Method and device for portable multi-terminal coordinated combat in wild areas

Abstract

The present invention discloses a method and device for portable multi-terminal collaborative operations in wild areas, including a sky end, including a flying platform, the flying platform is equipped with a collection unit and an airborne communication unit, the collection unit is used to collect the required images to form a data stream, and the data stream is shared with the ground end through the airborne communication unit; the ground end is networked with the airborne communication unit through the ground communication unit to form a communication link, and the communication link is used to send control instructions to the sky end to control the flying platform and / or the collection unit, and obtain the status information of the sky end and the collected data stream. Beneficial effect: The innovative use of a combination of micro-UAV + AI edge computing + back-end management platform can detect and identify targets while the UAV is sailing and shooting. Compared with the previous server background AI recognition method, it can provide faster recognition speed for the combat team, help the special operations team and the rear command center quickly grasp the battlefield situation, and improve the efficiency of battlefield reconnaissance.

Description

Technical Field

[0001] The present invention belongs to the technical field of information collection and processing, and in particular, relates to a method and device for portable multi-terminal collaborative operations in wild areas. Background Art

[0002] Compared with traditional warfare, modern warfare places more emphasis on informatization and networking, and places higher and higher requirements on combat commanders and soldiers. In field operations, soldiers need to obtain and share battlefield information, instructions, and tasks in a timely manner in order to better complete combat missions. At the same time, due to the particularity of the field environment, traditional communication equipment and combat equipment are not portable and flexible enough to meet the needs of field operations. Therefore, methods and devices for portable multi-terminal collaborative operations in field areas have emerged, aiming to improve the efficiency and coordination of field operations, so that soldiers can better conduct multi-terminal collaborative operations in field environments. Summary of the invention

[0003] In view of this, the present invention aims to propose a method and device for portable multi-terminal collaborative operations in wild areas to build target detection and recognition for multi-source images, improve the timeliness of information extraction and the accuracy of target recognition. In order to achieve this goal, the scheme builds a drone detection and recognition system.

[0004] To achieve the above object, the technical solution of the present invention is achieved as follows:

[0005] A portable multi-terminal collaborative combat device in the wild area, including:

[0006] The sky side includes a flight platform, which is equipped with a collection unit and an airborne communication unit. The collection unit is used to collect the required images to form a data stream, and the data stream is shared with the ground side through the airborne communication unit. The shared devices include a portable control terminal and an Android phone. The shared data includes: command data, status data, and video data;

[0007] The ground side forms a communication link by networking with the airborne communication unit through the ground communication unit, and uses the communication link to send control instructions to the sky side to control the flight platform and / or the collection unit, and obtain the status information of the sky side and the collected data stream.

[0008] Furthermore, the flight platform is equipped with an onboard lightweight processor, the acquisition unit includes a zoom wide-angle camera, and the onboard communication unit includes an ad hoc network radio.

[0009] Furthermore, the ground end includes a self-organizing network radio station and a human-machine interaction terminal, and there is a wired / wireless connection between the self-organizing network radio station and the human-machine interaction terminal;

[0010] The ground end connects and communicates with the onboard computer on the flight platform through the software installed on the human-computer interaction terminal to send control instructions to control the movement of the flight platform or the gimbal and camera of the load.

[0011] Furthermore, the obtaining of the status information of the sky terminal includes receiving the flight platform status data and the sensor data of the payload gimbal and the camera, and displaying the obtained status information on the screen of the human-computer interaction terminal;

[0012] The obtaining of the data stream collected by the sky end includes intercepting the collected information required for real-time detection and identification of the raw video stream in the data stream.

