WO2020258889A1

WO2020258889A1 - Tracking method of video tracking device, and video tracking device

Info

Publication number: WO2020258889A1
Application number: PCT/CN2020/075481
Authority: WO
Inventors: 高宗伟
Original assignee: 杭州海康微影传感科技有限公司
Priority date: 2019-06-25
Filing date: 2020-02-17
Publication date: 2020-12-30
Also published as: CN112132858A

Abstract

Embodiments of the present application provide a tracking method of a video tracking device, and a video tracking device. The video tracking device supports multiple tracking algorithms. The method comprises: obtaining a video image and a parameter representing the scene complexity of the video image from an imaging device; selecting a tracking algorithm suitable for the video image from the multiple tracking algorithms on the basis of the parameter; determining a tracking target in the video image according to the tracking algorithm suitable for the video image; extracting position information of the tracking target in the video image, and calculating the distance from the tracking target to the center of the video image on the basis of the position information; and adjusting a capturing angle of the imaging device according to the distance, so that the tracking target is located at the center of the next video image captured by the imaging device. By applying the solution provided in the embodiments of the present application for target tracking, adaptive tracking algorithm adjustment can be achieved depending on scene change, thereby ensuring a tracking effect.

Description

Tracking method of video tracking equipment and video tracking equipment

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on June 25, 2019 with the application number 201910555419.6 and the invention titled "A tracking method and video tracking device for a video tracking device", the entire content of which is incorporated by reference In this application.

Technical field

This application relates to the field of security technology, and in particular to a tracking method of a video tracking device and a video tracking device.

Background technique

Video tracking devices such as existing video trackers usually use a single tracking algorithm for target tracking. The main tracking algorithms are contrast tracking algorithm, correlation tracking algorithm and binary tracking algorithm. Among them, the related tracking algorithm can track multiple types of targets. When the tracked target has no boundaries, the motion is not very strong, and the scene is more complicated, the tracking effect of the related tracking algorithm is good. The contrast tracking algorithm can track fast-moving targets and is highly adaptable to changes in target posture. The binary tracking algorithm can automatically detect the target, the tracking gate is adaptive to the target size, the closed loop speed is fast, the tracking is stable, and it is suitable for tracking the air target.

Since each tracking algorithm has its focus on application scenarios, when a video tracking device uses a single tracking algorithm for target tracking, the tracking effect of the video tracking device will be different when used in different scenarios.

Summary of the invention

In view of this, the purpose of this application is to provide a tracking method for a video tracking device and a video tracking device, which can adaptively adjust the tracking algorithm according to scene changes, thereby ensuring the tracking effect.

In order to achieve the foregoing objectives, the embodiments of the present application provide the following technical solutions:

In the first aspect, an embodiment of the present application provides a tracking method for a video tracking device, the video tracking device supports multiple tracking algorithms; the method includes:

Acquiring a video image from an imaging device and a parameter that characterizes the scene complexity of the video image;

Selecting a tracking algorithm suitable for the video image from the multiple tracking algorithms based on the parameter;

Determining a tracking target in the video image according to a tracking algorithm applicable to the video image;

Extracting position information of the tracking target in the video image, and calculating the distance between the tracking target and the center of the video image based on the position information;

The shooting angle of the imaging device is adjusted according to the distance, so that the tracking target is at the center of the next video image captured by the imaging device.

In a second aspect, an embodiment of the present application provides a video tracking device that supports multiple tracking algorithms; the video tracking device includes a non-transitory computer-readable storage medium and a processor, wherein:

The non-transitory computer-readable storage medium is used to store instructions that can be executed by the processor, and when the instructions are executed by the processor, the processor is caused to:

It can be seen from the above technical solutions that, in the solution provided by the embodiments of the present application, for each frame of video image taken by the imaging device, the tracking algorithm suitable for the frame of video image is adaptively selected based on the parameters that characterize the scene complexity of the frame of video image , Based on the selected tracking algorithm to track the tracking target in the frame of the video image, and adjust the shooting angle of the imaging device according to the distance between the tracking target in the video image and the center of the video image, so that the tracking target is in the next video shot by the imaging device The center position of the image. Since the tracking algorithm can be switched at any time according to the parameters that characterize the scene complexity of each frame of video image, it can well adapt to the changes in the shooting scene of the imaging device, thereby ensuring the video tracking effect.

Description of the drawings

In order to explain the embodiments of the present application and the technical solutions of the prior art more clearly, the following briefly introduces the drawings needed in the embodiments and the prior art. Obviously, the drawings in the following description are only the present For some of the embodiments of the application, for those of ordinary skill in the art, other drawings can be obtained from these drawings without creative work.

FIG. 1 is a schematic diagram of the architecture of a video tracking system provided by an embodiment of the present application;

FIG. 2 is a flowchart of a tracking method of a video tracking device provided by an embodiment of the present application;

Fig. 3 is a schematic structural diagram of a video tracking device provided by an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions, and advantages of the present application clearer, the following further describes the present application in detail with reference to the drawings and embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of this application.

