CN114723781B - Target tracking method and system based on camera array - Google Patents
Target tracking method and system based on camera array Download PDFInfo
- Publication number
- CN114723781B CN114723781B CN202210215587.2A CN202210215587A CN114723781B CN 114723781 B CN114723781 B CN 114723781B CN 202210215587 A CN202210215587 A CN 202210215587A CN 114723781 B CN114723781 B CN 114723781B
- Authority
- CN
- China
- Prior art keywords
- computing
- camera
- camera shooting
- target tracking
- resolution
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Studio Devices (AREA)
Abstract
The invention provides a target tracking method and a target tracking system based on a camera array, and belongs to the technical field of target tracking. The system includes a local edge computing terminal and a plurality of computing sticks. The local edge computing terminal receives the image data, performs local edge computing processing and sends the image data to the remote control equipment; the remote control equipment activates the computing bar based on the local edge computing processing result; and the computing bar receives the image data corresponding to the local edge computing processing result and executes target tracking processing on the image data. Judging whether a current image acquired by current camera shooting acquisition equipment contains a target of a preset type; if yes, the current camera shooting acquisition equipment is closed, and the other camera shooting acquisition equipment is started to activate the first neural network computing stick at the same time. The invention realizes target tracking processing by combining local edge computing equipment and a computing rod, reduces hardware starting cost and ensures the adaptability of a tracking model while ensuring the real-time performance of target tracking.
Description
Technical Field
The invention belongs to the technical field of target tracking, and particularly relates to a target tracking method and system based on a camera array, a computer system for realizing the method and a storage medium.
Background
Target tracking is an important problem in the field of computer vision, and is widely applied to the fields of sports event rebroadcasting, security monitoring, unmanned aerial vehicles, unmanned vehicles, robots and the like at present. The target tracking may be classified into single target tracking, multi-target tracking, dynamic target tracking, and static target tracking.
In order to realize accurate target tracking, data acquired by a single camera device is usually not complete enough, so that a camera array formed by a plurality of camera devices is adopted for tracking, and especially when a plurality of different types of targets to be tracked exist in a tracking field of view, a plurality of cameras with different angles started by the camera array can capture images with different frame rates and different resolutions at the same time, so that the accuracy of target tracking is facilitated.
However, in the prior art, to achieve the accuracy, a plurality of camera devices are usually required to be started in real time, which increases hardware energy consumption and arrangement cost; meanwhile, since there are many possible target tracking objects, the tracking system needs a complex target tracking model to adapt to the number and type variations of the tracked targets. When the amount of image data generated by the camera array is large, the existing tracking system has a large delay, and the accuracy caused by the reduced adaptability of the single model is also greatly reduced.
Disclosure of Invention
In order to solve the above technical problems, the present invention provides a target tracking method and system based on a camera array, a computer system implementing the method, and a storage medium.
In a first aspect of the present invention, a camera array based object tracking system is proposed,
the target tracking system comprises a local edge computing terminal and a plurality of computing rods;
the camera array comprises a plurality of camera shooting acquisition devices with different resolutions, the camera shooting acquisition devices are in wireless communication with a remote control device, and the remote control device controls the camera shooting acquisition devices to be turned on or turned off through instructions;
the local edge computing terminal receives the image data acquired by the camera array to perform local edge computing processing, and sends a local edge computing processing result to the remote control equipment;
the plurality of computing sticks may communicate in parallel with the remote control device;
the remote control device activates at least one computing stick based on the local edge computing processing result;
and the activated at least one computing bar receives the image data corresponding to the local edge computing processing result and executes target tracking processing on the image data.
As a further improvement, the local edge computing terminal receives the image data acquired by the camera array to perform local edge computing processing, and sends a local edge computing processing result to the remote control device, specifically including:
the local edge computing terminal receives a first image acquired by first camera shooting acquisition equipment with a first resolution of the camera array, and identifies whether the first image contains a first preset type of target to be tracked;
and if the first image contains a target to be tracked of a first preset type, sending the identification number of the first camera shooting and collecting equipment and the first resolution to the remote control equipment.
