an alternative background image of an alternative video stream;
comparing the alternative background image with the region of the target object determined for the last time in the first video stream, and selecting the alternative background image with the highest feature similarity;
selecting a corresponding alternative video stream as a second video stream according to the alternative background image with the highest feature similarity;
wherein there are several of the alternative video streams.
Further, the step of determining the position of the target object in the second video stream and displaying the lock identification of the target object comprises:
predicting the position of the target object in the second current frame image according to the position, the moving direction and the moving speed of the target object which are determined for the last time in the first video stream;
displaying a locking identifier at the predicted position of the target object in the second current frame image;
wherein the second current frame image is a current frame image of a second video stream;
or the like, or, alternatively,
acquiring a second background image of a second video stream;
acquiring a second current frame image of a second video stream;
acquiring a second current frame foreground image according to a second background image and the second current frame image;
determining the position of a target object in a foreground image of a second current frame according to the characteristic information;
displaying a locking identifier at the position of the target object in the second current frame image;
the technical scheme adopted by the other aspect of the invention is as follows:
a cross-screen target tracking system, comprising:
the video stream acquisition module is used for acquiring a first video stream;
the target position determining and displaying module is used for determining a target object and the position thereof in a first current frame image of the first video stream and displaying a locking identifier at the position of the target object in the first current frame image;
the first tracking module is used for determining the position of a target object in a next frame image of the first video stream and displaying a locking identifier at the position of the target object in the next frame image;
the second video stream determining module is used for determining a second video stream according to the area of the target object determined in the first video stream for the last time;
and the second tracking module is used for determining the position of the target object in the second video stream and displaying the locking identification of the target object.
Further, still include:
and the characteristic acquisition module is used for acquiring the characteristic information of the target object.
The technical scheme adopted by the other aspect of the invention is as follows:
a cross-screen target tracking apparatus, comprising:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement the cross-screen target tracking method.
The technical scheme adopted by the other aspect of the invention is as follows:
a storage medium having stored therein processor-executable instructions, which when executed by a processor, are for performing the cross-screen target tracking method.
The invention has the beneficial effects that: according to the cross-screen target tracking method, the system, the device and the storage medium, the target object is continuously locked in the same video stream by determining the position of the target object in the next frame image of the first video stream and displaying the locking mark at the position of the target object in the next frame image, so that the automatic tracking of the target object is realized; and determining a second video stream according to the area of the target object determined in the first video stream for the last time, determining the position of the target object in the second video stream, and displaying the locking identifier of the target object, so that when the target object disappears from the first video stream, the second video stream is accurately found to continuously lock the target object, cross-screen target automatic tracking is realized, and the target object is prevented from being lost.
Drawings
FIG. 1 is a flowchart of a cross-screen target tracking method according to an embodiment of the present invention;
FIG. 2 is a flowchart of a cross-screen target tracking method according to another embodiment of the present invention;
FIG. 3 is a block diagram of a cross-screen target tracking system according to an embodiment of the present invention;
FIG. 4 is a block diagram of a cross-screen target tracking system according to another embodiment of the present invention;
fig. 5 is a block diagram of a cross-screen target tracking apparatus according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the specific embodiments. The step numbers in the following embodiments are provided only for convenience of illustration, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
In the description of the present invention, the meaning of a plurality is more than two, if there are first and second described for the purpose of distinguishing technical features, but not for indicating or implying relative importance or implicitly indicating the number of indicated technical features or implicitly indicating the precedence of the indicated technical features. In addition, the terms "upper", "middle" and "lower" in the following embodiments are used only for clearly describing the positional relationship, and are relative positional relationship, not absolute positional relationship, and can be adaptively adjusted according to the understanding of those skilled in the art.