In the prior art, different tracking algorithms are applicable to different scenarios, and different scenarios are mostly distinguished by the complexity of the scenarios. In the embodiment of the present application, the tracking algorithm is adaptively adjusted according to the different complexity of each frame of video image captured by the imaging device, so as to avoid the difference in tracking effect caused by the change of the imaging device shooting scene.

Refer to Fig. 1, which is a schematic diagram of the architecture of a video tracking system provided by an embodiment of the present application. The video tracking system includes: imaging equipment, video tracking equipment, display terminals, and follow-up control equipment. The following are introduced separately.

1) Imaging equipment

The imaging device is integrated with an imaging movement, which is used to shoot videos, and the imaging device sends each frame of video image taken by the imaging movement to the video tracking device. In addition, in order to enable the video tracking device to adaptively switch the tracking algorithm according to the changes in the shooting scene, the imaging device will not only send each frame of video image captured by the imaging core to the video tracking device, but also characterize each frame of video image. The parameters of the scene complexity are sent to the video tracking device. In an embodiment, the parameters that characterize the scene complexity of each frame of video image may be represented by the definition evaluation parameter of each frame of video image.

In one embodiment, the imaging core integrated with the imaging device may include a thermal imaging core and a visible light imaging core. Both the thermal imaging core and the visible light imaging core are used to capture video images, the former captures the thermal imaging video image, and the latter captures the visible light video image. The imaging device can send both the thermal imaging video image captured by the thermal imaging core and the visible light video image captured by the visible light imaging core to the video tracking device. In addition, if the subsequent video tracking device performs target tracking on the thermal imaging video image, the imaging device needs to send the thermal imaging video image to the video tracking device and also need to send the parameters that characterize the scene complexity of the thermal imaging video image to the video. Tracking equipment. In the same way, if the video tracking device performs target tracking on the visible light video image, the imaging device needs to send the visible light video image to the video tracking device and also need to send the parameters representing the scene complexity of the visible light video image to the video tracking device.

2) Video tracking equipment

Video tracking equipment can be implemented using a Digital Signal Processing (DSP) chip and supports multiple tracking algorithms.

For each frame of video image sent by the imaging device, the video tracking device selects an appropriate tracking algorithm for target tracking based on the parameters that characterize the scene complexity of each frame of video image, and obtains the position information of the tracking target in each frame of video image, according to the above The position information obtains the tracking result, and then adjusts the shooting angle of the imaging device according to the tracking result, so that the tracking target is at the center position of the video image subsequently captured by the imaging device. In specific implementation, the video tracking device can send the tracking result to the follow-up control device, and the follow-up control device adjusts the shooting angle of the imaging device.

Wherein, the tracking result may be: the distance between the tracking target and the center of the video image in the video image calculated based on the position information.

In one embodiment, the video tracking device performs target tracking on the thermal imaging video image sent by the imaging device, and the visible light video image sent by the imaging device is directly sent to the display device for display. In one embodiment, in order to ensure that the video images are processed synchronously, the imaging device and the video tracking device need to maintain timing synchronization. In specific implementation, the video tracking device can use the synchronization signal of the thermal imaging core or the visible light imaging core in the imaging device as the synchronization source, and use the aforementioned synchronization source as the basic working sequence of the video tracking device, so as to ensure that it is connected to the thermal imaging core or The visible light imaging movement is strictly synchronized. The video tracking equipment generates the working timing of each unit circuit on this basis, forming a unified timing synchronization system.

3) Follow-up control equipment

The follow-up control device executes the follow-up control operation according to the tracking result sent by the video tracking device. For example, the shooting angle of the imaging device is directly adjusted, or the shooting angle of the imaging device is adjusted by moving and/or rotating the follow-up control device itself, so as to ensure that the tracking target is located in the center of the video image captured by the imaging device.

The following describes in detail the tracking method of the video tracking device provided by the embodiment of the present application with reference to FIG. 2. Figure 2 is an example of target tracking of thermal imaging video images captured by the thermal imaging core in the imaging device. In actual implementation, the video tracking device can also capture the visible light captured by the visible light imaging core in the imaging device. Video image for target tracking.

In addition, the solution provided by the embodiment of the present application is applied to a video tracking device for tracking a tracking target in a video image.

Refer to FIG. 2, which is a flowchart of a tracking method of a video tracking device provided in an embodiment of the present application. As shown in FIG. 2, the method includes the following steps 201 to 205.

Step 201: Obtain a video image and a parameter that characterizes the scene complexity of the video image from the imaging device.

Since the imaging device sequentially shoots video images in chronological order, the video tracking device can also obtain video images from the imaging device frame by frame. For each video image acquired by the video tracking device from the imaging device, the target tracking method is the same, and all of them can be implemented through step 201 to step 205 provided in the embodiment of the present application.

In an embodiment, the definition evaluation parameter of the video image may be used as a parameter that characterizes the scene complexity of the video image. The definition evaluation parameter of the video image can be expressed by using an active image format descriptor (Active Format Description, AFD). The size of the AFD value can represent the complexity of the scene corresponding to the video image to a certain extent. Specifically, the larger the AFD value, the higher the scene complexity of the video image. Conversely, the smaller the AFD value, the lower the scene complexity of the video image. According to the AFD value of the video image, the complexity of the scene can be determined. For example, there are two frames of video images, the value range of the AFD value of the first frame of video image is [5000, 10000], and the value range of the AFD value of the second frame of video image is [500, 1000], then The scene complexity of one frame of video image is higher than the scene complexity of the second frame of video image.