As another improvement, the image pickup and acquisition device is in wireless communication with a remote control device, and the remote control device controls the image pickup and acquisition device to be turned on or turned off through an instruction, specifically including:
if a first image acquired by a first camera shooting acquisition device with a first resolution of the camera array, which is received by the local edge computing terminal, contains a first preset type of target to be tracked, and the first resolution is not the highest resolution of the camera shooting acquisition devices contained in the camera array, the remote control device turns off the first camera shooting acquisition device with the first resolution through an instruction, and simultaneously turns on a second camera shooting acquisition device;
the second resolution of the second camera capture device is greater than the first resolution and the second camera capture device is contiguous with the first camera capture device.
In the above technical solution of the present invention, the activating, by the remote control device, at least one computing stick based on the local edge computing processing result specifically includes:
the remote control device activates a first computing bar based on the local edge computing processing result;
the first computing wand configures a first target tracking model;
the first target tracking model may perform a target tracking process for a first predetermined type of target to be tracked.
The activated at least one computing rod receives the image data corresponding to the local edge computing processing result, and performs target tracking processing on the image data, specifically including:
the first computing bar receives a first image acquired by the first camera shooting and acquiring equipment and a second image acquired by the second camera shooting and acquiring equipment, and executes target tracking processing;
the second camera shooting acquisition equipment is started after the first camera shooting acquisition equipment with the first resolution is closed by the remote control equipment through an instruction.
In a second aspect of the present invention, there is provided a camera array based target tracking method, the method comprising:
s810: judging whether a current image acquired by current camera acquisition equipment contains a target of a preset type;
if so, closing the current camera shooting acquisition equipment, starting another camera shooting acquisition equipment, and taking the other camera shooting acquisition equipment as the current camera shooting acquisition equipment;
s820: activating at least one first neural network computing stick, the first neural network computing stick configuring a first target tracking model;
the first target tracking model may perform a target tracking process for the predetermined type of target to be tracked;
and the resolution ratio of the other camera shooting equipment is greater than that of the current camera shooting and collecting equipment.
Further, the method is implemented based on a camera array comprising a first camera acquisition device of a first resolution, a second camera acquisition device of a second resolution and a third camera acquisition device of a third resolution, the second resolution being greater than the first resolution and less than the third resolution;
and if the current camera shooting and collecting equipment is third camera shooting and collecting equipment and the current image collected by the current camera shooting and collecting equipment contains a target of a preset type, activating a plurality of first neural network computing sticks while keeping the current equipment in an opening state.
In a third aspect of the present invention, there is provided a computer system, the computer system includes a display device, the display device includes a USB interface and/or an HDMI interface, the USB interface and/or the HDMI interface can be connected to a plurality of computing sticks, and the computer system activates at least one computing stick through an instruction, so as to implement the camera array-based object tracking method according to the second aspect.
In a fourth aspect of the present invention, there is provided a camera array based object tracking device, the device comprising a processor and a memory, the memory having stored thereon computer executable program instructions, the executable program instructions being executable by the processor for implementing the method of the second aspect.
Further, in a fifth aspect of the present invention, the present invention may be embodied as a computer-readable storage medium having stored thereon computer program instructions for executing the method of the first aspect.
Similarly, in a sixth aspect of the present invention, the present invention can also be embodied as a computer program product, which is loaded into a computer storage medium and executed by a processor, thereby implementing the method of the first aspect.
The technical scheme of the invention realizes target tracking processing by combining local edge computing equipment and a computing stick, wherein the computing stick is a neural network computing stick, and different neural network computing sticks are configured with different types of neural network inference models; the real-time performance of target tracking is ensured, and meanwhile, the hardware starting cost is reduced, and the adaptability of a tracking model is ensured.