[0013] Furthermore, for the data stream acquired by the acquisition unit, data hard decoding and soft decoding are required, as follows:

[0014] The steps for streaming media transmission using RTSP are as follows: Step 1: Establish a network connection: Use TCP to establish a network connection with the streaming media server; Step 2: Handshake: The client (tablet) sends an RTSP request to the server to request to establish a session; Step 3: Get the session description: The tablet sends a "DESCRIBE" request to the server to request to obtain the session description information of the streaming media; Step 4: Establish a transmission channel: The client sends a "SETUP" request to the server to request to establish a transmission channel. This request specifies the transmission protocol (RTCP is used this time) and the transmission port number; Step 5: Start transmission: The client sends a "PLAY" request to the server to request to start transmitting streaming media data; Step 6: Control transmission: The client sends a "PAUSE" request to pause transmission; Step 7: Data transmission: After establishing the transmission channel, the server sends the streaming media data to the client in RTP (Real-time Transport Protocol) format. The client receives the RTP data and decodes and displays it.

[0015] Use FFMpeg for video capture and encoding;

[0016] Use the recognition terminal to obtain the pushed video stream data in real time and perform subsequent recognition and detection work.

[0017] Furthermore, the acquisition unit needs to perform coordinate calculation, including:

[0018] S1. Use the Yolo_V5 recognition algorithm to identify the target in the image. If the recognition is unsuccessful, the drone continues to fly to the waypoint; if the recognition is successful, according to the formula Get the pixel coordinates (X, Y) of the center point of the target object on the image, where X 1 , X 2 To identify the pixel coordinates of the target object in the X direction,1 , Y 2 To identify the pixel coordinates of the target object in the Y direction;

[0019] S2. Using formula Obtain the difference between the pixel coordinates and the pixel coordinates of the center point of the image , according to the formula , convert the pixel coordinate difference of the video stream resolution to the actual screen resolution , where X 中 is the X-coordinate of the center pixel of the entire image, Y 中 The Y coordinate of the center pixel of the entire image;

[0020] S3. According to the focal length of the camera , Wide , long distance , the UAV flight altitude, using the formula Solve the actual length of the screen and the actual width , where h is the flight altitude of the UAV;

[0021] S4. According to the formula Get the actual distance corresponding to the pixel coordinates ,in, and are the length and width of the screen respectively. is the actual length of the screen, The actual width of the screen.

[0022] S5. Function formula , calculate the angles that the camera needs to rotate in two directions, and then use the formula , calculate the angle, and control the camera according to The angle of rotation is θ1, which is the X-axis rotation angle and θ2, which is the Y-axis rotation angle.

[0023] S6. Calculate the length of each direction from the target point to the center point according to the pixel coordinates returned by the search target. The formula is: ;

[0024] S7. According to the coordinate inverse calculation formula, the longitude and latitude coordinates returned by the drone are known to solve the plane coordinates on the Gaussian plane. , and combined with the calculated , and the yaw angle of the drone relative to the north direction , and finally get the absolute coordinates of the static target point , whose formula is , and finally the Gaussian plane coordinates calculated , use the Gaussian coordinate calculation formula to get the final longitude and latitude of the target object .

[0025] Furthermore, the mobile phone application, the command and control center and the portable control terminal are combined into a three-end collaborative operation module, and real-time data synchronization and sharing can be achieved among the three-end collaborative operation modules.

[0026] Furthermore, the present invention discloses a method for a portable multi-terminal coordinated operation in a wild area, and a device for implementing a portable multi-terminal coordinated operation in a wild area, comprising:

[0027] The command center sends the task to the portable control terminal. After receiving the task instruction, the control terminal sends the instruction to the onboard lightweight processor through point-to-point radio communication, calls the PSDK control service, and issues and executes the task.

[0028] The control terminal interface displays the real-time location and map data of the drone by calling the GIS map service and data middleware on the onboard lightweight processor; the real-time video stream returned by the gimbal camera is viewed and broadcast live by calling the video streaming service; the suspicious targets in the video frame are quickly identified by calling the lightweight recognition and detection service and the results are given to the portable control terminal;

[0029] The control terminal displays the recognition result in the video frame in real time and distributes the video stream to the mobile phone;

[0030] Log in to the authorized app on the mobile phone to quickly view the identification results and provide navigation, guiding the operator to quickly approach the suspicious target for real-time tracking and attack.

