Unmanned aerial vehicle cruise target recognition device and method

Abstract

The invention discloses an unmanned aerial vehicle cruise target recognition device and method, and belongs to the technical field of unmanned aerial vehicle target recognition. The unmanned aerial vehicle cruise target recognition device comprises a data acquisition module, a wireless communication module, an image recognition module and a data processing module, the data acquisition module comprises a main camera and an auxiliary camera, the main camera is arranged at the center of the bottom surface of a mounting box, the auxiliary camera is arranged at the lower part of the main camera, and the wireless communication module is connected with the data processing module. Environment cameras are respectively arranged at four corners of the bottom surface of the mounting box, and the data acquisition module and the image recognition module are connected with the data processing module through the wireless communication module; according to the invention, through the arrangement of the auxiliary camera, under the condition that the target is not easy to identify, the auxiliary camera can be utilized to change the focal length and carry out amplification identification, and the main camera is not influenced to continuously identify other areas, so that the problem that the efficiency is influenced sometimes when the unmanned aerial vehicle identifies the target in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle target recognition, and in particular to a device and method for unmanned aerial vehicle cruise target recognition. Background Art

[0002] Unmanned aerial vehicles, also known as "drones" ("UAV"), are unmanned aerial vehicles that are controlled by radio remote control equipment and self-contained program control devices. Their current applications in aerial photography, agriculture, plant protection, power inspection, disaster relief, film and television shooting, military and other fields have greatly expanded the use of drones themselves and have become a new tool that various industries are keen on.

[0003] With the advancement of AI algorithms, the use of drones for target recognition has also developed. However, in existing technologies, when performing target recognition, usually only one camera is used in conjunction with an AI algorithm. In some special cases, such as looking for a specific target in a complex environment, the camera is required to check one by one, and even the drone must be lowered or the focal length adjusted to magnify the image for inspection. This will result in a smaller recognition range and affect efficiency, so improvement is needed. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that the efficiency of UAV target recognition is sometimes affected in the prior art, and to propose a UAV cruise target recognition device and method.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A drone cruise target recognition device comprises a data acquisition module, a wireless communication module, an image recognition module and a data processing module. The data acquisition module comprises a main camera and a secondary camera. The main camera is arranged at the center of the bottom surface of an installation box, and the secondary camera is arranged below the main camera. Environmental cameras are respectively arranged at the four corners of the bottom surface of the installation box. The data acquisition module and the image recognition module are connected via the wireless communication module and the data processing module.

[0006] Preferably, the installation box is connected to the bottom of the drone via a connector.

[0007] Preferably, a second support is rotatably provided at the bottom center of the installation box, and the main camera is rotatably provided on the second support.

[0008] Preferably, a mounting bracket is provided at the bottom of the second support, a first support is rotatably provided on the bottom surface of the mounting bracket, and the auxiliary camera is rotatably provided on the first support.

[0009] Preferably, the second support and the mounting box, the main camera and the second support, the first support and the mounting frame, and the first support and the auxiliary camera are all driven to rotate by motors.

[0010] Preferably, both the main camera and the auxiliary camera are provided with image recognition modules.

[0011] Preferably, the image collected by the data collection module and the recognition data of the image recognition module are transmitted to the data processing module via the wireless communication module.

[0012] Preferably, the method comprises the following steps: S1. Train a neural network for the data acquisition module to obtain a detection model, and detect and identify the target according to the detection model; S2, input the target information into the image recognition module, and use the drone to fly to the target area for cruise identification; S3, first use the main camera to acquire images and identify them. When a suspected target appears in the image acquisition range, if the target's surrounding environment is not complicated, such as the target is in an open area, it is easy to shoot the target for identification, or the target is relatively large and can be identified without adjusting the height of the drone or changing the focal length of the main camera, then jump to step S4; if the target's surrounding environment is complex, such as other objects blocking part of the target, which is not conducive to accurate target identification, or the target itself is small, and the drone height needs to be adjusted or the main camera needs to change the focal length for identification, then jump to step S5; S4. When the main camera recognizes and confirms the target, it can lock the target, and the main camera can be rotated under the drive of the motor to ensure that the main camera can always keep the target within the shooting range of the main camera; S5: When the main camera can only roughly identify the target, the secondary camera changes the focal length and zooms in on the suspected target for identification, while the main camera continues to search and identify other areas; S6, the environment camera, is used to shoot the surrounding environment during the flight of the drone. It uses oblique photography to shoot. After the target is recognized in S4 and S5, the environment camera focuses on recording the surrounding environment; S7. The data and identification data collected in steps S3-S6 are transmitted to the data processing module through the wireless communication module for data aggregation, and the control center issues commands after identification, such as tracking, warning, attack, etc.