Further advantages of the invention will be apparent from the detailed description of embodiments which follows, when considered in conjunction with the accompanying drawings.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic diagram of a partial block architecture of a camera array based object tracking system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a portion of the data acquisition and transmission of the camera array based object tracking system of FIG. 1;
FIG. 3 is a schematic diagram of a portion of the operation of the camera array based target tracking system of FIG. 1;
FIG. 4 is a flowchart illustrating a target tracking method based on a camera array according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a computer-readable storage medium implementing the method of FIG. 4;
FIG. 6 is a block diagram of a computer system that implements the method of FIG. 4.
Detailed Description
The invention is further described with reference to the following drawings and detailed description.
Referring to fig. 1, fig. 1 is a schematic diagram of a partial structural module of a target tracking system based on a camera array according to an embodiment of the present invention.
In fig. 1, the target tracking system based on camera array according to the present embodiment is shown to include a local edge computing terminal, a plurality of computing sticks, and a remote control device.
The plurality of computing sticks may communicate in parallel with the remote control device;
specifically, the remote control device is provided with a USB interface and/or an HDMI interface, and the USB interface and/or the HDMI interface can be connected to a plurality of computing sticks simultaneously.
It should be noted that although the USB interface and/or the HDMI interface may connect a plurality of computing sticks at the same time, in the technical solution of the present invention, it is not required to call all the computing sticks that have been connected at the same time, but activate the computing stick that meets the requirement according to the actual call requirement.
It should be noted that the introduction of a computing stick in the field of object tracking is one of the outstanding improvements of the present invention, and a brief description of the computing stick used in the various embodiments of the present invention follows.
In popular terms, the computing stick looks like a U disk, is internally provided with advanced chips such as a memory, an arithmetic unit, a processor and the like, and can be regarded as a microcomputer. The computer is a complete computer only by inserting the computer on a display with an HDMI interface and connecting the computer with peripherals such as a power supply and the like.
But the computing stick does not replace a full computer because its processing power is not as powerful as a full computer.
Taking a neural network computing stick as an example, the computing stick can configure a trained neural network to directly perform inference computation, such as target tracking, but cannot perform adaptive training or updating of the neural network by itself.
With a computing stick, low power consumption is a major advantage. The host processor is released on the premise of not increasing power consumption, and deep learning capability is endowed.
In particular embodiments of the present invention, the plurality of computing sticks may be neural network computing sticks; different neural network computing sticks are configured with different types of neural network inference models.
The different types of the neural network reasoning model can be that the first computing stick is configured with a neural network reasoning model for back propagation, and the second computing stick is configured with a neural network reasoning model for deep learning;
in another aspect, as a further preference, the plurality of computing sticks are neural network computing sticks that configure a target tracking model; different types of targets that can be tracked by target tracking models of different neural network computing rod configurations are different.
As a specific example, the first computing stick may be configured with a neural network inference model that is a first target tracking model of a type of animal, and the second computing stick may be configured with a neural network inference model that is a second target tracking model of a type of animal.
Of course, it will be understood by those skilled in the art that the types of objects to be tracked are not limited to people and animals, but may include other categories, such as static object tracking models, dynamic object tracking models, single object tracking models, multi-object tracking models, or hybrid models.
Based on the architecture of fig. 1, the technical solution of the present invention can be summarized as follows:
the local edge computing terminal receives the image data acquired by the camera array to perform local edge computing processing, and sends a local edge computing processing result to the remote control equipment;
the plurality of computing sticks may communicate in parallel with the remote control device;
the remote control device activates at least one computing stick based on the local edge computing processing result;
and the activated at least one computing bar receives the image data corresponding to the local edge computing processing result and executes target tracking processing on the image data.
The local edge computing terminal, which may be referred to as an edge computing terminal for short in the present invention, is a generic name of a device terminal adapted to perform edge computing.