Among them, the AFD value of a frame of video image can be obtained through the thermal imaging core. Specifically, the processor of the thermal imaging core performs high-frequency filtering processing on the parameters that characterize the clarity of a frame of video image to obtain a set of data. Therefore, each data in the group of data can be used to characterize the clarity of the video image. The larger the data in the group of data, the higher the clarity of the video image. On the contrary, the smaller the data in the group of data, the higher the clarity of the video image. The lower the level of clarity. Each data in this set of data can be referred to as the aforementioned AFD value.

In an embodiment, the processor of the thermal imaging core may be a Field Programmable Gate Array (FPGA).

In one embodiment, the imaging device may integrate a thermal imaging core, a visible light core, or other imaging cores capable of performing video shooting, and use the integrated imaging core to capture video images and send each frame of the captured video images to Video tracking equipment. In addition, the imaging device also analyzes each frame of video image captured by the imaging core to determine its definition evaluation parameters, and sends the definition evaluation parameters of each frame of video image as a parameter that characterizes the scene complexity of each frame of video image to the video. Tracking equipment.

In one embodiment, only one imaging core is integrated in the imaging device. After the imaging device sends each frame of video image captured by the imaging core to the video tracking device, the video tracking device needs to track each frame of video image on the one hand. On the other hand, it can also send each frame of video image to the display terminal for display.

In another embodiment, multiple imaging cores can be integrated in the imaging device, the imaging device sends each frame of video image captured by one of the imaging cores to the video tracking device, and the video tracking device performs target tracking on each frame of video image; In addition, each frame of video images captured by other imaging cores can be sent to a video tracking device, and the video tracking device performs other video processing on each frame of video images captured by other imaging cores, for example, sending them to a display terminal for display.

When multiple imaging cores are integrated in an imaging device, for example, a thermal imaging core and a visible light imaging core are integrated in the imaging device. The imaging device integrating the visible light imaging core and the thermal imaging core performs video shooting of the current scene, where the thermal imaging core in the imaging device captures the thermal imaging video image of the current scene, and the visible light imaging core in the imaging device captures Obtain the visible light video image of the current scene. In addition, the imaging device also analyzes the thermal imaging video image of the current scene to determine its sharpness evaluation parameters. The imaging device sends each frame of thermal imaging video image captured by the thermal imaging core and each frame of visible light video image captured by the visible light imaging core to the video tracking device, and also uses the definition evaluation parameters of each frame of thermal imaging video image as a characterization The parameters of the scene complexity of each thermal imaging video image are sent to the video tracking device.

Step 202: Select a tracking algorithm suitable for the video image from among multiple tracking algorithms based on the parameter that characterizes the scene complexity of the video image.

In order to enable the video tracking device to switch tracking algorithms according to changes in the shooting scene, the video tracking device can support multiple tracking algorithms such as contrast tracking algorithms, related tracking algorithms, and binary tracking algorithms.

Since different tracking algorithms are suitable for different shooting scenes, different shooting scenes are distinguished according to the complexity of the scene. In view of this, in one embodiment, the parameter value range corresponding to each tracking algorithm can be preset. The scene complexity of the applicable shooting scene is consistent with the scene complexity represented by the parameter value range corresponding to each tracking algorithm.

For example, the definition evaluation parameter of the video image can be used as a parameter that characterizes the scene complexity of the video image. In this case, the preset value range of the parameter corresponding to each tracking algorithm is the definition corresponding to each tracking algorithm Evaluation parameter value range.

For example, the AFD value output by the imaging device can be divided into multiple levels according to the order of magnitude:...n-1th level, nth level, n+1th level..., where n represents the serial number of the level. Assume that the n-1th level is AFD=40,000, the nth level is AFD=70000, and the n+1th level is AFD=100,000. When the AFD value is near the n-1th level, that is, when the AFD value is near 40,000, for example, 40,000±15,000, it means that the current scene is a simple scene, such as a sky scene, a sea scene, and the video tracking device selects binary tracking The algorithm is more suitable for target tracking. When the AFD value is near the nth level, that is, when the AFD value is near 70,000, for example, 70,000±15000, it indicates that the current scene is more complicated, and the video tracking device chooses the contrast tracking algorithm for target tracking; when the AFD value is in the nth When the +1 level is near, that is, when the AFD value is near 100000, for example, 100000±15000, it indicates that the current scene is very complicated, and it is more appropriate for the video tracking device to select the relative tracking algorithm for target tracking.