Referring to fig. 1, an embodiment of the present invention provides a cross-screen target tracking method, including the following steps:
s101, acquiring a first video stream;
specifically, the first video stream is acquired by a video capture device, which includes but is not limited to a camera, a video camera and a closed circuit television, and the camera is taken as an example for description in the description of the embodiment of the present invention. The target object is found at the first time by acquiring the first video stream shot by the first camera so as to be identified, and automatic tracking is carried out subsequently.
S102, determining a target object and the position thereof in a first current frame image of a first video stream, and displaying a locking identifier at the position of the target object in the first current frame image;
specifically, a target object may be determined in the first current frame image through target recognition, and a locking identifier may be displayed at a position of the image where the target object is located to indicate that the target object is the target object, where the locking identifier includes, but is not limited to, a rectangular frame, a circular frame, and an oval frame.
S103, determining the position of a target object in a next frame image of the first video stream, and displaying a locking identifier at the position of the target object in the next frame image;
specifically, after the position (i.e., the current position) of the target object is acquired and locked in the first current frame image, the position (i.e., the next frame position) of the target object in the next frame image of the first video stream can be determined through target recognition, target prediction and the like, and the locking identifier is displayed, so that the target object is continuously locked in the same video stream, and automatic tracking of the target object is realized.
S104, determining a second video stream according to the area of the target object determined for the last time in the first video stream;
specifically, since the area monitored by a single camera is limited, when the target object disappears in the monitoring range of the first camera, the target object needs to be continuously tracked by another camera. In the embodiment of the invention, the second video stream is determined according to the area where the target object tracked last time in the first video stream is located, and the second camera for continuously tracking the target object can be quickly found.
And S105, determining the position of the target object in the second video stream, and displaying the locking identification of the target object.
Specifically, in the second video stream, the position of the target object may be determined in the second video stream in a manner similar to that of the first video stream (i.e., by means of target recognition, target prediction, etc.), and the locking identifier may be continuously displayed, so that the target object is continuously locked in the second video stream, and cross-screen target automatic tracking is achieved.
In the embodiment of the invention, when the target object disappears from the monitoring range of the second camera, the third video stream for cross-screen target tracking can be determined according to the area where the target object tracked last time in the second video stream is located, the target object is continuously and automatically tracked in the third video stream, and the like, so that the multi-point monitoring cross-screen target automatic tracking is realized, and the target object is prevented from losing.
Referring to fig. 2, further as an alternative embodiment, the method further includes the following steps:
and S100, acquiring characteristic information of the target object.
Specifically, before target tracking, characteristic information of a target object can be acquired for identification; because the characteristics of the target object may change, the characteristic information of the target object also needs to be updated in real time in the target tracking process.
As a further optional implementation manner, the step S102 of determining the target object and its position in the first current frame image of the first video stream, and displaying the locking identifier at the position of the target object in the first current frame image includes:
s1021, acquiring a first background image of the first video stream;
specifically, the first background image is an image composed of a plurality of regions in the first video stream, where no change occurs or the change is smaller than a preset value. Optionally, the video frames of the first video stream may be compared based on the preset value to obtain the first background image.
S1022, acquiring a first current frame image of the first video stream;
specifically, a video frame image of the first video stream at the current time is obtained, which is the first current frame image.
Optionally, after the first current frame image is obtained, the first current frame image may be preprocessed by using processing methods such as gray processing, binarization, median filtering, and the like, so that the target object can be accurately identified in the following process.
S1023, acquiring a foreground image of a first current frame according to the first background image and the first current frame image;
specifically, the first current frame image may be filtered according to the first background image, so as to obtain a first current frame foreground image, where the first current frame foreground image includes all objects, such as people or objects, in a motion state in the first current frame image.
S1024, determining a target object and the position of the target object in the foreground image of the first current frame according to the characteristic information, and extracting the contour of the target object;
specifically, the feature information of the target object may be compared with the feature information of all objects in the foreground image of the first current frame, so as to determine the target object to be tracked and the position of the target object, and simultaneously, noise except the target object in the foreground image of the first current frame may be removed, so as to extract the contour of the target object.