Edge computing means that an open platform integrating network, computing, storage and application core capabilities is adopted on one side close to an object or a data source to provide nearest-end services nearby. The application program is initiated at the edge side, so that a faster network service response is generated, and the basic requirements of the industry in the aspects of real-time business, application intelligence, safety, privacy protection and the like are met. The edge computation is between the physical entity and the industrial connection, or on top of the physical entity. And the cloud computing still can access the historical data of the edge computing.
It can be seen that the processing power of the edge computing device is low relative to the cloud, but the advantage is in the real-time nature of the response.
Specifically, in the embodiments of the present invention, the edge computing terminal configures a fast target type identification model, which can quickly identify whether the input image frame or video frame includes the target to be tracked and the type, number, etc. of the target to be tracked, but does not need to perform a complicated target tracking process. Thus, using a local edge computing terminal can respond quickly and make simple decisions.
In particular, see fig. 2 on the basis of fig. 1. FIG. 2 is a schematic diagram of a portion of the device data acquisition and data transmission of the camera array based object tracking system of FIG. 1.
The local edge computing terminal receives the image data acquired by the camera array to perform local edge computing processing, and sends a local edge computing processing result to the remote control equipment;
the local edge calculation result includes a result of identifying whether the image data acquired by the camera array includes the target to be tracked, the type of the target to be tracked, the number of the targets to be tracked, the static target to be tracked, the dynamic target to be tracked, and the like.
The local edge computing terminal receives a first image acquired by first camera shooting acquisition equipment with a first resolution of the camera array, and identifies whether the first image contains a first preset type of target to be tracked;
and if the first image contains a target to be tracked of a first preset type, sending the identification number of the first camera shooting and collecting equipment and the first resolution to the remote control equipment.
In fig. 3, the image capture and acquisition device is in wireless communication with a remote control device, and the remote control device controls the image capture and acquisition device to be turned on or off through an instruction, which specifically includes:
if a first image acquired by a first camera shooting acquisition device with a first resolution of the camera array, which is received by the local edge computing terminal, contains a first preset type of target to be tracked, and the first resolution is not the highest resolution of the camera shooting acquisition devices contained in the camera array, the remote control device turns off the first camera shooting acquisition device with the first resolution through an instruction, and simultaneously turns on a second camera shooting acquisition device;
the second resolution of the second camera capture device is greater than the first resolution and the second camera capture device is contiguous with the first camera capture device.
The remote control device activates at least one computing stick based on the local edge computing processing result, and specifically includes:
the remote control device activates a first computing bar based on the local edge computing processing result;
the first computing bar configuring a first target tracking model;
the first target tracking model may perform a target tracking process for a first predetermined type of target to be tracked.
As an example, if the image data acquired by the camera array is displayed by the local edge calculation processing result to include that the type of the target to be tracked is a person, activating a first computing bar configured with a first person tracking model;
as an example, if the local edge calculation processing result shows that the image data acquired by the camera array includes the type of the target to be tracked as a character and includes a first number of multiple characters, a second computing stick configured with a second multiple-target character tracking model is activated, or a first number of multiple third computing sticks configured with a third character tracking model are activated.
In the above embodiment, the activated at least one computing stick receives image data corresponding to the local edge computing processing result, and performs target tracking processing on the image data, specifically including:
the first computing stick receives a first image acquired by the first camera acquisition equipment and a second image acquired by the second camera acquisition equipment, and executes target tracking processing;
the second camera shooting acquisition equipment is started after the first camera shooting acquisition equipment with the first resolution is closed by the remote control equipment through an instruction.
Obviously, according to the improvement, the low-resolution camera shooting acquisition equipment is turned off, and the high-resolution camera shooting acquisition equipment is turned on, so that unnecessary repetition of data and repeated turning-on of hardware are avoided while the continuity of target tracking is ensured.
On the basis of the hardware and software architectures of fig. 1-3, referring to fig. 4, fig. 4 is a flowchart illustrating a target tracking method based on a camera array according to an embodiment of the present invention, and it can be understood that the method illustrated in fig. 4 can be implemented based on the hardware architecture or principle of fig. 1-3.