According to the level of the above AFD value, the following parameter value ranges can be set:..., (25000, 55000), (55000, 85000), (85000, 115000)..., among them, the parameter value range corresponding to the binary tracking algorithm Yes (25000, 55000), the parameter value range corresponding to the contrast tracking algorithm is (55000, 85000), and the parameter value range corresponding to the correlation tracking algorithm is (85000, 115000). Therefore, if the AFD value of a frame of thermal imaging video image received by the video tracking device from the imaging device is 50000, since 50000 belongs to the parameter value range (25000, 55000), the parameter value range corresponds to the binary tracking algorithm, so The video tracking device determines the binary tracking algorithm as a tracking algorithm suitable for the frame of thermal imaging video image, and uses the binary tracking algorithm to track the tracking target in the frame of thermal imaging video image.

Since the parameter value range corresponding to each tracking algorithm is preset, the video tracking device can evaluate the definition of the frame of video image after obtaining a frame of video image and the definition evaluation parameters of the frame of video image from the imaging device The parameter is compared with the parameter value range corresponding to each tracking algorithm, and the parameter value range corresponding to the definition evaluation parameter of the frame of video image is found, and the tracking algorithm corresponding to the parameter value range is determined as suitable for the frame Video image tracking algorithm.

Specifically, the parameter value range corresponding to the definition evaluation parameter of the video image may be the parameter value range to which the definition evaluation parameter of the video image belongs.

In view of the foregoing, in one embodiment, in this step 202, a specific method for selecting a tracking algorithm suitable for a video image from among multiple tracking algorithms based on a parameter that characterizes the scene complexity of the video image is shown in the following S11-S12.

S11. Compare the above-mentioned parameters with the parameter value ranges corresponding to each preset tracking algorithm.

By comparing the above parameters with the parameter value ranges corresponding to various tracking algorithms, it can be determined which parameter value range the above parameters belong to.

For example, assuming that the above parameters are AFD values, and the AFD value is 44000, the parameter value ranges corresponding to various tracking algorithms are: the parameter value range corresponding to algorithm 1 (25000, 55000), the parameter value range corresponding to algorithm 2 ( 55000, 85000), the parameter value range corresponding to Algorithm 3 (85000, 115000).

The value range of the parameter to which 44000 belongs is: (25000, 55000).

S12. Determine the tracking algorithm corresponding to the parameter value range to which the above-mentioned parameter belongs as the tracking algorithm suitable for the video image.

In the example shown in S11 above, the parameter value range of 44000 is: (25000, 55000), and the tracking algorithm corresponding to the parameter value range of (25000, 55000) is Algorithm 1, so it is suitable for video images The tracking algorithm is Algorithm 1.

Step 203: Determine the tracking target in the video image according to the tracking algorithm suitable for the video image.

Generally, a frame of video image includes more than one type of object. In order to identify the tracking target in it, it is necessary to use all the objects as the detection target, and identify whether they are tracking targets one by one.

The following describes the specific tracking algorithm:

I) Contrast tracking algorithm

For the contrast tracking algorithm, when using the contrast tracking algorithm to track a frame of video image, it is necessary to extract the tracking information for the contrast tracking algorithm corresponding to each detected target from the frame of the video image, for example, edge information, contour The length, area, center of gravity, and/or centroid, etc., and then compare the tracking information of all detected targets with the tracking information of the detected target in the previous frame of video image to find out the tracking information and the detected target in the previous frame of video image The detection target with the greatest matching degree of tracking information is determined as the tracking target in the frame of video image.

Among them, before extracting the tracking information for the contrast tracking algorithm corresponding to each detection target from a frame of video image, the frame of video image can also be preprocessed, and the specific method is shown as X1-X3.

X1. De-noise processing on the video image.

For example, in this step, Gaussian filtering can be used to achieve noise removal processing, and Gaussian filtering of the video image can remove scattered noise in the video image.

X2. Perform edge detection on the video image that has been denoised.

Edge detection is mainly to identify pixels with obvious brightness changes in video images. For example, the algorithm for edge detection can include two types of edge detection algorithms based on search and zero-crossing.

X3. Determine the segmentation threshold range of the edge-detected video image, and perform binarization processing on the edge-detected video image according to the segmentation threshold range.

The determination of the segmentation threshold range is for the subsequent binarization of the edge-detected video image. For example, the segmentation threshold range can be determined according to the gray limit value of the video image and the maximum video signal amplitude. Wherein, the gray limit value can be the maximum gray value or the minimum gray value. Specifically, the lower limit T _{V_min of the} segmentation threshold range can be _calculated using T _{V_min} =P-α·V _P ; the upper limit T _{V_max of the} segmentation threshold range can be _calculated using the formula T _{V_max} =P+α·V _P ; where P is The gray limit value of the video image; V _P is the maximum video signal amplitude of the video image, for example, 700 mV; α is the preset contrast parameter value, and the value range can be 5% to 15% of the value of V _P.

When the edge-detected video image is binarized, the gray values of pixels with gray values lower than T _{V_min} can be uniformly set to 0, and the gray values of pixels with gray values greater than T _{V_max} can be uniform. Set to 255, and the gray value of the pixel with gray value between T _{V_min} and T _{V_max} can remain unchanged.

After the video image is preprocessed, the tracking information for the contrast tracking algorithm corresponding to each detection target can be extracted from the preprocessed video image, so that according to the extracted tracking information of all detection targets, from the video image The tracking target is filtered out of all detection targets.