[0031] Compared with the prior art, the method and device for field portable multi-terminal collaborative operations described in the present invention have the following beneficial effects:

[0032] (1) The method and device for portable multi-terminal collaborative combat in wild areas described in the present invention ensure the timeliness of information extraction: the system is guided by the concept of portable reconnaissance and innovatively adopts a combination of micro-UAV + AI edge computing + back-end management platform. It can detect and identify targets while the UAV is flying and shooting. Compared with the previous server background AI recognition method, it can provide faster recognition speed for the combat team, help the special operations team and the rear command center quickly grasp the battlefield situation, and improve the efficiency of battlefield reconnaissance;

[0033] (2) The method and device for portable multi-terminal collaborative operations in wild areas described in the present invention improve the accuracy of target recognition: At present, most drones have poor image matching effects under conventional conditions and complex shooting conditions, and local areas cannot be matched. The application of this algorithm can determine the main direction and size of key points and generate feature points. Then the improved yolo5 algorithm is used to describe the produced feature points. Finally, the improved algorithm with multiple constraints is used to obtain the optimal matching set, thereby improving the accuracy of target detection and recognition;

[0034] (3) The method and device for field portable multi-terminal collaborative operations described in the present invention improve the accuracy of target positioning: in the identification and reconnaissance mission, the drone is used for precise positioning, and the pixel-based coordinate conversion algorithm is used to accurately locate and identify the target, providing behavioral decision-making and action guidelines for the reconnaissance team;

[0035] (4) The method and device for portable multi-terminal collaborative operations in wild areas described in the present invention, by combining dynamic frame selection, adaptive resolution adjustment and spatiotemporal synchronization processing, not only improves the efficiency of image cutting, but also significantly reduces the distortion during image processing and ensures the synchronization of image cutting and AI recognition tasks, and has significant innovation and practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:

[0037] Figure 1 A schematic diagram of the hardware composition of the system according to an embodiment of the present invention;

[0038] Figure 2 The present invention is a schematic diagram of a logic flow of the detection and identification module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0040] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0041] This solution aims to realize the construction of target detection and recognition for multi-source images, improve the timeliness of information extraction and the accuracy of target recognition. In order to achieve this goal, this solution builds a drone detection and recognition system. The main structure is divided into two parts: the sky end and the ground end.

[0042] The main body of the sky end is a multi-rotor flight platform, which is equipped with a zoom wide-angle camera through a gimbal assembly, and integrates the onboard lightweight processor and ad hoc network radio. After being integrated into one, it is fixed to the upper part of the flight platform with a non-standard bracket and connected to the data interface and power interface of the flight platform.

[0043] The ground end mainly consists of a self-organizing radio station and a human-computer interaction terminal, which can be connected via wireless WIFI or network cable. The sky end and the ground end are networked through a self-organizing radio station to form a communication link. Through the software installed on the human-computer interaction terminal, it connects and communicates with the onboard computer on the flight platform, sends control instructions, controls the flight platform action or the gimbal and camera action of the load; quickly identifies and detects, intercepts the video raw stream to detect and identify the target vehicle on the ground end in real time, and receives the flight platform status data and the sensor data of the gimbal and camera of the load at the same time, and displays it on the screen of the human-computer interaction terminal. The data stream obtained from the sky end is divided into two paths, one for cutting and identifying on the onboard lightweight processor, and the other for directly pushing the video raw stream to the ground end, where:

[0044] In view of the problems that the existing technology may face in the process of image cutting, such as computing power limitation, image distortion, and real-time performance, this solution designs a more efficient innovative image cutting solution that is adapted to the low computing power of the airborne end and synchronized with the AI ​​recognition service. An innovative image cutting method based on dynamic frame selection and adaptive resolution adjustment is proposed, which mainly includes:

[0045] a. Dynamic frame selection strategy:

[0046] In order to reduce the burden on the onboard lightweight processor, a dynamic frame selection algorithm is used. This algorithm dynamically selects frames in the video stream based on the data of the flight platform and load sensors. Specifically, the algorithm analyzes the state of the platform (such as altitude, speed, pitch angle, etc.) and the movement of the camera (such as the pan / tilt rotation angle), identifies scenes that are more critical to the recognition task, and prioritizes cutting these frames. This method can reduce the processing of invalid frames, thereby improving processing efficiency and accuracy. Specifically:

