Grassland grazing forbidding system and method based on unmanned aerial vehicle

Abstract

The invention discloses a grassland grazing forbidding system and method based on an unmanned aerial vehicle, and belongs to the field of grassland grazing forbidding management. The system comprises an unmanned aerial vehicle, a ground control station, an AI identification module, a communication module, an alarm module and a management and protection personnel terminal. The method comprises the steps that the unmanned aerial vehicle collects data and transmits the data to the ground control station; the ground control station transmits the data to the AI recognition module, and the AI recognition module recognizes and judges whether grazing is stolen or not; and if the grazing is stolen, the alarm module generates alarm information and pushes the alarm information to the management and protection personnel terminal through the communication module, and the management and protection personnel dispose. According to the method, the unmanned aerial vehicle and the AI technology are utilized, the efficiency, the accuracy and the real-time performance of grassland grazing forbidding management are improved, human subjective factors are avoided, comprehensive and effective supervision on the grassland is achieved, and the ecology of the grassland is protected.

Description

Technical Field

[0001] This invention relates to the field of grassland grazing ban management technology, and in particular to a grassland grazing ban system and method based on unmanned aerial vehicles (UAVs). Background Technology

[0002] Grasslands, as an important ecosystem, play an irreplaceable role in maintaining ecological balance and preventing desertification.

[0003] Traditional grassland grazing ban management mainly relies on manual patrols, which has many drawbacks. Grasslands have complex terrain, vast areas, and scattered management units, making it difficult for manual patrols to cover all areas comprehensively. This results in low efficiency, high cost, and inability to detect and deal with illegal grazing in a timely manner. Furthermore, subjective factors such as kinship between patrol personnel and illegal grazing perpetrators may lead to cover-ups.

[0004] In some areas, video surveillance systems are used for inspections, but these systems are mostly installed high in the air on towers, making them significantly affected by the camera's viewing distance, terrain, and wind speed. A camera "standing high" doesn't necessarily "see far," and wind speed can cause the camera to shake, resulting in "unclear images." Furthermore, ultra-long-range, star-level industrial cameras are extremely expensive and their widespread application is limited by the location of on-site operator base stations.

[0005] Therefore, traditional grassland grazing ban management methods are inadequate in terms of efficiency, cost, coverage, real-time performance, and objectivity, and cannot meet the needs of efficient and precise grassland grazing ban management. This invention aims to solve these problems. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a grassland grazing ban system and method based on unmanned aerial vehicles (UAVs) to improve the efficiency and accuracy of grassland grazing ban management.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A grassland grazing ban system based on drones includes drones, a ground control station, an AI recognition module, a communication module, an alarm module, and a terminal for management personnel.

[0008] The drone is equipped with a high-definition camera, GPS positioning module, and wireless communication module. The high-definition camera can be a model with at least 12 megapixels, such as the Sony IMX686, ensuring clear images and videos of the grassland. The GPS positioning module uses a product with an accuracy of within 1 meter, such as the UBLOX ZED-F9P, to accurately obtain the drone's location. The wireless communication module can be a 4G / 5G module, such as the Quectel EC200S, to achieve data transmission with the ground control station and communication module. The drone can fly according to a preset route or commands from the ground control station, achieving all-round, real-time patrols. Its flexibility overcomes terrain limitations, covering areas that are difficult for humans to reach.

[0009] Ground control station: Employs a high-performance industrial computer, such as the Advantech IPC-810, equipped with professional flight control software, such as the DJI GS Pro. Its functions include planning flight routes, including optimal patrol paths based on grassland terrain features and key grazing-prohibited areas; and sending flight commands, such as adjusting altitude, speed, and heading. Simultaneously, the ground control station receives data transmitted from the drone and forwards it to the AI ​​recognition module, serving as the system's control center.

[0010] AI Recognition Module: This module utilizes a server equipped with a GPU, such as the NVIDIA Jetson AGX Xavier, which possesses powerful data processing capabilities. It receives data transmitted from the ground control station and uses AI algorithms to identify animals such as cattle and sheep. The deep learning algorithms employed, such as Convolutional Neural Networks (CNNs), are trained on a large number of grassland animal images to accurately identify targets, distinguish between normal grazing and illegal grazing, improve recognition accuracy and robustness, and reduce false positives.

[0011] Communication module: It adopts a communication solution that integrates 4G, 5G and Wi-Fi, and the core module can be Huawei ME909s-821. 4G / 5G ensures real-time data transmission under wide coverage, while Wi-Fi can be used for fast exchange of large amounts of data at short distances, ensuring stable data communication between modules and maintaining connectivity even in remote grassland areas.

[0012] Alarm module: Composed of an embedded system, such as a development board based on STM32H743. When the AI ​​recognition module detects grazing violations, this module integrates information such as the location, time, and images of the grazing violations to generate alarm information, ensuring information integrity and providing accurate information for management personnel to take action.

[0013] Maintenance personnel terminal: Select a smartphone or tablet with good communication capabilities and long battery life, such as Huawei Mate 60 Pro or Apple iPad Pro, and install a dedicated management app. The terminal receives alarm information and displays relevant maps, drone locations, historical records, etc., enabling maintenance personnel to respond quickly and go to the site in a timely manner.

[0014] This invention also provides a grassland grazing ban method based on the above system, the steps of which are as follows: Data Acquisition: The drone flies along a preset route or according to commands from the ground control station, capturing image and video data using a high-definition camera. At night or in low light conditions, optional infrared cameras such as the FLIR Boson 640 and multispectral cameras such as the MicaSenseRedEdge-MX can assist in analyzing vegetation conditions and indirectly determine the risk of overgrazing. The collected data is transmitted to the ground control station via a wireless communication module, enabling comprehensive and real-time acquisition of grassland information.

[0015] Image Recognition: The ground control station transmits data to the AI ​​recognition module. The module first preprocesses the image, such as using Gaussian filtering to reduce noise and histogram equalization to enhance contrast and improve image quality. Then, it uses target detection algorithms such as YOLO and Faster R-CNN to detect and locate cattle and sheep, and count their numbers. If the number of animals in the prohibited grazing area exceeds a user-defined threshold, it is determined to be illegal grazing. This step uses AI technology to achieve intelligent recognition, overcoming the subjectivity and fatigue of human judgment.

[0016] Alarm generation: After the AI ​​recognition module determines that grazing has occurred, it triggers the alarm module to generate alarm information that includes the location of the grazing (combined with drone GPS information), time, and on-site images, providing key clues for subsequent handling.

[0017] Alarm push notification: The alarm module pushes alarm information to the maintenance personnel's terminal through the communication module. The terminal APP provides instant reminders, and the maintenance personnel can quickly plan routes to stop the alarm, record the situation, etc., to achieve rapid response.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. Drones have the characteristics of long flight distance and high flexibility. They can conduct all-round patrols of grasslands according to preset routes or instructions from ground control stations. They can quickly cover large grassland areas, including areas with complex terrain that are difficult for humans to reach. They overcome the drawbacks of traditional manual patrols, which are limited by terrain and have limited coverage. They greatly improve the comprehensiveness and efficiency of patrols and reduce the time and cost required for manual patrols.

[0019] 2. Drones can collect images and video data of the grassland in real time and quickly transmit them to the ground control station and AI recognition module via the communication module, enabling managers to monitor grassland dynamics in real time. When the AI ​​recognition module detects illegal grazing, the alarm module can immediately generate alarm information including the location, time, and images of the illegal grazing, and push it to the management personnel's terminal, facilitating rapid on-site response by management personnel. This solves the problem that traditional manual patrols and video monitoring systems cannot detect and handle violations in a timely manner, improving the response speed of grassland grazing ban management.

[0020] 3. AI algorithms (such as convolutional neural networks, YOLO, Faster R-CNN, etc.) are used to analyze images collected by drones. Through extensive sample training, the system can accurately identify animals such as cattle and sheep, effectively distinguishing between normal grazing and illegal grazing, with high accuracy and robustness. Furthermore, the entire identification process is automated, avoiding the possibility of subjective factors such as personal relationships leading to the cover-up of illegal grazing during manual inspections, thus ensuring the fairness and objectivity of management.

[0021] 4. Compared to traditional manual patrols which require significant manpower and resources, and video surveillance systems which necessitate the installation of expensive, long-range cameras and are limited by base station locations, this system fully utilizes drones and mature wireless communication technologies (4G, 5G, Wi-Fi, etc.). This ensures effective monitoring while reducing equipment purchase and operation / maintenance costs. Furthermore, the system is flexible in deployment, less restricted by terrain and base station locations, making it easier to promote and apply across vast grassland areas.