In fig. 4, the method comprises the steps of:
s810: judging whether a current image acquired by current camera acquisition equipment contains a target of a preset type;
if so, closing the current camera shooting acquisition equipment, starting another camera shooting acquisition equipment, and taking the other camera shooting acquisition equipment as the current camera shooting acquisition equipment;
s820: activating at least one first neural network computing stick, the first neural network computing stick configuring a first target tracking model;
the first target tracking model may perform target tracking processing for the predetermined type of target to be tracked;
and the resolution of the other camera shooting equipment is greater than that of the current camera shooting acquisition equipment.
As a further preference, in the step S810,
if the current image acquired by the current camera shooting acquisition equipment contains a target of a preset type and the current camera shooting acquisition equipment is the camera shooting acquisition equipment with the highest resolution in the camera array, keeping the starting state of the current camera shooting acquisition equipment unchanged;
otherwise, closing the current camera shooting acquisition equipment, starting another camera shooting acquisition equipment with higher resolution, and taking the other camera shooting acquisition equipment as the current camera shooting acquisition equipment;
if the current image acquired by the current camera shooting acquisition equipment does not contain the target of the preset type and the current camera shooting acquisition equipment is not the camera shooting acquisition equipment with the lowest resolution in the camera array, closing the current camera shooting acquisition equipment, starting another camera shooting acquisition equipment with the lower resolution and taking the other camera shooting acquisition equipment as the current camera shooting acquisition equipment;
and if the current image acquired by the current camera shooting acquisition equipment does not contain the target of the preset type and the current camera shooting acquisition equipment is the camera shooting acquisition equipment with the lowest resolution in the camera array, keeping the opening state of the current camera shooting acquisition equipment unchanged.
As a further example, a camera array including three image capture apparatuses is taken as an example.
The method is implemented based on a camera array comprising a first camera capture device of a first resolution, a second camera capture device of a second resolution, and a third camera capture device of a third resolution, the second resolution being greater than the first resolution and less than the third resolution;
and if the current camera shooting and collecting equipment is third camera shooting and collecting equipment and the current image collected by the current camera shooting and collecting equipment contains a target of a preset type, activating a plurality of first neural network computing sticks while keeping the current equipment in an opening state.
For example, if a current image acquired by the current camera acquisition equipment contains a target of a person type, a first computing stick configured with a first person tracking model is activated;
for example, if the current image captured by the current image capture device includes a target to be tracked of a type of person and a first number of multiple persons, the second computing stick configured with the second multi-target person tracking model is activated, or the first number of multiple third computing sticks configured with the third person tracking model is activated.
The technical scheme of the invention can be automatically realized by computer equipment based on computer program instructions. Similarly, the present invention can also be embodied as a computer program product, which is loaded on a computer storage medium and executed by a processor to implement the above technical solution.
Further embodiments therefore include a computer device comprising a memory storing a computer executable program and a processor configured to perform the steps of the above method.
Specifically, referring to fig. 6, a computer system is provided, where the computer system includes a display device, the display device includes a USB interface and/or an HDMI interface, the USB interface and/or the HDMI interface can be connected to a plurality of computing sticks, and the computer system activates at least one computing stick through an instruction, so as to implement the camera array based object tracking method described in fig. 3 or fig. 4.
The target tracking processing is realized by combining local edge computing equipment and a computing bar, the computing bar is a neural network computing bar, and different neural network computing bars are configured with different types of neural network inference models; the technical scheme of the invention can reduce the hardware starting cost and ensure the adaptability of the tracking model while ensuring the real-time performance of target tracking.
It should be noted that the present invention can solve a plurality of technical problems or achieve corresponding technical effects, but does not require that each embodiment of the present invention solves all the technical problems or achieves all the technical effects, and an embodiment that separately solves one or several technical problems or achieves one or more improved effects also constitutes a separate technical solution.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
In the present invention, the contents of the module structure or the technical terms not specifically defined are subject to the description of the prior art. The prior art mentioned in the background section can be used as part of the invention to understand the meaning of some technical features or parameters. The scope of the present invention is defined by the claims.
Claims (8)
1. A target tracking system based on a camera array comprises a plurality of camera shooting acquisition devices with different resolutions, wherein the camera shooting acquisition devices are in wireless communication with a remote control device, and the remote control device controls the camera shooting acquisition devices to be turned on or turned off through instructions;
the method is characterized in that:
the target tracking system also comprises a local edge computing terminal and a plurality of computing rods;
the local edge computing terminal receives the image data acquired by the camera array to perform local edge computing processing, and sends a local edge computing processing result to the remote control equipment;
the plurality of computing sticks in parallel communication with the remote control device;
the remote control device activates at least one computing stick based on the local edge computing processing result;
the activated at least one computing rod receives image data corresponding to the local edge computing processing result and executes target tracking processing on the image data;
the local edge computing terminal receives the image data acquired by the camera array to perform local edge computing processing, and sends a local edge computing processing result to the remote control device, and the method specifically includes:
the local edge computing terminal receives a first image acquired by first camera shooting acquisition equipment with a first resolution of the camera array, and identifies whether the first image contains a first preset type of target to be tracked;
if the first image contains a target to be tracked of a first preset type, sending the identification number of the first camera shooting and collecting device and the first resolution to the remote control device;
the remote control device activates at least one computing stick based on the local edge computing processing result, and specifically includes:
the remote control device activates a first computing bar based on the local edge computing processing result; the first computing bar configuring a first target tracking model;
the first target tracking model performs target tracking processing for a first predetermined type of target to be tracked.
2. A camera array based object tracking system as claimed in claim 1, wherein:
the plurality of computing sticks are neural network computing sticks;
different neural network computing sticks are configured with different types of neural network inference models.
3. A camera array based object tracking system as claimed in claim 1, wherein:
the plurality of computing rods are neural network computing rods for configuring a target tracking model;
the target tracking models of different neural network computing rod configurations track different types of targets.
4. A camera array based object tracking system as claimed in claim 1, wherein:
the camera shooting and collecting equipment is in wireless communication with the remote control equipment, the remote control equipment controls the camera shooting and collecting equipment to be opened or closed through an instruction, and the method specifically comprises the following steps:
if a first image acquired by a first camera shooting acquisition device with a first resolution of the camera array, which is received by the local edge computing terminal, contains a first preset type of target to be tracked, and the first resolution is not the highest resolution of the camera shooting acquisition devices contained in the camera array, the remote control device turns off the first camera shooting acquisition device with the first resolution through an instruction, and simultaneously turns on a second camera shooting acquisition device;
the second resolution of the second camera capture device is greater than the first resolution and the second camera capture device is contiguous with the first camera capture device.
5. A camera array based object tracking system as claimed in claim 1, wherein:
the activated at least one computing rod receives the image data corresponding to the local edge computing processing result, and performs target tracking processing on the image data, specifically including:
the first computing stick receives a first image acquired by the first camera acquisition equipment and a second image acquired by the second camera acquisition equipment, and executes target tracking processing;
the second camera shooting acquisition equipment is started after the first camera shooting acquisition equipment with the first resolution is closed by the remote control equipment through an instruction.
6. A target tracking method based on camera array, which is implemented based on a target tracking system based on camera array as claimed in any one of claims 1-5;
characterized in that the method comprises:
s810: judging whether a current image acquired by current camera acquisition equipment contains a target of a preset type;
if so, closing the current camera shooting acquisition equipment, starting another camera shooting acquisition equipment, and taking the other camera shooting acquisition equipment as the current camera shooting acquisition equipment;
s820: activating at least one first neural network computing stick, the first neural network computing stick configuring a first target tracking model;
the first target tracking model executes target tracking processing aiming at the target to be tracked of the preset type;
and the resolution ratio of the other camera shooting acquisition equipment is greater than that of the current camera shooting acquisition equipment.