[0047] ;

[0048] in,

[0049] The frame sequence that needs to be cut;

[0050] is the speed of the flying platform;

[0051] is the rotation angle of the gimbal;

[0052] is the motion parameter of the camera;

[0053] For time;

[0054] According to the above formula, the system calculates the frame sequence that needs to be cut in real time, giving priority to key frames with higher recognition value;

[0055] b. Adaptive resolution adjustment:

[0056] In the prior art, there may be a resolution mismatch problem in the images cut from the raw video stream. To solve this problem, an adaptive resolution adjustment algorithm is proposed to dynamically adjust the image resolution according to the onboard computing power and network bandwidth. By maintaining the ratio of the original picture, the algorithm can adapt the image resolution to the processor's load capacity without distortion, ensuring that the AI ​​recognition system can be processed in real time on low-computing power devices. Specifically:

[0057] ;

[0058] in,

[0059] is the resolution after adaptation;

[0060] is the resolution of the original image;

[0061] The current processor computing power usage;

[0062] The maximum computing power of the processor;

[0063] Through the above formula, the adaptive algorithm can dynamically reduce the image resolution to adapt to the capabilities of the current processing environment;

[0064] c. Time and space synchronization processing mechanism:

[0065] In order to meet the needs of real-time image cutting and recognition, a spatiotemporal synchronization processing mechanism is adopted. This mechanism performs image cutting and transmission while ensuring the frame rate. The image cutting process is 10 frames per second. During the processing, the timestamp synchronization technology is used to ensure that the cut image is seamlessly connected with the AI ​​recognition service to avoid asynchrony during data stream transmission. Specifically:

[0066] ;

[0067] in,

[0068] The time difference between image segmentation and AI recognition services;

[0069] Timestamp for image segmentation;

[0070] Timestamp for AI recognition service calls;

[0071] Through the above formula, the system can ensure that the image cutting time and the recognition service call time are synchronized at the microsecond level, thereby achieving efficient real-time processing;

[0072] Through the above method, the system can ensure that the image cutting time and the recognition service call time are synchronized at the microsecond level, thereby achieving efficient real-time processing. At the same time, the image resolution is dynamically adjusted according to the computing power and network conditions of the onboard equipment to ensure that the image is not distorted in a low computing power environment. At the same time, the time synchronization mechanism of the image cutting and recognition service of this solution improves real-time performance and accuracy.

[0073] The software of the human-computer interaction terminal provides a secondary development interface, which can be combined with other systems through subsequent expansion and development to achieve coordinated control between the drone detection and identification system and other systems. Figure 1 as shown).

[0074] 2. Design module

[0075] 2.1 Detection and Identification Module

[0076] The video stream data collected by the H20t camera is used for image cutting. Since the image transmission or conversion system causes incomplete image resolution and image detail contours, the video stream data needs to be hard-decoded and soft-decoded, mainly including streaming media services and AI recognition services. Currently, the common live broadcast protocols are RTMP, HLS, and HTTP-FLV. They are compared in terms of latency, whether they are private protocols, and plug-in support. Considering the performance consumption of the server and the delay of live broadcast distribution, the RTSP protocol was selected. The final technical selection plan is as follows (see the specific process for details). Figure 2 ):

[0077] (1) Streaming media transmission: RTSP;

[0078] (2) Video capture and encoding: FFMpeg is a cross-platform audio and video recording and transcoding solution;

[0079] (3) Recognition end: obtains pushed video stream data in real time and performs subsequent recognition and detection work;

[0080] 2.2 Coordinate solution module

[0081] The identification and solution steps are as follows:

[0082] Step 1: The algorithm recognition module uses the Yolo_V5 recognition algorithm to identify the target in the image. If the recognition is unsuccessful, the drone continues to fly to the waypoint; if the recognition is successful, according to the formula Get the pixel coordinates (X, Y) of the center point of the target object on the image.