[0022] 5. A three-level linkage technical architecture of "air-ground-human" has been constructed, which forms a complete closed loop of data and management process from front-end data collection of UAVs, data transfer of ground control stations, intelligent analysis of AI recognition modules, information generation of alarm modules, and handling feedback of management personnel terminals. This makes grassland grazing ban management more standardized and systematic, helps to consolidate the achievements of grassland ecological construction, and ensures the restoration and stability of grassland ecosystems. Attached Figure Description

[0023] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart illustrating the process from the drone to the AI ​​recognition module in an embodiment of the present invention. Figure 2 This is a flowchart of the process from the AI ​​recognition module to the terminal application layer in an embodiment of the present invention. Detailed Implementation

[0024] The technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0025] like Figure 1 , Figure 2 As shown, a grassland grazing ban system based on drones includes drones, a ground control station, an AI recognition module, a communication module, an alarm module, and a terminal for management personnel.

[0026] The drone uses a DJI Matrice 300 RTK, equipped with a 20-megapixel high-definition camera capable of clearly capturing images and videos of the grassland. Its GPS positioning module boasts centimeter-level accuracy, precisely determining the drone's location. The wireless communication module supports 4G / 5G and Wi-Fi, ensuring stable data transmission. The drone can automatically fly along routes planned by the ground control station and can also receive real-time commands from the station to adjust its flight path, enabling comprehensive patrols of the grassland.

[0027] The ground control station uses an Advantech IPC-610L industrial computer and installs Mission Planner flight control software. Operators use this software to plan the drone's flight path, such as setting waypoints, and control the drone's takeoff, landing, hovering, and other flight states. Simultaneously, the ground control station receives image and video data transmitted from the drone in real time and promptly transmits this data to the AI ​​recognition module.

[0028] The AI ​​recognition module utilizes an NVIDIA DGX Station server, whose GPU can efficiently run convolutional neural network (CNN) algorithms. Trained on a large number of grassland cattle and sheep image samples, this module achieves extremely high recognition accuracy. Upon receiving image and video data transmitted from the ground control station, it uses the trained CNN algorithm to identify cattle, sheep, and other animals in the images.

[0029] The communication module uses Huawei's MH5000-31 5G module, which supports multiple frequency bands and can achieve stable wireless communication in complex environments such as grasslands, ensuring real-time data transmission between drones, ground control stations, AI recognition modules, alarm modules, and management personnel terminals.

[0030] The alarm module uses a Raspberry Pi 4 development board based on the ARM architecture. When the AI ​​recognition module detects grazing behavior, the Raspberry Pi 4 development board will quickly generate alarm information based on the received information such as the location, time, and images of the grazing.

[0031] The management personnel use Huawei Mate 60 Pro smartphones as their terminals, with a specially developed grassland grazing ban management app installed. This app can receive alarm information sent by the alarm module and display information such as a map of the grassland area, the real-time location of the drone, and historical alarm records, making it convenient for management personnel to understand the situation and take timely action.

[0032] The specific steps for grassland grazing ban methods based on the above system are as follows: Data Acquisition: Staff use Mission Planner software at the ground control station to plan flight routes for the DJI Matrice 300RTK drone, which then flies along the preset routes. During flight, the drone's 20-megapixel high-definition camera collects images and video data of the grassland area. In nighttime or low-light environments, an optional FLIRVue Pro R infrared camera is used for data acquisition; if grassland vegetation conditions need to be analyzed, a Parrot Sequoia multispectral camera can be used. The GPS positioning module acquires the drone's location information in real time, and the wireless communication module transmits the collected image and video data, along with the drone's location information, to the ground control station.

[0033] Image Recognition: The ground control station transmits received image and video data to the NVIDIA DGX Station server of the AI ​​recognition module. The AI ​​recognition module first preprocesses the images, removing noise through noise reduction algorithms and improving image quality through contrast enhancement algorithms. Then, the YOLO object detection algorithm is used to detect and locate animals such as cattle and sheep in the preprocessed images, and the number of animals is counted. Finally, based on the detection results, it determines whether there is illegal grazing. If cattle and sheep are detected in a prohibited grazing area and their numbers exceed a user-defined threshold (e.g., 5 animals / animal), it is determined to be illegal grazing.

[0034] Alarm generation: When the AI ​​recognition module determines that there is grazing theft, it transmits the relevant information to the Raspberry Pi 4 development board of the alarm module. The Raspberry Pi 4 development board generates alarm information based on the location, time, and images of the grazing theft.

[0035] Alarm Push Notification: The alarm module pushes alarm information to the grassland grazing ban management APP on the Huawei Mate 60 Pro mobile phone of the management personnel via the Huawei MH5000-31 5G module. After receiving the alarm information, the management personnel can view the specific location, time and images of the illegal grazing through the APP, promptly go to the grazing site to stop it, and record the handling situation on the APP.

[0036] By adopting the above systems and methods, efficient and precise grazing ban management of grasslands can be achieved, greatly improving the efficiency and accuracy of grassland grazing ban management and effectively reducing the damage to grasslands caused by illegal grazing.

[0037] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A grassland grazing ban system based on unmanned aerial vehicles (UAVs), characterized in that, This includes drones, ground control stations, AI recognition modules, communication modules, alarm modules, and maintenance personnel terminals; The drone is equipped with a high-definition camera, a GPS positioning module, and a wireless communication module. The high-definition camera is used to collect images and video data of the grassland area, the GPS positioning module is used to obtain the drone's location information, and the wireless communication module is used to transmit data with the ground control station and the communication module. The ground control station is used to control the flight of the drone, receive image and video data transmitted by the drone, and transmit them to the AI ​​recognition module. The AI ​​recognition module receives image and video data transmitted from the ground control station and uses AI algorithms to identify poaching animals in the images; The communication module is used to enable data communication between the drone, ground control station, AI recognition module, alarm module and maintenance personnel terminal; The alarm module generates an alarm message when the AI ​​recognition module detects that the animal is illegally grazing. The maintenance personnel terminal is used to receive alarm information sent by the alarm module.

2. The grassland grazing ban system based on unmanned aerial vehicles as described in claim 1, characterized in that, The drone can fly according to a preset flight route or according to the instructions of the ground control station to conduct all-round, real-time patrols of the grassland.

3. A grassland grazing ban system based on unmanned aerial vehicles as described in claim 1, characterized in that, The ground control station is used to plan flight routes and send flight commands to control the flight status of the UAV.

4. A grassland grazing ban system based on unmanned aerial vehicles as described in claim 1, characterized in that, The AI ​​recognition module employs a deep learning algorithm, specifically a CNN convolutional neural network, which is trained on a large number of grassland images containing grazing animals to improve the accuracy and robustness of the recognition.

5. A grassland grazing ban system based on unmanned aerial vehicles as described in claim 1, characterized in that, The communication module employs one or more wireless communication technologies, such as 4G, 5G, and Wi-Fi, to achieve real-time data transmission and stable connection.

6. A grassland grazing ban system based on unmanned aerial vehicles as described in claim 1, characterized in that, The alarm information includes the location, time, and image-related information of the grazing theft.

7. A grassland grazing ban system based on unmanned aerial vehicles as described in claim 1, characterized in that, The terminal for the management personnel is a mobile phone or tablet computer, which can display a map of the grassland area, the real-time location of the drone, and historical alarm records.

8. A method for grassland grazing ban based on unmanned aerial vehicles (UAVs), characterized in that, Using the grazing prohibition system as described in any one of claims 1 to 7, comprising: Data acquisition: The UAV flies according to the preset flight route or according to the instructions of the ground control station, uses a high-definition camera to collect image and video data of the grassland area, and transmits the data to the ground control station through the wireless communication module; Image recognition: The ground control station transmits the collected images and video data to the AI ​​recognition module, which uses AI algorithms to identify the prohibited grazing animals in the images to determine whether there is any illegal grazing. Alarm generation: When the AI ​​recognition module detects grazing behavior, the alarm module generates alarm information based on the location, time, and images of the grazing. Alarm push: The alarm module pushes alarm information to the maintenance personnel's terminal through the communication module, and the maintenance personnel will take action after receiving the alarm information.

9. A method for grassland grazing ban based on unmanned aerial vehicles as described in claim 8, characterized in that, The high-definition camera can be equipped with an infrared camera or a multispectral camera. The infrared camera is suitable for data acquisition at night or in low-light environments, while the multispectral camera can be used to analyze grassland vegetation conditions.

10. A method for grassland grazing ban based on unmanned aerial vehicles as described in claim 8, characterized in that, The AI ​​recognition module preprocesses the image and uses a target detection algorithm to detect and locate the animals in the prohibited grazing area. Based on the detection results, it determines whether there is illegal grazing. If cattle, sheep, or other animals are detected in the prohibited grazing area and their numbers exceed a custom threshold, it is determined to be illegal grazing. The target detection algorithm is either You Only Look Once or Faster R-CNN.

Images