And S1025, displaying a locking mark at the position of the target object in the first current frame image according to the contour of the target object.
Specifically, the locking identifier of the target object is determined according to the contour of the target object, so that the locking identifier is prevented from being too small or too large, and the position of the target object can be accurately locked.
In the embodiment of the invention, a first video stream shot by a first camera is obtained, then a first background image, a first current frame image and a background image are obtained, a first current frame foreground image is obtained through comparison, and then the characteristic information of all objects in the first current frame foreground image is obtained and compared with the characteristic information of a target object, so that the target object and the position thereof in the first current frame image are determined, and meanwhile, the noise except the target object in the first current frame foreground image is removed, so that the external contour of the target object is conveniently extracted, and the target object is accurately and automatically tracked.
In practical application, when the invention is used for tracking a criminal suspect, since there are many other unrelated persons or objects besides the criminal suspect to be tracked in the same video frame, it is necessary to eliminate the interference of other unrelated factors. According to the embodiment of the invention, the foreground image of the current frame is extracted, the target object is identified through comparison, and other irrelevant noises are removed, so that the interference of irrelevant factors is avoided, and the accuracy of target tracking is further ensured.
In addition, since the monitoring environment is constantly changing, the background image of the first video stream also needs to be updated in real time to ensure accurate identification of the target object.
As a further alternative, the step S103 of determining the position of the target object in the next frame image of the first video stream and displaying the locking identifier at the position of the target object in the next frame image may be implemented by adopting two schemes, which are specifically described as follows.
As for the first embodiment of step S103, it includes:
s1031, determining the moving direction and the moving speed of the target object according to the video frames of the first video stream;
specifically, the moving direction and the moving speed of the target object may be calculated according to the position change and the interval time of the target object in two or more adjacent video frames in the first video stream.
S1032, predicting the position of the target object in the next frame image of the first video stream according to the position, the moving direction and the moving speed of the target object in the first current frame image;
specifically, the position of the target object in the next frame image may be calculated according to the position of the target object in the first current frame image, and the moving direction and moving speed obtained in the above step S1031, in combination with the time interval between the first current frame image and the next frame image.
And S1033, displaying a locking identifier in the next frame image according to the predicted position of the target object.
Specifically, the embodiment of the present invention may determine a locking identifier of the target object according to the contour of the target object, and display the locking identifier at the predicted position, where the locking identifier includes, but is not limited to, a rectangular frame, a circular frame, and an oval frame.
As can be seen from the above, the first embodiment of step S103 predicts the position of the target object in the next frame image of the first video stream according to the position, the moving direction and the moving speed of the target object in the first current frame image, and can more quickly and efficiently locate the position of the target object in the next frame image of the first video stream and display the lock identifier.
For the second embodiment of step S103, it includes:
s1034, acquiring a next frame image of the first video stream;
specifically, a video frame image at a moment when the first video stream flows is obtained, which is the next frame image.
Optionally, after the next frame image is obtained, the next frame may be preprocessed by using processing methods such as gray processing, binarization, median filtering, and the like, so that the target object can be accurately identified in the following process.
S1035, acquiring a next frame foreground image according to the first background image and the next frame image;
specifically, the next frame of image may be filtered according to the first background image, so as to obtain a next frame of foreground image, where the next frame of foreground image includes all objects, such as people or objects, in a motion state in the next frame of image.
S1036, determining the position of a target object in the next frame of foreground image according to the characteristic information;
specifically, the feature information of the target object may be compared with the feature information of all objects in the foreground image of the next frame, so as to determine the position of the target object.
And S1037, displaying a locking mark at the position of the target object in the next frame of image.
Specifically, the embodiment of the present invention may determine a locking identifier of the target object according to the contour of the target object, and display the locking identifier at the determined position, where the locking identifier includes, but is not limited to, a rectangular frame, a circular frame, and an oval frame.
As can be seen from the above, the second embodiment of step S103 determines the position of the target object in the foreground image of the next frame according to the feature information, and can accurately locate the position of the target object in the foreground image of the next frame of the first video stream and display the locking identifier.
In practical applications, two embodiments of step S103 may be combined: and for each frame of image in the first video stream, predicting the position of the target object in the next frame of image according to the position, the moving direction and the moving speed of the target object in each frame of image, and determining the position of the target object in the video frame images selected by the system at intervals of a plurality of frames according to the characteristic information of the target object so as to adjust the predicted position (namely, the displayed locking identifier is also adjusted synchronously). For example, in the video frames of 15 consecutive frames, the 2 nd frame to the 15 th frame predict the position of the target object according to the position, the moving direction and the moving speed of the target object in the image of the previous frame, and in the 4 th frame, the 7 th frame, the 10 th frame and the 13 th frame which are separated by 3 frames, the position of the target object is determined according to the feature information of the target object, and the predicted position is adjusted.
The comprehensive scheme can quickly and efficiently position the target object in the next frame of image of the first video stream and display the locking mark, and can adjust the predicted position at a proper frequency, so that the target object can be quickly and efficiently continuously tracked, and the accuracy of target tracking is considered.
As a further optional implementation manner, the step S104 of determining the second video stream according to the last determined region to the target object in the first video stream includes:
s1041, obtaining an alternative background image of an alternative video stream;
specifically, in the embodiment of the present invention, a camera closer to the first camera in a physical position (that is, an actual spatial position, which can be represented by latitude and longitude coordinates, etc.) may be selected as the candidate camera, a video stream of the candidate camera is obtained as the candidate video stream, and then a method similar to the above S1021 is adopted to obtain a background image of the candidate video stream, that is, the candidate background image.
Optionally, the candidate background image may be compared with the first background image according to the moving direction of the target object, and the unmatched candidate background image is removed. According to the embodiment of the invention, the alternative video stream is filtered firstly by the method, so that the efficiency of selecting the second video stream is improved.
S1042, comparing the alternative background image with the region of the target object determined for the last time in the first video stream, and selecting the alternative background image with the highest feature similarity;
specifically, the region of the target object is a region to which the position of the target object belongs, and since only the image of the position of the target object is compared with the alternative background image, it is difficult to find the alternative background image with a certain feature similarity, and the image of the region to which the position of the target object belongs is compared with the alternative background image, it is easier to find the alternative background image with a certain feature similarity.
And S1043, selecting a corresponding alternative video stream as a second video stream according to the alternative background image with the highest feature similarity.
Wherein there are several of the alternative video streams.
In the embodiment of the invention, when the target object disappears from the monitoring range of the first camera, the target object is in the boundary area of the monitoring range of the first camera, and at the moment, the alternative background image and the first background image are compared in the moving direction of the target object, so that unmatched alternative background images are filtered; comparing the alternative background image with the background image of the boundary area where the target object disappears in the first video stream, and selecting the alternative background image with the highest feature similarity; and selecting the corresponding alternative video stream as a second video stream according to the alternative background image with the highest feature similarity, so that a second camera for continuously tracking the target object can be quickly found. In addition, when the target object disappears from the monitoring range of the second camera shooting the second video stream, the third video stream and the third camera for continuous tracking can be selected according to the method, and the like, so that the cross-screen target automatic tracking of multipoint monitoring can be realized.
As a further alternative, the step S105 of determining the position of the target object in the second video stream and displaying the locking identifier of the target object may be implemented by using two schemes, as shown below.
For the first embodiment of step S105, it specifically is:
s1051, predicting the position of the target object in the second current frame image according to the position, the moving direction and the moving speed of the target object which are determined for the last time in the first video stream;
s1052, displaying a locking mark at the predicted position of the target object in the second current frame image;
and the second current frame image is a current frame image of the second video stream.
Specifically, the position of the target object in the second current frame image can be calculated according to the position, the moving direction and the moving speed of the target object tracked last time in the first video stream and by combining the interval time between the target object tracked last time in the first video stream and the second current frame image, so that the position of the target object can be quickly and efficiently positioned in the current frame image of the second video stream, and the locking mark can be displayed.
For the second embodiment of step S105, it is specifically:
s1053, obtaining a second background image of the second video stream;
s1054, obtaining a second current frame image of a second video stream;
s1055, obtaining a foreground image of a second current frame according to the second background image and the second current frame image;
s1056, determining the position of the target object in the foreground image of the second current frame according to the characteristic information;
and S1057, displaying a locking mark at the position of the target object in the second current frame image.
Specifically, the feature information of the target object may be compared with the feature information of all objects in the current frame foreground image of the second video stream to determine the position of the target object, so that the position of the target object may be accurately located in the current frame image of the second video stream and the locking identifier may be displayed.
In practical applications, two embodiments of step S105 may also be combined, such as: predicting the position of a target object in a current frame image of a second video stream shot by a second camera according to the position, the moving direction and the moving speed of the target object tracked in the first video stream for the last time; and further determining the position of the target object and displaying a locking identifier of the target object according to the characteristic information of the target object in the foreground image of the region to which the predicted position of the target object belongs.
The comprehensive scheme can predict the area to which the target object position belongs in the current frame image of the second video stream relatively quickly and efficiently, and then accurately position the target object in the current frame image of the second video stream and display the locking identification, so that the efficiency can be improved while the accuracy of the target object position is considered.
In summary, the cross-screen target tracking method provided in the embodiments of the present invention can continuously perform automatic tracking on a target object in the same video stream, and can also perform cross-screen automatic tracking on a target object in multiple video streams. It should be understood that embodiments of the present invention have a variety of applications including, but not limited to, automatic tracking of criminal suspects, automatic tracking of hit vehicles by transportation departments, tracking shots of contestants on a field, etc.
Referring to fig. 3, an embodiment of the present invention provides a cross-screen target tracking system, including:
the video stream acquisition module is used for acquiring a first video stream;
the target position determining and displaying module is used for determining a target object and the position thereof in a first current frame image of the first video stream and displaying a locking identifier at the position of the target object in the first current frame image;
the first tracking module is used for determining the position of a target object in a next frame image of the first video stream and displaying a locking identifier at the position of the target object in the next frame image;
the second video stream determining module is used for determining a second video stream according to the area of the target object determined in the first video stream for the last time;
and the second tracking module is used for determining the position of the target object in the second video stream and displaying the locking identification of the target object.
Referring to fig. 4, as a further alternative embodiment, the method further includes:
and the characteristic acquisition module is used for acquiring the characteristic information of the target object.
As a further optional implementation, the target position determining and displaying module includes:
a first acquiring unit configured to acquire a first background image of a first video stream;
a second obtaining unit, configured to obtain a first current frame image of the first video stream;
the third acquisition unit is used for acquiring a foreground image of the first current frame according to the first background image and the first current frame image;
the target object identification unit is used for determining a target object and the position thereof in the foreground image of the first current frame according to the characteristic information and extracting the outline of the target object;
and the target object display unit is used for displaying the locking identification at the position of the target object in the first current frame image according to the contour of the target object.
As a further optional implementation, the first tracking module includes:
a first prediction unit for determining a moving direction and a moving speed of a target object from video frames of a first video stream;
a second prediction unit for predicting a position of the target object in a next frame image of the first video stream based on the position, the moving direction, and the moving speed of the target object in the first current frame image;
a target tracking display unit for displaying a locking mark according to the predicted position of the target object in the next frame image;
or the like, or, alternatively,
a next frame image acquisition unit configured to acquire a next frame image of the first video stream;
a next frame foreground image obtaining unit, configured to obtain a next frame foreground image according to a first background image and the next frame image;
the target tracking identification unit is used for determining the position of a target object in the foreground image of the next frame according to the characteristic information;
and the target tracking display unit is used for displaying a locking mark at the position of the target object in the next frame image.