[0013] Compared with the prior art, the present invention provides a UAV cruise target identification device and method, which have the following beneficial effects.

[0014] 1. The present invention, by setting up a secondary camera, can change the focal length of the secondary camera to zoom in and identify the target when it is difficult to identify the target, while not affecting the main camera to continue to identify other areas, thereby solving the problem of efficiency sometimes being affected when identifying drone targets in the prior art.

[0015] Other advantages, objectives and features of the present invention will be described in part in the following description; and in part, will be apparent to those skilled in the art based on an examination of the following; or, may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0017] Figure 2 It is a schematic diagram of the overall structure of the present invention.

[0018] Figure 3 It is a schematic diagram of the partially disassembled structure of the present invention.

[0019] In the figure: 1. Connector; 2. Installation box; 3. Environmental camera; 4. Main camera; 5. Auxiliary camera; 6. First support; 7. Mounting bracket; 8. Second support. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0021] Reference Figure 1-3 A drone cruise target recognition device includes a data acquisition module, a wireless communication module, an image recognition module and a data processing module. The data acquisition module includes a main camera 4 and a sub-camera 5. The main camera 4 is arranged at the center of the bottom surface of the installation box 2, and the sub-camera 5 is arranged at the lower part of the main camera 4. The four corners of the bottom surface of the installation box 2 are respectively provided with environmental cameras 3. The data acquisition module and the image recognition module are connected through the wireless communication module and the data processing module.

[0022] In a specific embodiment of the present invention, the wireless communication module adopts the conventional setting of the prior art, and is mainly used for data transmission between the data acquisition module and the data processing module. The data processing module is arranged in the control center for data storage, analysis and comparison. The image recognition module adopts the algorithm of the prior art, and can use the AI ​​face recognition algorithm to perform face recognition, so as to facilitate the use of drones to cruise and find specific people. The environmental camera 3 is mainly based on oblique photography technology, and is used to shoot the specific environment and terrain around the target, so as to facilitate other work after locking and identifying the target.

[0023] The installation box 2 is connected to the bottom of the drone via the connector 1 .

[0024] A second support 8 is rotatably disposed at the bottom center of the installation box 2 , and the main camera 4 is rotatably disposed on the second support 8 .

[0025] A mounting frame 7 is provided at the bottom of the second support 8 , a first support 6 is rotatably provided on the bottom surface of the mounting frame 7 , and the auxiliary camera 5 is rotatably provided on the first support 6 .

[0026] Motors are used to drive the rotation between the second support 8 and the installation box 2, between the main camera 4 and the second support 8, between the first support 6 and the mounting frame 7, and between the first support 6 and the auxiliary camera 5. This arrangement allows the main camera 4 and the auxiliary camera 5 to adjust their angles, making it easier to lock on to the target. The control program for controlling the motor can use the technology of existing cameras with tracking functions.

[0027] Both the main camera 4 and the auxiliary camera 5 are provided with image recognition modules.

[0028] The images collected by the data collection module and the recognition data of the image recognition module are transmitted to the data processing module through the wireless communication module.

[0029] A recognition method for a drone cruise target recognition device comprises the following steps: S1. Train a neural network for the data acquisition module to obtain a detection model, and detect and identify the target based on the detection model. This step is mainly used to establish a database so that the data acquisition module can compare and identify the collected information with the database. When the main camera 4 is working, all objects in the field of view can be compared and identified with the data in the database.

[0030] S2. Input the target information into the image recognition module, and use the drone to fly to the target area for cruise recognition. In this step, the target information can be a category, such as a person, a car, etc., or a specific person, a specific car, etc.

[0031] S3. First, use the main camera 4 to acquire and identify the image. When a suspected target appears within the image acquisition range, if the surrounding environment of the target is not complicated, such as the target is in an open area and is easy to shoot and identify, or the target is relatively large and can be identified without adjusting the height of the drone or changing the focal length of the main camera 4, then jump to step S4. If the surrounding environment of the target is complex, such as other objects blocking part of the target, which is not conducive to accurate target identification, or the target itself is small and needs to adjust the height of the drone or change the focal length of the main camera 4 for identification, then jump to step S5; in this step, the suspected target (referring to the target information in S2) is identified.

[0032] S4, when the main camera 4 recognizes and confirms the target, it can lock the target, and under the drive of the motor, the main camera 4 can be rotated to ensure that the main camera 4 can always keep the target within the shooting range of the main camera 4; S5. When the main camera 4 can only roughly determine the target, the sub-camera 5 changes the focal length and zooms in on the suspected target for identification, while the main camera 4 continues to search and identify other areas; the main camera 4 is mainly used for large-area scanning and identification, and the sub-camera 5 is used for local precise identification. The sub-camera 5 can adjust the angle and focal length at will to compare more features, but it does not affect the large-area identification of the main camera 4, so it can make finding specific targets in the area more efficient.

[0033] S6, the environment camera 3 is used to shoot the surrounding environment during the flight of the drone, using oblique photography. After the target is recognized in S4 and S5, the environment camera 3 focuses on recording the surrounding environment; S7. The data and identification data collected in steps S3-S6 are transmitted to the data processing module through the wireless communication module for data aggregation, and the control center issues commands after identification, such as tracking, warning, attack, etc.

[0034] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0035] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0036] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A drone cruise target recognition device, characterized in that: The invention comprises a data acquisition module, a wireless communication module, an image recognition module and a data processing module, wherein the data acquisition module comprises a main camera (4) and a secondary camera (5), wherein the main camera (4) is arranged at the center of the bottom surface of the installation box (2), and the secondary camera (5) is arranged below the main camera (4). Environmental cameras (3) are respectively arranged at the four corners of the bottom surface of the installation box (2), and the data acquisition module and the image recognition module are connected via the wireless communication module and the data processing module.

2. The UAV cruise target identification device according to claim 1, characterized in that: The installation box (2) is connected to the bottom of the drone via a connecting piece (1).

3. The UAV cruise target identification device according to claim 1, characterized in that: A second support (8) is rotatably arranged at the bottom center of the installation box (2), and the main camera (4) is rotatably arranged on the second support (8).

4. The UAV cruise target identification device according to claim 3 is characterized in that: A mounting frame (7) is arranged at the bottom of the second support (8), a first support (6) is rotatably arranged on the bottom surface of the mounting frame (7), and the auxiliary camera (5) is rotatably arranged on the first support (6).

5. The UAV cruise target identification device according to claim 4, characterized in that: The second support (8) and the installation box (2), the main camera (4) and the second support (8), the first support (6) and the installation frame (7), and the first support (6) and the auxiliary camera (5) are all driven to rotate by a motor.

6. The UAV cruise target identification device according to claim 1, characterized in that: The main camera (4) and the auxiliary camera (5) are both provided with image recognition modules.

7. The UAV cruise target identification device according to claim 1, characterized in that: The images collected by the data collection module and the recognition data of the image recognition module are transmitted to the data processing module through the wireless communication module.

8. The identification method of the UAV cruise target identification device according to any one of claims 1 to 7, characterized in that: The steps include: S1. Train a neural network for the data acquisition module to obtain a detection model, and detect and identify the target according to the detection model; S2, input the target information into the image recognition module, and use the drone to fly to the target area for cruise identification; S3, first use the main camera (4) to acquire an image and identify it. When a suspected target appears within the image acquisition range, if the target's surrounding environment is not complicated, such as the target is in an open area and is easy to photograph and identify, or the target is relatively large and can be identified without adjusting the height of the drone or changing the focal length of the main camera (4), then jump to step S4. If the target's surrounding environment is complicated, such as other objects blocking part of the target, which is not conducive to accurate target identification, or the target itself is small and needs to adjust the height of the drone or change the focal length of the main camera (4) for identification, then jump to step S5. S4, when the main camera (4) recognizes and confirms the target, it can lock the target, and under the drive of the motor, the main camera (4) can be rotated to ensure that the main camera (4) can always keep the target within the shooting range of the main camera (4); S5, when the main camera (4) can only roughly identify the target, the secondary camera (5) changes the focal length and zooms in on the suspected target for identification, while the main camera (4) continues to search and identify other areas; S6, the environment camera (3) is used to photograph the surrounding environment during the flight of the UAV, using an oblique photography method. After the target is identified in S4 and S5, the environment camera (3) focuses on recording the surrounding environment; S7. The data and identification data collected in steps S3-S6 are transmitted to the data processing module through the wireless communication module for data aggregation, and the control center issues the identified command.

Images