Therefore, when the tracking algorithm suitable for the video image is a contrast tracking algorithm, the tracking target in the video image is determined according to the tracking algorithm suitable for the video image, which specifically includes:

S21: For each detection target in the video image, extract tracking information of the detection target used for the contrast tracking algorithm;

S22: According to the tracking information used for the contrast tracking algorithm of all the detected targets, the tracking target is selected from all the detected targets.

It can be seen from the above description that when the contrast tracking algorithm is used to track the target, it is necessary to use the tracking information of the detected target in the previous frame of video image, but for the first frame of video image to start tracking the target, there is no previous frame. One frame of video image. In view of this, in one embodiment, the tracking target can be determined in the first frame of video image by manual designation.

II) Binary tracking algorithm

For the binary tracking algorithm, when using the binary tracking algorithm to track a frame of video image, it is necessary to extract the tracking information for the binary tracking algorithm corresponding to each detection target from the video image, for example, the aforementioned tracking information It can include edge information, contour length, area, center of gravity, and/or centroid, and then compare the tracking information of all detected targets with the tracking information of the detected target in the previous frame of video image to find out the tracking information and the previous frame of video The detection target in the image with the greatest matching degree of the tracking information of the detection target is determined as the tracking target in the frame of the video image.

Among them, before extracting the tracking information for the contrast tracking algorithm corresponding to each detection target from a frame of video image, the frame of video image can also be preprocessed, and the preprocessing method is the same as that in the contrast tracking algorithm . After the preprocessing, it is also necessary to perform area filling for each detection target in the frame of the video image after the preprocessing. The area filling can more highlight each detection target in the video image.

After the video image is preprocessed and area filled, the tracking information for the binary tracking algorithm corresponding to each detection target can be extracted from the preprocessed and area filled video image, so as to correspond to all the extracted detection targets The tracking information is selected from all the detected targets in the frame of video image.

Therefore, when the tracking algorithm suitable for the video image is a binary tracking algorithm, the tracking target in the video image is determined according to the tracking algorithm suitable for the video image, which specifically includes the following S31 and S32.

S31. For each detection target in the video image, extract tracking information corresponding to the detection target for the binary tracking algorithm;

S32. According to the tracking information used for the binary tracking algorithm of all the detection targets, the tracking target is selected from all the detection targets.

It can be seen from the above description that when a binary tracking algorithm is used to track a target, it is necessary to use the tracking information of the detected target in the previous frame of video image, but for the first frame of video image to start tracking the target, there is no In view of this, in one embodiment, the previous frame of video image may be manually designated to determine the tracking target in the first frame of video image.

III) Related tracking algorithms

For related tracking algorithms, there is no need to extract the tracking information of each detection target from the video image, just match the preselected template image containing the tracking target with each detection target in the video image to maximize the degree of matching The detection target is used as the tracking target in the video image. Among them, before matching the pre-selected template image containing the tracking target with each detection target in the video image, the video image can also be preprocessed, and the preprocessing method is the same as the preprocessing method in the contrast tracking algorithm.

Here, the pre-selected template image containing the tracking target can be preset or extracted from previous video images. For example, the image containing the tracking target is extracted from the video image where the tracking target appears for the first time as the template image , Or extract the image containing the tracking target from the previous frame of video image as a template image.

Therefore, when the tracking algorithm suitable for the video image is the related tracking algorithm, the tracking target in the video image is determined according to the tracking algorithm suitable for the video image, which specifically includes the following S41.

S41. According to a pre-selected template image containing a tracking target, a detection target with the greatest degree of matching with the template image is selected from all detection targets in the video image, and the screened detection target is determined as the tracking target of the video image.

Step 204: Extract the position information of the tracking target in the video image, and calculate the distance between the tracking target and the center of the video image based on the position information.

In an embodiment, the location information of the tracking target mainly includes information such as the height, width, coordinates of the tracking target. Among them, the height and width of the tracking target can be determined according to the projection of the video image on the x-axis and y-axis in the three-dimensional coordinate system. For example, the projection on the x-axis falls into the interval [x1, x2], on the y-axis If the projection falls within the interval [y1, y2], it can be determined that the width of the detection target is x2-x1, the height is y2-y1, and the center point coordinates of the tracking target are: ((x1+x2)/2, (y1+y2 )/2).

After the position information of the tracking target in the video image is determined, since the coordinates of the center of the video image are known, the distance between the tracking target and the center of the video image can be obtained by calculating the distance between the coordinates.

In addition, in one embodiment, the center of gravity and/or centroid of the tracking target can also be used as the location information of the tracking target. Furthermore, the location of a specific point on the tracking target can also be used as the location information of the tracking target. Such as tracking a corner point or protruding end point on the edge of the target.

Step 205: Adjust the shooting angle of the imaging device according to the above distance, so that the tracking target is at the center of the next video image shot by the imaging device.

In one embodiment, the specific implementation of adjusting the shooting angle of the imaging device according to the above distance is: sending the distance between the tracking target and the center of the video image to the follow-up control device equipped with the video tracking device, so that the follow-up control device is based on the foregoing The distance performs the follow-up control operation to adjust the shooting angle of the imaging device.

In one case, the position between the imaging device and the video tracking device is very close or directly integrated, and both are installed on the follow-up control device and move with the movement of the follow-up control device.

After the video tracking device determines the distance between the tracking target in the video image of the current scene and the center of the video image, it can send this distance to the follow-up control device, and the follow-up control device can drive the imaging device to move by controlling its own movement, or directly The imaging device is controlled to rotate or move so that the tracking target is located at the center of the next frame of video image captured by the imaging device.

It can be seen from the above that, in the solutions provided by the foregoing embodiments, for each frame of video image captured by the imaging device, the tracking algorithm suitable for the frame of video image is adaptively selected based on the parameter that characterizes the scene complexity of the frame of video image. The tracking algorithm tracks the tracking target in this frame of video image, and adjusts the shooting angle of the imaging device according to the distance between the tracking target in the video image and the center of the video image, so that the tracking target is at the center of the next video image captured by the imaging device position. Since the tracking algorithm can be switched at any time according to the parameters that characterize the scene complexity of each frame of video image, it can well adapt to the changes in the shooting scene of the imaging device, thereby ensuring the video tracking effect.

The tracking method of the video tracking device provided by the embodiment of the application is described in detail above, and the embodiment of the application also provides a video tracking device, which is described in detail below with reference to FIG. 3.

Referring to FIG. 3, FIG. 3 is a schematic structural diagram of a video tracking device provided by an embodiment of the present application. As shown in FIG. 3, the video tracking device 300 includes a processor 301 and a non-transitory computer-readable storage medium 302, wherein,

The non-transitory computer-readable storage medium 302 is configured to store instructions that can be executed by the processor 301, and when the instructions are executed by the processor 301, the processor 301 is caused to:

Acquiring a video image from an imaging device and a parameter characterizing the scene complexity of the frame video image;

In one embodiment, in the device shown in FIG. 3, an imaging core is integrated in the imaging device;

The processor 301, when acquiring a video image and a parameter characterizing the scene complexity of the video image from an imaging device, includes:

Receive the frame of the video image taken by the imaging core sent by the imaging device, and receive the definition evaluation parameter of the video image sent by the imaging device, wherein the definition evaluation parameter represents the video image The parameter of the scene complexity.

In one embodiment, in the device shown in FIG. 3, when the processor 301 selects a tracking algorithm suitable for the video image from among the multiple tracking algorithms based on the parameter, the method includes:

Compare the parameter value range corresponding to each preset tracking algorithm with the parameter;

The tracking algorithm corresponding to the parameter value range to which the parameter belongs is determined as the tracking algorithm suitable for the video image.

In one embodiment, in the device shown in FIG. 3, the multiple tracking algorithms include: a contrast tracking algorithm, a binary tracking algorithm, and a related tracking algorithm;

When the target tracking algorithm suitable for the video image is a contrast tracking algorithm or a binary tracking algorithm, when the processor 301 determines the tracking target in the video image according to the tracking algorithm suitable for the video image, it includes:

For each detection target in the video image, extract tracking information corresponding to the detection target for the target tracking algorithm;

Screening out the tracking target from all the detection targets according to the tracking information of all the detection targets used in the target tracking algorithm;

When the target tracking algorithm is a related tracking algorithm, when the processor 301 determines the tracking target in the video image according to the tracking algorithm applicable to the video image, the method includes:

According to the pre-selected template image containing the tracking target, the detection target with the greatest degree of matching with the template image is selected from all the detection targets of the video image, and the screened detection target is determined as the video image. Track the target.

In one embodiment, in the device shown in FIG. 3, when the processor 301 adjusts the shooting angle of the imaging device according to the distance, it includes: sending the distance to a follow-up control equipped with the video tracking device Device, so that the follow-up control device performs a follow-up control operation according to the distance, thereby adjusting the shooting angle of the imaging device.

It can be seen from the above that, when the video tracking device provided by the foregoing embodiments performs target tracking, for each frame of video image captured by the imaging device, adaptive selection is applied to the frame of video image based on the parameter that characterizes the scene complexity of the frame of video image. Tracking algorithm, based on the selected tracking algorithm to track the tracking target in the frame of video image, and adjust the shooting angle of the imaging device according to the distance between the tracking target in the video image and the center of the video image, so that the tracking target is under the shooting of the imaging device The center position of a video image. Since the video tracking device can switch the tracking algorithm at any time according to the parameters characterizing the scene complexity of each frame of video image, it can well adapt to changes in the shooting scene of the imaging device, thereby ensuring the video tracking effect.

Corresponding to the tracking method of the video tracking device described above, an embodiment of the present application also provides a tracking device of the video tracking device. The following describes the tracking device of the video tracking device provided in the embodiment of the present application.

In one embodiment, the aforementioned video tracking device supports multiple tracking algorithms; the tracking device of the aforementioned video tracking device includes:

An information acquisition module for acquiring a video image from an imaging device and parameters that characterize the scene complexity of the video image;

An algorithm selection module, configured to select a tracking algorithm suitable for the video image among the multiple tracking algorithms based on the parameter;

A target determination module, configured to determine a tracking target in the video image according to a tracking algorithm applicable to the video image;

A distance calculation module, configured to extract position information of the tracking target in the video image, and calculate the distance between the tracking target and the center of the video image based on the position information;

The angle adjustment module is configured to adjust the shooting angle of the imaging device according to the distance, so that the tracking target is at the center of the next video image shot by the imaging device.

In one embodiment, an imaging core is integrated in the imaging device;

The information acquisition module is specifically used for:

Receive a video image taken by the imaging core sent by an imaging device, and receive a sharpness evaluation parameter of the video image sent by the imaging device, wherein the sharpness evaluation parameter is a scene characterizing the video image Complexity parameter.

In an embodiment, the algorithm selection module is specifically used for:

In an embodiment, the multiple tracking algorithms include: a contrast tracking algorithm, a binary tracking algorithm, and a related tracking algorithm;

When the target tracking algorithm suitable for the video image is a contrast tracking algorithm or a binary tracking algorithm, the target determination module is specifically used for:

When the target tracking algorithm is a related tracking algorithm, the target determination module is specifically used for:

According to the pre-selected template image containing the tracking target, the detection target with the greatest degree of matching with the template image is selected from all detection targets in the video image, and the selected detection target is determined as the video image Tracking target.

In an embodiment, the angle adjustment module is specifically used for:

The distance is sent to a follow-up control device equipped with the video tracking device, so that the follow-up control device performs a follow-up control operation according to the distance, thereby adjusting the shooting angle of the imaging device.

Corresponding to the tracking method of the video tracking device described above, an embodiment of the present application also provides a computer-readable storage medium in which a computer program is stored, and the computer program is executed by a processor to realize this The steps of the tracking method of the video tracking device described in the application embodiment.

Corresponding to the tracking method of the video tracking device described above, an embodiment of the present application also provides a computer program product containing instructions that, when it runs on a computer, causes the computer to perform the tracking of the video tracking device described in the embodiment of the application. method.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in the form of a computer program product in whole or in part. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium 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 (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as 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 data center integrated with one or more available media. The usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, and a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)).

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply one of these entities or operations. There is any such actual relationship or order between. Moreover, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but also includes Other elements of, or also include elements inherent to this process, method, article or equipment. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other same elements in the process, method, article, or equipment including the element.

The various embodiments in this specification are described in a related manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of video tracking equipment, devices, computer-readable storage media, and computer program products, since they are basically similar to the method embodiments, the description is relatively simple. For related details, please refer to the description of the method embodiments .

The above are only the preferred embodiments of this application and are not intended to limit this application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in this application Within the scope of protection.

Claims

A tracking method for a video tracking device, characterized in that the video tracking device supports multiple tracking algorithms; the method includes:

Acquiring a video image from an imaging device and a parameter that characterizes the scene complexity of the video image;

Selecting a tracking algorithm suitable for the video image from the multiple tracking algorithms based on the parameter;

Determining a tracking target in the video image according to a tracking algorithm applicable to the video image;

Extracting position information of the tracking target in the video image, and calculating the distance between the tracking target and the center of the video image based on the position information;

The shooting angle of the imaging device is adjusted according to the distance, so that the tracking target is at the center of the next video image captured by the imaging device.
The method of claim 1, wherein an imaging core is integrated in the imaging device;

The step of obtaining a video image and a parameter characterizing the scene complexity of the video image from an imaging device includes:

Receive a video image taken by the imaging core sent by an imaging device, and receive a sharpness evaluation parameter of the video image sent by the imaging device, wherein the sharpness evaluation parameter is a scene characterizing the video image Complexity parameter.
The method according to claim 1, wherein the step of selecting a tracking algorithm suitable for the video image from the multiple tracking algorithms based on the parameter comprises:

Compare the parameter value range corresponding to each preset tracking algorithm with the parameter;

The tracking algorithm corresponding to the parameter value range to which the parameter belongs is determined as the tracking algorithm suitable for the video image.
The method according to claim 1, wherein the multiple tracking algorithms include: a contrast tracking algorithm, a binary tracking algorithm, and a correlation tracking algorithm;

When the target tracking algorithm suitable for the video image is a contrast tracking algorithm or a binary tracking algorithm, the step of determining the tracking target in the video image according to the tracking algorithm suitable for the video image includes:

For each detection target in the video image, extract tracking information corresponding to the detection target for the target tracking algorithm;

Filter out the tracking target from all the detection targets according to the tracking information corresponding to all the detection targets for the target tracking algorithm;

When the target tracking algorithm is a related tracking algorithm, the step of determining the tracking target in the video image according to the tracking algorithm applicable to the video image includes:

According to the pre-selected template image containing the tracking target, the detection target with the greatest degree of matching with the template image is selected from all detection targets in the video image, and the selected detection target is determined as the video image Tracking target.
The method according to any one of claims 1-4, wherein the step of adjusting the shooting angle of the imaging device according to the distance comprises:

The distance is sent to a follow-up control device equipped with the video tracking device, so that the follow-up control device performs a follow-up control operation according to the distance.
A video tracking device, characterized in that the video tracking device supports multiple tracking algorithms; the video tracking device includes a non-transitory computer-readable storage medium and a processor, wherein,

The non-transitory computer-readable storage medium is used to store instructions that can be executed by the processor, and when the instructions are executed by the processor, the processor is caused to:

Acquiring a video image from an imaging device and a parameter that characterizes the scene complexity of the video image;

Selecting a tracking algorithm suitable for the video image from the multiple tracking algorithms based on the parameter;

Determining a tracking target in the video image according to a tracking algorithm applicable to the video image;

Extracting position information of the tracking target in the video image, and calculating the distance between the tracking target and the center of the video image based on the position information;

The shooting angle of the imaging device is adjusted according to the distance, so that the tracking target is at the center of the next video image captured by the imaging device.
The device according to claim 6, wherein:

An imaging core is integrated in the imaging device;

When the processor acquires a video image and a parameter representing the complexity of the scene of the video image from an imaging device, it includes:

Receive a video image taken by the imaging core sent by an imaging device, and receive a sharpness evaluation parameter of the video image sent by the imaging device, wherein the sharpness evaluation parameter is a scene characterizing the video image Complexity parameter.
The device according to claim 6, wherein:

When the processor selects a tracking algorithm suitable for the video image from the multiple tracking algorithms based on the parameter, the method includes:

Compare the parameter value range corresponding to each preset tracking algorithm with the parameter;

The tracking algorithm corresponding to the parameter value range to which the parameter belongs is determined as the tracking algorithm suitable for the video image.
The device according to claim 6, wherein:

The multiple tracking algorithms include: contrast tracking algorithm, binary tracking algorithm, and related tracking algorithm;

When the target tracking algorithm applicable to the video image is a contrast tracking algorithm or a binary tracking algorithm, when the processor determines the tracking target in the video image according to the tracking algorithm applicable to the video image, it includes:

For each detection target in the video image, extract tracking information corresponding to the detection target for the target tracking algorithm;

Screening out the tracking target from all the detection targets according to the tracking information of all the detection targets used in the target tracking algorithm;

When the target tracking algorithm is a related tracking algorithm, when the processor determines the tracking target in the video image according to the tracking algorithm applicable to the video image, the method includes:

According to the pre-selected template image containing the tracking target, the detection target with the greatest degree of matching with the template image is selected from all the detection targets of the video image, and the screened detection target is determined as the video image. Track the target.
The device according to any one of claims 6-9, characterized in that:

When the processor adjusts the shooting angle of the imaging device according to the distance, it includes: sending the distance to a follow-up control device equipped with the video tracking device, so that the follow-up control device is Perform follow-up control operations from a distance.
A tracking device for a video tracking device, wherein the video tracking device supports multiple tracking algorithms; the device includes:

An information acquisition module for acquiring a video image from an imaging device and parameters that characterize the scene complexity of the video image;

An algorithm selection module, configured to select a tracking algorithm suitable for the video image from the multiple tracking algorithms based on the parameter;

A target determination module, configured to determine a tracking target in the video image according to a tracking algorithm applicable to the video image;

A distance calculation module, configured to extract position information of the tracking target in the video image, and calculate the distance between the tracking target and the center of the video image based on the position information;

The angle adjustment module is configured to adjust the shooting angle of the imaging device according to the distance, so that the tracking target is at the center of the next video image shot by the imaging device.
The apparatus according to claim 11, wherein an imaging core is integrated in the imaging device;

The information acquisition module is specifically used for:

Receive a video image taken by the imaging core sent by an imaging device, and receive a sharpness evaluation parameter of the video image sent by the imaging device, wherein the sharpness evaluation parameter is a scene characterizing the video image Complexity parameter.
The device according to claim 11, wherein the algorithm selection module is specifically configured to:

Compare the parameter value range corresponding to each preset tracking algorithm with the parameter;

The tracking algorithm corresponding to the parameter value range to which the parameter belongs is determined as the tracking algorithm suitable for the video image.
The device according to claim 11, wherein the multiple tracking algorithms include: a contrast tracking algorithm, a binary tracking algorithm, and a correlation tracking algorithm;

When the target tracking algorithm suitable for the video image is a contrast tracking algorithm or a binary tracking algorithm, the target determination module is specifically used for:

For each detection target in the video image, extract tracking information corresponding to the detection target for the target tracking algorithm;

Screening out the tracking target from all the detection targets according to the tracking information of all the detection targets used in the target tracking algorithm;

When the target tracking algorithm is a related tracking algorithm, the target determination module is specifically used for:

According to the pre-selected template image containing the tracking target, the detection target with the greatest degree of matching with the template image is selected from all detection targets in the video image, and the selected detection target is determined as the video image Tracking target.
The device according to any one of claims 11-14, wherein the angle adjustment module is specifically configured to:

The distance is sent to a follow-up control device equipped with the video tracking device, so that the follow-up control device performs a follow-up control operation according to the distance.
A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method steps according to any one of claims 1-5 are realized.