[0083] Step 2: Use the formula Obtain the difference between the pixel coordinates and the pixel coordinates of the center point of the image Since the resolution of the acquired video stream is inconsistent with the resolution of the tablet itself, according to the formula , convert the pixel coordinate difference of the video stream resolution to the actual screen resolution .

[0084] Step 3: According to the focal length of the camera (default value is 4.5), Width (Default value is 6.29), Long Distance (Default value is 4.71) Drone flight altitude ( ), according to the formula Solve the actual length of the screen and the actual width ;

[0085] Step 4: According to the formula (in, and are the length and width of the screen respectively) to obtain the actual distance corresponding to the pixel coordinates ;

[0086] Step 5: Function formula , calculate the angles that the camera needs to rotate in two directions, and then use the formula , calculate the angle, and control the camera according to Turn to the target object at an angle of

[0087] Step 6: Calculate the length of each direction from the target point to the center point based on the pixel coordinates returned by the search target. The formula is: ;

[0088] Step 7: According to the coordinate inverse calculation formula, the longitude and latitude coordinates returned by the drone are known to solve the plane coordinates on the Gaussian plane , and combined with the calculated , and the yaw angle of the drone relative to the north direction , and finally get the absolute coordinates of the static target point , whose formula is , and finally the Gaussian plane coordinates calculated , use the Gaussian coordinate calculation formula to get the final longitude and latitude of the target object ;

[0089] 2.3 Multi-terminal collaboration module

[0090] The three-terminal collaborative operation module refers to the sharing of data and the issuance of commands to three different terminal devices in the information system: mobile application (App), command center (PC) and portable control terminal. These three terminal devices can achieve real-time data synchronization and sharing, so that users can obtain the same information and functions on different devices.

[0091] In field combat missions, the command and control center, as a management and monitoring system, can view the real-time location and status information of the drone in real time, and better command and monitor the mobile terminals (drone equipment) and control terminals. As the field main control terminal, the portable control terminal can quickly monitor and identify suspicious targets in the field environment, provide the command and control center with important front-line data and respond quickly. The mobile app can identify the target and location through sharing, and can quickly navigate to the target location, truly realizing the purpose of three-terminal linkage and rapid response.

[0092] 2.4 Implementation steps

[0093] S1. The command center sends tasks to the portable control terminal. After receiving the task instructions, the control terminal sends the instructions to the onboard lightweight processor through point-to-point radio communication, calls the PSDK control service, and issues and executes the tasks.

[0094] S2. The control terminal interface displays the real-time location and map data of the drone by calling the GIS map service and data middleware on the onboard lightweight processor; the real-time video stream returned by the gimbal camera is viewed and broadcast live by calling the video streaming service; suspicious targets in the video frame are quickly identified by calling the lightweight recognition and detection service and the results are given to the portable control terminal.

[0095] S3. The control terminal displays the recognition result in the video frame in real time and distributes the video stream to the mobile phone.

[0096] S4. Log in to the authorized app on the mobile phone to quickly view the identification results and provide navigation, guiding the operator to quickly approach the suspicious target for real-time tracking and attack.

[0097] Those of ordinary skill in the art will appreciate that the units and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0098] In the several embodiments provided in the present application, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the division of the units described above is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The above-mentioned units may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

[0100] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A portable multi-terminal cooperative combat device in wild areas, characterized in that: include: The sky end includes a flight platform, which is equipped with a collection unit and an airborne communication unit. The collection unit collects the required images to form a data stream, and the data stream is shared with the ground end through the airborne communication unit. The ground side forms a communication link by networking with the airborne communication unit through the ground communication unit, and uses the communication link to send control instructions to the sky side to control the flight platform and / or the acquisition unit, and obtain the status information of the sky side and the collected data stream; The flying platform is equipped with an onboard lightweight processor, a collection unit including a zoom wide-angle camera, and an onboard communication unit including an ad hoc network radio; The ground end includes a self-organizing network radio station and a human-computer interaction terminal, and there is a wired / wireless connection between the self-organizing network radio station and the human-computer interaction terminal; The ground end connects and communicates with the onboard computer on the flight platform through the software installed on the human-computer interaction terminal to send control instructions to control the flight platform movement or the gimbal and camera movement of the load; The obtaining of the status information of the air terminal includes receiving the flight platform status data and the sensor data of the payload gimbal and camera, and displaying the obtained status information on the screen of the human-computer interaction terminal; The data stream obtained from the sky end is divided into two paths. One path is used for image recognition on the onboard lightweight processor, and the other path is used to directly push the raw video stream to the ground end. Coordinate calculation is required for the acquisition unit, including: S1. Use the Yolo_V5 recognition algorithm to identify the target in the image. If the recognition is unsuccessful, the drone continues to fly to the waypoint; if the recognition is successful, according to the formula Obtain the pixel coordinates (X, Y) of the center point of the target object on the image, where X1 and X2 are the pixel coordinates of the identified target object in the X direction, and Y1 and Y2 are the pixel coordinates of the identified target object in the Y direction; S2. Using formula Obtain the difference between the pixel coordinates and the pixel coordinates of the center point of the image , according to the formula , convert the pixel coordinate difference of the video stream resolution to the actual screen resolution , where X 中 is the X-coordinate of the center pixel of the entire image, Y 中 The Y coordinate of the center pixel of the entire image; S3. According to the focal length of the camera , Wide , long distance , the UAV flight altitude, using the formula Solve the actual length of the screen and the actual width , where h is the flight altitude of the UAV; S4. According to the formula Get the actual distance corresponding to the pixel coordinates ,in, and are the length and width of the screen respectively, where L1 is the actual length corresponding to the screen, and L2 is the actual length corresponding to the screen; S5. Function formula , calculate the angles that the camera needs to rotate in two directions, and then use the formula , calculate the angle, and control the camera according to The angle of rotation is θ1, which is the X-axis rotation angle and θ2, which is the Y-axis rotation angle. S6. Calculate the length of each direction from the target point to the center point according to the pixel coordinates returned by the search target. The formula is: ; S7. According to the coordinate inverse calculation formula, the longitude and latitude coordinates returned by the drone are known to solve the plane coordinates on the Gaussian plane. , and combined with the calculated , and the yaw angle of the drone relative to the north direction , and finally get the absolute coordinates of the static target point , whose formula is , and finally the Gaussian plane coordinates calculated , use the Gaussian coordinate calculation formula to get the final longitude and latitude of the target object .

2. The portable multi-terminal cooperative combat device in wild areas according to claim 1 is characterized in that: For the data stream obtained by the acquisition unit, data hard decoding and soft decoding are required, as follows: Use RTSP for streaming media transmission; Use FFMpeg for video capture and encoding; Use the recognition terminal to obtain the pushed video stream data in real time and perform subsequent recognition and detection work.

3. The portable multi-terminal cooperative combat device in wild areas according to claim 1 is characterized in that: The mobile phone application, command center and portable control terminal are combined into a three-end collaborative operation module, which can achieve real-time data synchronization and sharing among the three-end collaborative operation modules.

4. A method for field portable multi-terminal collaborative operations, used to implement the field portable multi-terminal collaborative operations device described in any one of claims 1 to 3, characterized in that: include: The command center sends the task to the portable control terminal. After receiving the task instruction, the control terminal sends the instruction to the onboard lightweight processor through point-to-point radio communication, calls the PSDK control service, and issues and executes the task. The control terminal interface displays the real-time location and map data of the drone by calling the GIS map service and data middleware on the onboard lightweight processor; By calling the video streaming service, the real-time video stream returned by the PTZ camera can be viewed and broadcast live; by calling the lightweight recognition and detection service, suspicious targets in the video frame can be quickly identified and the results can be sent to the portable control terminal; The control terminal displays the recognition result in the video frame in real time and distributes the video stream to the mobile phone; Log in to the authorized app on the mobile phone to quickly view the identification results and provide navigation, guiding the operator to quickly approach the suspicious target for real-time tracking and attack.

Images