7. A camera array based target tracking method as claimed in claim 6, wherein:
the method is implemented based on a camera array comprising a first camera capture device of a first resolution, a second camera capture device of a second resolution and a third camera capture device of a third resolution, the second resolution being greater than the first resolution and less than the third resolution;
and if the current camera shooting and collecting equipment is third camera shooting and collecting equipment and the current image collected by the current camera shooting and collecting equipment contains a target of a preset type, activating a plurality of first neural network computing sticks while keeping the current equipment in an opening state.
8. A computer system, the computer system comprising a display device, the display device comprising a USB interface and/or an HDMI interface, the USB interface and/or the HDMI interface being connected to a plurality of computing sticks, the computer system activating at least one computing stick by an instruction for implementing a camera array based object tracking method according to any one of claims 6 or 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210215587.2A CN114723781B (en) | 2022-03-07 | 2022-03-07 | Target tracking method and system based on camera array |
| PCT/CN2022/141927 WO2023169053A1 (en) | 2022-03-07 | 2022-12-26 | Target tracking method and system based on camera array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210215587.2A CN114723781B (en) | 2022-03-07 | 2022-03-07 | Target tracking method and system based on camera array |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114723781A CN114723781A (en) | 2022-07-08 |
| CN114723781B true CN114723781B (en) | 2022-10-28 |
Family
ID=82237356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210215587.2A Active CN114723781B (en) | 2022-03-07 | 2022-03-07 | Target tracking method and system based on camera array |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114723781B (en) |
| WO (1) | WO2023169053A1 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114723781B (en) * | 2022-03-07 | 2022-10-28 | 北京拙河科技有限公司 | Target tracking method and system based on camera array |
| CN115361499B (en) * | 2022-08-16 | 2024-03-12 | 小视科技(江苏)股份有限公司 | Dual-machine cooperative border defense target recognition and tracking system and method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909648A (en) * | 2019-11-15 | 2020-03-24 | 华东师范大学 | People flow monitoring method implemented on edge computing equipment by using neural network |
| CN112532870A (en) * | 2020-11-09 | 2021-03-19 | 福建福瑞康信息技术有限公司 | Multi-terminal self-adaptive data processing method and system |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11017539B2 (en) * | 2018-04-30 | 2021-05-25 | Intel Corporation | Object tracking and identification using intelligent camera orchestration |
| TWI768735B (en) * | 2019-02-01 | 2022-06-21 | 國立中央大學 | Identification method of moveable apparatus, identification and parking detection method of moveable apparatus and identification and parking detection system of moveable apparatus |
| CN110161932A (en) * | 2019-06-11 | 2019-08-23 | 浙江嘉科电子有限公司 | A kind of intelligent vehicle mounted terminal based on edge calculations and multisensor interface |
| CN110705395A (en) * | 2019-09-18 | 2020-01-17 | 北京邮电大学 | Detection system based on raspberry group |
| CN111259461A (en) * | 2020-01-20 | 2020-06-09 | 北京科芮智盈技术有限公司 | Anti-stealing-lighting method, terminal equipment and system |
| AU2021305674A1 (en) * | 2020-07-10 | 2023-03-02 | TerraClear Inc. | Intelligent multi-visual camera system and method |
| US20210081271A1 (en) * | 2020-09-25 | 2021-03-18 | Intel Corporation | Dynamic tracing control |
| US20220004351A1 (en) * | 2020-09-26 | 2022-01-06 | Intel Corporation | Hardware architecture for multi-display video synchronization |
| CN112419401A (en) * | 2020-11-23 | 2021-02-26 | 上海交通大学 | Aircraft surface defect detection system based on cloud edge cooperation and deep learning |
| US20210117697A1 (en) * | 2020-12-26 | 2021-04-22 | Francesc Guim Bernat | Edge automatic and adaptive processing activations |
| CN113327219B (en) * | 2021-06-21 | 2022-01-28 | 易成功(厦门)信息科技有限公司 | Image processing method and system based on multi-source data fusion |
| CN113837222A (en) * | 2021-08-23 | 2021-12-24 | 南京信息职业技术学院 | Cloud-edge cooperative machine learning deployment application method and device for millimeter wave radar intersection traffic monitoring system |
| CN114723781B (en) * | 2022-03-07 | 2022-10-28 | 北京拙河科技有限公司 | Target tracking method and system based on camera array |
-
2022
- 2022-03-07 CN CN202210215587.2A patent/CN114723781B/en active Active
- 2022-12-26 WO PCT/CN2022/141927 patent/WO2023169053A1/en unknown
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909648A (en) * | 2019-11-15 | 2020-03-24 | 华东师范大学 | People flow monitoring method implemented on edge computing equipment by using neural network |
| CN112532870A (en) * | 2020-11-09 | 2021-03-19 | 福建福瑞康信息技术有限公司 | Multi-terminal self-adaptive data processing method and system |
Non-Patent Citations (2)
| Title |
|---|
| "Privacy-Aware Edge Computing System For People Tracking";J. Yrjanainen等;《2020 IEEE International Conference on Image Processing (ICIP)》;20200930;全文 * |
| "无人机实时图像应用***的计算模式研究";窦晓磊等;《小型微型计算机***》;20210630;第42卷(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023169053A1 (en) | 2023-09-14 |
| CN114723781A (en) | 2022-07-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114723781B (en) | Target tracking method and system based on camera array | |
| EP3471021B1 (en) | Method for determining target intelligently followed by unmanned aerial vehicle, unmanned aerial vehicle and remote controller | |
| US10728570B2 (en) | Apparatus and methods for real time estimation of differential motion in live video | |
| WO2021043112A1 (en) | Image classification method and apparatus | |
| US20190057244A1 (en) | Method for determining target through intelligent following of unmanned aerial vehicle, unmanned aerial vehicle and remote control | |
| US20190304102A1 (en) | Memory efficient blob based object classification in video analytics | |
| US8977582B2 (en) | Spiking neuron network sensory processing apparatus and methods | |
| US9070039B2 (en) | Temporal winner takes all spiking neuron network sensory processing apparatus and methods | |
| US20140156574A1 (en) | Rate stabilization through plasticity in spiking neuron network | |
| US20140365414A1 (en) | Spiking network apparatus and method with bimodal spike-timing dependent plasticity | |
| US10818016B2 (en) | Systems and methods for predictive/reconstructive visual object tracker | |
| Hoermann et al. | Object detection on dynamic occupancy grid maps using deep learning and automatic label generation | |
| US20210382542A1 (en) | Screen wakeup method and apparatus | |
| KR20200118076A (en) | Biometric detection method and device, electronic device and storage medium | |
| US11461992B2 (en) | Region of interest selection for object detection | |
| Martinez-Alpiste et al. | Benchmarking machine-learning-based object detection on a UAV and mobile platform | |
| CN113065645A (en) | Twin attention network, image processing method and device | |
| CN113111782A (en) | Video monitoring method and device based on salient object detection | |
| CN109963072B (en) | Focusing method, focusing device, storage medium and electronic equipment | |
| CN112651267A (en) | Recognition method, model training, system and equipment | |
| CN109714531A (en) | A kind of image processing method, device and unmanned plane | |
| CN115298705A (en) | License plate recognition method and device, electronic equipment and storage medium | |
| CN114979497A (en) | Unmanned aerial vehicle linkage tracking method and system based on pole loading and cloud platform | |
| CN115147818A (en) | Method and device for identifying mobile phone playing behaviors | |
| WO2021189321A1 (en) | Image processing method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |