Image recognition and target tracking method and system based on unmanned aerial vehicle

Abstract

The invention relates to an image recognition and target tracking method and system based on an unmanned aerial vehicle, and aims to provide a technical scheme capable of realizing efficient and accurate target recognition and tracking. According to the invention, the high-resolution camera is installed on the unmanned aerial vehicle, and the recognition and tracking of targets such as the ground, buildings, vehicles and the like are realized by using an advanced image recognition algorithm. According to the image recognition algorithm, the deep learning technology is adopted, image features can be automatically extracted, feature points do not need to be manually selected, and the recognition accuracy and real-time performance are improved. Meanwhile, a multi-target tracking technology is adopted in the target tracking algorithm, multiple targets can be tracked at the same time, and the target tracking effect is better than that of a traditional technology. The system also has a series of advantages. Firstly, the system can achieve the functions of automatic take-off, autonomous flight, automatic landing and the like, and the autonomy of the unmanned aerial vehicle is improved. And secondly, the system can automatically plan the optimal flight path by analyzing the target position and the environment information, so that the efficiency of the unmanned aerial vehicle is improved. And finally, the system can automatically alarm the user, measures are taken in time, and accidents are avoided.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and more particularly to image recognition and target tracking technology based on UAVs. Background Technology

[0002] With the continuous development of drone technology, drones have been widely used in both civilian and military fields. However, current drones often require manual operation, and the accuracy and autonomy of target tracking need to be improved. While some image recognition and target tracking systems exist, they often require human intervention and have poor adaptability to complex environments, failing to meet practical application needs. Summary of the Invention

[0003] This invention provides an image recognition and target tracking method and system based on unmanned aerial vehicles (UAVs). The system includes an UAV, an image acquisition module, an image processing module, a target tracking module, a system obstacle avoidance module, and an anomaly alarm module.

[0004] The image acquisition module captures images of the area surrounding the drone and transmits them to the image processing module. The image processing module preprocesses and analyzes the acquired images, extracts target information, and transmits the results to the target tracking module. The target tracking module automatically adjusts the drone's flight path and speed based on the prediction of the target's position and speed to track the target in a timely manner. The obstacle avoidance module detects obstacles around the drone and automatically avoids them, ensuring safe flight. The anomaly alarm module detects deviations from the flight path or other abnormal situations and automatically alerts the user.

[0005] The image acquisition module includes a high-definition camera, an image transmission device, and an image processor. The high-definition camera is used to acquire images of the area surrounding the drone, and the image transmission device transmits the images to the image processor. The image processor preprocesses and analyzes the acquired images, extracts target information, and transmits the results to the target tracking module.

[0006] The target tracking module includes a target predictor, a flight controller, and an adjuster. The target predictor predicts the target's next position and velocity based on changes in the target's position and velocity. The flight controller automatically adjusts the UAV's flight path and speed based on the prediction results, ensuring the target remains within the field of view. The adjuster automatically adjusts the flight attitude and speed based on real-time environmental information, ensuring stable flight of the UAV and accurate target tracking.

[0007] The system's obstacle avoidance module includes an obstacle detector, an obstacle avoider, and a path planner. The obstacle detector detects obstacles around the drone and transmits this information to the obstacle avoider. The obstacle avoider automatically avoids obstacles based on their position and shape, ensuring safe flight for the drone. The path planner automatically plans the optimal flight path based on the target location and environmental information, improving the drone's autonomy and efficiency.

[0008] The anomaly alarm module includes an anomaly detector and an alarm. The anomaly detector can detect when a target deviates from its flight path or other abnormal situations and transmit the information to the alarm. The alarm can automatically alert the user so that timely action can be taken.

[0009] Drone platform: includes a drone with autonomous flight capability and a controllable remote controller.

[0010] Camera: A high-resolution camera mounted on the drone can capture images of the target area.

[0011] Image recognition algorithm: Using deep learning technology, it analyzes images and automatically extracts feature points to recognize targets such as the ground, buildings, and vehicles.

[0012] Target tracking algorithm: Employing multi-target tracking technology, it can track multiple targets simultaneously and adaptively adjust the tracking algorithm parameters during the tracking process to improve tracking performance.

[0013] Control algorithm: Based on the target location information and environmental information provided by the target tracking algorithm, the optimal flight path of the UAV is automatically planned and the UAV is controlled to fly along the path, thereby achieving target tracking.

[0014] User interface: The interface displays real-time video captured by the camera and information about the tracked target to the user, making it convenient for the user to monitor and control the target.

[0015] Early warning system: When the system detects abnormal behavior of the target or abnormal conditions of the drone, it will automatically issue an alarm to the user and take corresponding measures to ensure flight safety.

[0016] The system of this invention has a series of advantages. It enables UAVs to autonomously track targets and avoid obstacles, improving the autonomy and safety of the UAVs; it can process and analyze the acquired images in real time to extract target information, thereby tracking targets more accurately; it can automatically adjust flight attitude and speed according to real-time environmental information to ensure stable flight and target tracking; it can automatically alert the user and take timely measures to avoid accidents; and it can automatically plan the optimal flight path according to the target location and environmental information, improving the efficiency and autonomy of the UAVs. Attached Figure Description

[0017] Figure 1 is a flowchart of the image recognition and target tracking method based on UAV of the present invention; Figure 2 is a schematic diagram of the image recognition and target tracking system based on UAV of the present invention.

[0018] Figure 3 is a schematic diagram of the structure of the UAV of the present invention. Implementation

[0019] The specific implementation steps of the present invention are as follows: (1) Start the drone platform and connect the remote controller to the drone platform.

[0020] (2) Preset or input relevant parameters for the UAV flight mission, including takeoff point, target area, flight altitude, flight speed, etc.

[0021] (3) The system starts the image recognition algorithm and inputs the real-time video stream from the camera into the algorithm for processing.

[0022] (4) The image recognition algorithm automatically extracts the feature points of the target area and classifies and recognizes the target through a pre-trained model.

[0023] (5) The system starts the target tracking algorithm and automatically tracks the target based on the target location information provided by the image recognition algorithm. If the target is occluded or moves too fast, the system will automatically adjust the parameters of the tracking algorithm to improve the tracking effect.

[0024] (6) The control algorithm plans the optimal flight path of the UAV based on the target location information and environmental information, and controls the UAV to fly along the path.

[0025] (7) The user interface displays real-time video captured by the camera and tracking target information to the user, making it convenient for the user to monitor and control the target.

[0026] (8) The early warning system monitors the status of the UAV platform and target, and issues an alarm to the user when an abnormal situation is detected.

[0027] (9) After the mission is completed, the UAV platform will automatically return to home and complete the flight mission.

[0028] The above-described specific embodiments are merely preferred embodiments of this invention and are not intended to limit this invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the protection scope of this invention.

Claims

1. This invention provides an image recognition and target tracking method and system based on unmanned aerial vehicles (UAVs), characterized in that... It includes one or more drones, each equipped with an image acquisition device and sensors; the image acquisition device is capable of acquiring images and transmitting them to a target tracking system; the target tracking system includes an image recognition module and a target tracking module, used to process image data acquired from the drone's image acquisition device; The image recognition module can automatically identify targets in an image and classify them; The target tracking module can track the identified target and generate its motion trajectory.

2. The method and system according to claim 1 are characterized in that... The image acquisition device includes a camera (1) and a thermal imaging camera (2); the image acquisition device can acquire images under different lighting, weather and environmental conditions, and the image recognition module can use artificial intelligence technology, such as deep learning, convolutional neural networks, etc. The target tracking module can use a variety of algorithms, such as Kalman filtering and particle filtering; the target tracking module can predict the future position of the target so that the UAV can adjust its flight path and speed in advance.

3. The method and system according to claim 1 are characterized in that... The target tracking module automatically adjusts the drone's altitude and direction to keep the target within the camera's field of view; it automatically alerts the user when the target deviates from the flight path or other abnormalities occur; it automatically avoids obstacles during drone flight and can make an emergency landing when necessary; and it transmits image data to the user's terminal device, such as a computer, mobile phone, or tablet, in real time. The target tracking module can identify parameters such as the target's speed, direction, and size to enable the drone to better track the target; the image recognition module can identify and record the target's attribute information, such as color and shape, while recognizing the target. The system can support the simultaneous tracking and identification of multiple targets; the system can automatically switch and shift when a target enters or leaves the camera's field of view to ensure that the target is not lost. The system can automatically adjust the number and deployment location of drones according to user needs to achieve better target tracking results.

Images