As a further optional implementation, the second video stream determining module includes:
a fourth obtaining unit, configured to obtain an alternative background image of the alternative video stream;
the first selection unit is used for comparing the alternative background image with the region of the target object determined for the last time in the first video stream and selecting the alternative background image with the highest feature similarity;
a second selecting unit, configured to select, according to the candidate background image with the highest feature similarity, a corresponding candidate video stream as a second video stream;
wherein there are several of the alternative video streams.
As a further optional implementation, the second tracking module includes:
a third prediction unit for predicting the position of the target object in the second current frame image according to the position, the moving direction and the moving speed of the target object determined last time in the first video stream;
the cross-screen tracking display unit is used for displaying a locking identifier at the predicted position of the target object in the second current frame image;
wherein the second current frame image is a current frame image of a second video stream;
or the like, or, alternatively,
a fifth acquiring unit, configured to acquire a second background image of the second video stream;
a sixth obtaining unit, configured to obtain a second current frame image of the second video stream;
a seventh obtaining unit, configured to obtain a second current frame foreground image according to the second background image and the second current frame image;
the cross-screen tracking identification unit is used for determining the position of a target object in a foreground image of a second current frame according to the characteristic information;
and the cross-screen tracking display unit is used for displaying the locking identification at the position of the target object in the second current frame image.
The contents in the above method embodiments are all applicable to the present system embodiment, the functions specifically implemented by the present system embodiment are the same as those in the above method embodiment, and the beneficial effects achieved by the present system embodiment are also the same as those achieved by the above method embodiment.
Referring to fig. 5, an embodiment of the present invention further provides a cross-screen target tracking apparatus, including:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement the cross-screen target tracking method.
The contents in the above method embodiments are all applicable to the present apparatus embodiment, the functions specifically implemented by the present apparatus embodiment are the same as those in the above method embodiments, and the advantageous effects achieved by the present apparatus embodiment are also the same as those achieved by the above method embodiments.
Furthermore, a storage medium is provided in an embodiment of the present invention, in which processor-executable instructions are stored, and when executed by a processor, the processor-executable instructions are used for executing the cross-screen target tracking method.
According to the cross-screen target tracking method, the system, the device and the storage medium, the target object is continuously locked in the same video stream by determining the position of the target object in the next frame image of the first video stream and displaying the locking mark at the position of the target object in the next frame image, so that the automatic tracking of the target object is realized; and determining a second video stream according to the area of the target object determined in the first video stream for the last time, determining the position of the target object in the second video stream, and displaying the locking identifier of the target object, so that when the target object disappears from the first video stream, the second video stream is accurately found to continuously lock the target object, cross-screen target automatic tracking is realized, and the target object is prevented from being lost. The invention can be applied to scenes such as automatic tracking of criminal suspects, automatic tracking of traffic departments to hit-and-run vehicles, tracking and shooting of contestants on a competition field and the like.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A cross-screen target tracking method is characterized by comprising the following steps:
acquiring a first video stream;
determining a target object and the position thereof in a first current frame image of a first video stream, and displaying a locking identifier at the position of the target object in the first current frame image;
determining the position of a target object in a next frame image of a first video stream, and displaying a locking mark at the position of the target object in the next frame image;
determining a second video stream according to the region of the target object determined for the last time in the first video stream;
the position of the target object is determined in the second video stream and the locking identification of the target object is displayed.
2. The cross-screen target tracking method according to claim 1, further comprising the steps of:
and acquiring characteristic information of the target object.
3. The cross-screen target tracking method of claim 2, wherein the step of determining the target object and its position in the first current frame image of the first video stream, and displaying the lock flag at the position of the target object in the first current frame image comprises:
acquiring a first background image of a first video stream;
acquiring a first current frame image of a first video stream;
acquiring a first current frame foreground image according to the first background image and the first current frame image;
determining a target object and the position thereof in the foreground image of the first current frame according to the characteristic information, and extracting the contour of the target object;
and displaying a locking mark at the position of the target object in the first current frame image according to the outline of the target object.
4. The method of claim 3, wherein the step of determining the position of the target object in the next frame of image of the first video stream and displaying the lock indicator at the position of the target object in the next frame of image comprises:
determining the moving direction and the moving speed of a target object according to the video frames of the first video stream;
predicting the position of a target object in a next frame image of the first video stream according to the position, the moving direction and the moving speed of the target object in the first current frame image;
displaying a locking mark in the next frame image according to the predicted position of the target object;
or the like, or, alternatively,
acquiring a next frame image of the first video stream;
acquiring a next frame of foreground image according to the first background image and the next frame of image;
determining the position of a target object in the next frame of foreground image according to the characteristic information;
and displaying a locking mark at the position of the target object in the next frame of image.
5. The method of claim 1, wherein the step of determining the second video stream according to the last determined area of the target object in the first video stream comprises:
acquiring an alternative background image of an alternative video stream;
comparing the alternative background image with the region of the target object determined for the last time in the first video stream, and selecting the alternative background image with the highest feature similarity;
selecting a corresponding alternative video stream as a second video stream according to the alternative background image with the highest feature similarity;
wherein there are several of the alternative video streams.
6. The method according to any one of claims 1 to 5, wherein the step of determining the position of the target object in the second video stream and displaying the lock identification of the target object comprises:
predicting the position of the target object in the second current frame image according to the position, the moving direction and the moving speed of the target object which are determined for the last time in the first video stream;
displaying a locking identifier at the predicted position of the target object in the second current frame image;
wherein the second current frame image is a current frame image of a second video stream;
or the like, or, alternatively,
acquiring a second background image of a second video stream;
acquiring a second current frame image of a second video stream;
acquiring a second current frame foreground image according to a second background image and the second current frame image;
determining the position of a target object in a foreground image of a second current frame according to the characteristic information;
and displaying the locking identification at the position of the target object in the second current frame image.
7. A cross-screen target tracking system, comprising:
the video stream acquisition module is used for acquiring a first video stream;
the target position determining and displaying module is used for determining a target object and the position thereof in a first current frame image of the first video stream and displaying a locking identifier at the position of the target object in the first current frame image;
the first tracking module is used for determining the position of a target object in a next frame image of the first video stream and displaying a locking identifier at the position of the target object in the next frame image;
the second video stream determining module is used for determining a second video stream according to the area of the target object determined in the first video stream for the last time;
and the second tracking module is used for determining the position of the target object in the second video stream and displaying the locking identification of the target object.
8. The cross-screen target tracking system of claim 7, further comprising:
and the characteristic acquisition module is used for acquiring the characteristic information of the target object.
9. A cross-screen target tracking apparatus, comprising:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement a cross-screen target tracking method as claimed in any one of claims 1 to 6.
10. A storage medium having stored therein processor-executable instructions, which when executed by a processor, are configured to perform a cross-screen target tracking method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010372081.3A CN111709978A (en) | 2020-05-06 | 2020-05-06 | Cross-screen target tracking method, system, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010372081.3A CN111709978A (en) | 2020-05-06 | 2020-05-06 | Cross-screen target tracking method, system, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111709978A true CN111709978A (en) | 2020-09-25 |
Family
ID=72536583
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010372081.3A Pending CN111709978A (en) | 2020-05-06 | 2020-05-06 | Cross-screen target tracking method, system, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111709978A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115826898A (en) * | 2023-01-03 | 2023-03-21 | 南京芯驰半导体科技有限公司 | Cross-screen display method, system, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106920247A (en) * | 2017-01-19 | 2017-07-04 | 博康智能信息技术有限公司上海分公司 | A kind of method for tracking target and device based on comparison network |
| WO2018121287A1 (en) * | 2016-12-30 | 2018-07-05 | 纳恩博（北京）科技有限公司 | Target re-identification method and device |
| CN109325967A (en) * | 2018-09-14 | 2019-02-12 | 腾讯科技（深圳）有限公司 | Method for tracking target, device, medium and equipment |
| CN110516620A (en) * | 2019-08-29 | 2019-11-29 | 腾讯科技（深圳）有限公司 | Method for tracking target, device, storage medium and electronic equipment |
| CN110866480A (en) * | 2019-11-07 | 2020-03-06 | 浙江大华技术股份有限公司 | Object tracking method and device, storage medium and electronic device |
-
2020
- 2020-05-06 CN CN202010372081.3A patent/CN111709978A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018121287A1 (en) * | 2016-12-30 | 2018-07-05 | 纳恩博（北京）科技有限公司 | Target re-identification method and device |
| CN106920247A (en) * | 2017-01-19 | 2017-07-04 | 博康智能信息技术有限公司上海分公司 | A kind of method for tracking target and device based on comparison network |
| CN109325967A (en) * | 2018-09-14 | 2019-02-12 | 腾讯科技（深圳）有限公司 | Method for tracking target, device, medium and equipment |
| CN110516620A (en) * | 2019-08-29 | 2019-11-29 | 腾讯科技（深圳）有限公司 | Method for tracking target, device, storage medium and electronic equipment |
| CN110866480A (en) * | 2019-11-07 | 2020-03-06 | 浙江大华技术股份有限公司 | Object tracking method and device, storage medium and electronic device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115826898A (en) * | 2023-01-03 | 2023-03-21 | 南京芯驰半导体科技有限公司 | Cross-screen display method, system, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104106260B (en) | Control based on geographical map | |
| JP5001260B2 (en) | Object tracking method and object tracking apparatus | |
| CN101918989B (en) | Video surveillance system with object tracking and retrieval | |
| US8886634B2 (en) | Apparatus for displaying result of analogous image retrieval and method for displaying result of analogous image retrieval | |
| CN110706259A (en) | Space constraint-based cross-shot tracking method and device for suspicious people | |
| CN112883819A (en) | Multi-target tracking method, device, system and computer readable storage medium | |
| US9092699B2 (en) | Method for searching for objects in video data received from a fixed camera | |
| EP3084577A1 (en) | Selection and tracking of objects for display partitioning and clustering of video frames | |
| CN107392932A (en) | Update the method and apparatus of the background model of the background subtraction for image | |
| CN1107282C (en) | Method of detecting moving objects in chronologically successive images | |
| CN101141633A (en) | Moving object detecting and tracing method in complex scene | |
| EP2951783B1 (en) | Method and system for detecting moving objects | |
| JP3970877B2 (en) | Tracking device and tracking method | |
| KR101645959B1 (en) | The Apparatus and Method for Tracking Objects Based on Multiple Overhead Cameras and a Site Map | |
| GB2443739A (en) | Detecting image regions of salient motion | |
| CN104680504A (en) | Scene change detection method and device thereof | |
| KR20110074107A (en) | Method for detecting object using camera | |
| KR20110035662A (en) | Intelligent image search method and system using surveillance camera | |
| CN116311063A (en) | Personnel fine granularity tracking method and system based on face recognition under monitoring video | |
| JP2017125974A (en) | Display device, display method and program | |
| CN109816700B (en) | Information statistical method based on target identification | |
| KR101826669B1 (en) | System and method for video searching | |
| CN109086717B (en) | Violent behavior detection system and method based on human skeleton and motion signal characteristics | |
| CN111709978A (en) | Cross-screen target tracking method, system, device and storage medium | |
| JP2014110020A (en) | Image processor, image processing method and image processing program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |