Unmanned aerial vehicle forest fire prevention smoke intelligent early warning method, device, equipment and medium

Abstract

The invention provides an unmanned aerial vehicle forest fire prevention smoke intelligent early warning method, device and equipment and a medium, and the method comprises the steps: detecting a smoke concentration value in a target inspection region at a current moment, when the smoke concentration value is greater than a preset concentration value, acquiring a multi-frame image sequence in a first preset time period and acquiring a multi-frame image sequence in a second preset time period; smog feature analysis is carried out on the multi-frame image sequence in the first preset time period and the multi-frame image sequence in the second preset time period to obtain smog multi-dimensional features; a thermal imaging data sequence in a second preset time period is collected, and flame multi-dimensional features are extracted; based on the smog multi-dimensional features and the flame multi-dimensional features, a fireproof early warning grade is judged; if the fireproof early warning level is greater than the preset early warning level, fusing the fire source navigation positioning information and the fire source dead reckoning information to obtain a fire source position coordinate; and generating a fireproof early warning prompt based on the fire source position coordinates, and sending the fireproof early warning prompt to a ground control center. Therefore, the forest fire prevention efficiency is effectively improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of fire warning, and in particular, to a UAV forest fire prevention smoke intelligent warning method, device, equipment and medium. BACKGROUND

[0002] Forest fires not only destroy the ecological environment, but also cause serious economic losses. Through timely forest fire detection, measures can be taken to control the fire and reduce the damage to forest resources, thereby protecting biodiversity and maintaining ecological balance. In related technologies, fire detection and warning mainly rely on personnel patrol, that is, corresponding patrol personnel are assigned to different areas for regular patrol, and if a fire is found during the patrol, it will be notified.

[0003] However, the forest fire prevention efficiency is not high using the prior art. SUMMARY

[0004] The embodiments described herein provide a UAV forest fire prevention smoke intelligent warning method, device, equipment and medium, which overcome the above problems.

[0005] In a first aspect, according to the content of the present disclosure, a UAV forest fire prevention smoke intelligent warning method is provided, applied to a UAV device, wherein the UAV device is equipped with a smoke sensor, a high-definition visible light camera, a thermal imaging camera, a navigation positioning sensor and an inertial navigation system; the method comprises:

[0006] During UAV inspection of a forest inspection area, the smoke sensor detects the smoke concentration value in the target inspection area at the current time, and when the smoke concentration value is greater than a preset concentration value, a plurality of image sequences of the target inspection area in a first preset time period are acquired by the high-definition visible light camera, and a plurality of image sequences of the target inspection area in a second preset time period are acquired by the high-definition visible light camera, the first preset time period being a time period before the current time, and the second preset time period being a time period after the current time;

[0007] The plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period are subjected to smoke feature analysis to obtain a smoke multi-dimensional feature corresponding to the target inspection area;

[0008] acquire a thermal imaging data sequence of the target inspection area within a second preset time period through the thermal imaging camera, and identify whether a fire point heat signal appears in the target inspection area based on the thermal imaging data sequence; if the fire point heat signal appears in the target inspection area, extract a flame multi-dimensional feature corresponding to the target inspection area from the thermal imaging data sequence;

[0009] determine a fire prevention warning level corresponding to the target inspection area based on the smoke multi-dimensional feature and the flame multi-dimensional feature corresponding to the target inspection area; if the fire prevention warning level is greater than a preset warning level, locate a fire source in the target inspection area through the navigation positioning sensor to obtain fire source navigation positioning information, and locate the fire source in the target inspection area through the inertial navigation system to obtain fire source dead reckoning information;

[0010] fuse the fire source navigation positioning information and the fire source dead reckoning information to obtain fire source position coordinates;

[0011] generate a fire prevention warning prompt based on the fire source position coordinates, and send the fire prevention warning prompt to a ground control center, so that the ground control center performs fire prevention treatment at the fire source position coordinates through a fire prevention warning device.

[0012] In a second aspect, according to the content of the present disclosure, an unmanned aerial vehicle forest fire prevention smoke intelligent warning device is provided, which is applied to an unmanned aerial vehicle device, and the unmanned aerial vehicle device is loaded with a smoke sensor, a high-definition visible light camera, a thermal imaging camera, a navigation positioning sensor and an inertial navigation system; the device comprises:

[0013] The detection module is configured to, during unmanned aerial vehicle inspection of a forest inspection area, detect a smoke concentration value in a target inspection area at a current time through the smoke sensor, and when the smoke concentration value is greater than a preset concentration value, acquire a plurality of image sequences of the target inspection area within a first preset time period collected by the high-definition visible light camera, and acquire a plurality of image sequences of the target inspection area within a second preset time period through the high-definition visible light camera, the first preset time period being a time period before the current time, and the second preset time period being a time period after the current time.

[0014] The analysis module is configured to perform smoke feature analysis on the plurality of image sequences of the target inspection area within the first preset time period and the plurality of image sequences of the target inspection area within the second preset time period, and obtain a smoke multi-dimensional feature corresponding to the target inspection area.

[0015] The collection module is configured to collect, by the thermal imaging camera, a thermal imaging data sequence of the target inspection area in a second preset time period, and identify whether a fire point thermal signal appears in the target inspection area based on the thermal imaging data sequence; if the fire point thermal signal appears in the target inspection area, extract a flame multi-dimensional feature corresponding to the target inspection area from the thermal imaging data sequence;

[0016] The judgment module is configured to judge a fire prevention warning level corresponding to the target inspection area based on the smoke multi-dimensional feature and the flame multi-dimensional feature corresponding to the target inspection area; if the fire prevention warning level is greater than a preset warning level, locate a fire source in the target inspection area by the navigation positioning sensor to obtain fire source navigation positioning information, and locate the fire source in the target inspection area by the inertial navigation system to obtain fire source dead reckoning information.

[0017] The fusion module is configured to fuse the fire source navigation positioning information and the fire source dead reckoning information to obtain fire source position coordinates.

[0018] The generation module is configured to generate a fire prevention warning prompt based on the fire source position coordinates, and send the fire prevention warning prompt to a ground control center, so that the ground control center performs fire prevention treatment on the fire source position coordinates by a fire prevention warning device.

[0019] In a third aspect, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the unmanned aerial vehicle forest fire smoke intelligent warning method in any one of the above embodiments when executing the computer program.

[0020] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the processor implements the steps of the unmanned aerial vehicle forest fire smoke intelligent warning method in any one of the above embodiments when executing the computer program.

[0021] The unmanned aerial vehicle forest fire prevention smoke intelligent early warning method provided by the embodiments of the application can detect the smoke concentration value in the target inspection area at the current time through a smoke sensor during unmanned aerial vehicle inspection of the forest inspection area, and when the smoke concentration value is greater than a preset concentration value, a plurality of image sequences of the target inspection area in a first preset time period are acquired by a high-definition visible light camera, and a plurality of image sequences of the target inspection area in a second preset time period are acquired by the high-definition visible light camera, the first preset time period is a time period before the current time, and the second preset time period is a time period after the current time; smoke feature analysis is performed on the plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection area; a thermal imaging data sequence of the target inspection area in the second preset time period is acquired by a thermal imaging camera, and whether a fire point thermal signal appears in the target inspection area is identified based on the thermal imaging data sequence; if the fire point thermal signal appears in the target inspection area, flame multi-dimensional features corresponding to the target inspection area are extracted from the thermal imaging data sequence; a fire prevention early warning level corresponding to the target inspection area is determined based on the smoke multi-dimensional features and the flame multi-dimensional features corresponding to the target inspection area; if the fire prevention early warning level is greater than a preset early warning level, a fire source in the target inspection area is positioned by a navigation positioning sensor to obtain fire source navigation positioning information, and the fire source in the target inspection area is positioned by an inertial navigation system to obtain fire source dead reckoning information; the fire source position coordinates are obtained by fusing the fire source navigation positioning information and the fire source dead reckoning information; a fire prevention early warning prompt is generated based on the fire source position coordinates, and the fire prevention early warning prompt is sent to a ground control center, so that the ground control center performs fire prevention treatment at the fire source position coordinates by a fire prevention early warning device. In this way, the unmanned aerial vehicle is equipped with a high-definition camera and various advanced sensors, and can quickly discover smoke and fire sources in combination with feature recognition, realize early warning and rapid response, and effectively improve the efficiency and accuracy of forest fire prevention.

[0022] The above description is only a summary of the technical solutions of the embodiments of the application. In order to more clearly understand the technical means of the embodiments of the application, the embodiments of the application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the embodiments of the application more obvious and easy to understand, the specific embodiments of the application are described below. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly described below. It should be noted that the drawings described below only relate to some embodiments of the present disclosure, but not limit the present disclosure, wherein:

[0024] Figure 1 is a flowchart of an unmanned aerial vehicle forest fire prevention smoke intelligent early warning method provided by the present disclosure.

[0025] Figure 2 is a structural schematic diagram of a forest fire smoke intelligent early warning device of the present disclosure.

[0026] Figure 3 is a structural schematic diagram of a computer device of the present disclosure.

[0027] It should be noted that the elements in the drawings are schematic and not drawn to scale. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions of the embodiments of the present disclosure will be described clearly and completely below in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by a person skilled in the art without any inventive effort also belong to the scope of protection of the present disclosure.

[0029] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this present subject matter belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the specification and relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein. As used herein, the statement that two or more parts are "connected" or "coupled" together refer to an indirect or direct connection or coupling.

[0030] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase "in an embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. A person of ordinary skill in the art will readily recognize from the disclosure herein, given the total volume of this application that one or more passages that are described as an embodiment is / are also an embodiment of another embodiment.

[0031] The term "and / or", merely describes an associated relationship of associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of existence of A, existence of A and B, and existence of B. In addition, the character " / " herein generally represents that the front and rear associated objects are in an "or" relationship. Terms such as "first" and "second" are merely used to distinguish one component (or part of a component) from another component (or another part of a component).

[0032] In the description of the present application, unless otherwise specified, the meaning of "a plurality of" refers to more than two (including two), and similarly, "a plurality of groups" refers to more than two groups (including two groups).

[0033] In order for those skilled in the art to better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings.

[0034] Figure 1 is a flowchart of a forest fire smoke intelligent early warning method provided by an embodiment of the present disclosure. The forest fire smoke intelligent early warning method is applied to a UAV device. A patrol pod is mounted on the UAV device. The patrol pod integrates a smoke sensor, a high-definition visible light camera, a thermal imaging camera, a navigation positioning sensor, and an inertial navigation system. The UAV platform has a safe flight control capability, high endurance capability, and stable flight performance. It can adapt to complex terrain and harsh weather conditions, and can stably operate in an extreme environment of -20°C to 45°C to ensure the smooth execution of the forest fire smoke intelligent early warning patrol task. It supports intelligent path planning and autonomous flight. The UAV can flexibly fly and cover a large area of forest area, especially some remote areas that are difficult to reach, and can ensure the comprehensiveness of fire monitoring.

[0035] As shown in Figure 1 , the specific process of the forest fire smoke intelligent early warning method includes:

[0036] S110, in the process of UAV patrol in the forest patrol area, detecting the smoke concentration value in the target patrol area at the current time through the smoke sensor, and when the smoke concentration value is greater than the preset concentration value, acquiring a plurality of image sequences of the target patrol area in a first preset time period collected by the high-definition visible light camera, and acquiring a plurality of image sequences of the target patrol area in a second preset time period collected by the high-definition visible light camera.

[0037] Among them, the first preset time period is a historical time period before the current time, and the second preset time period is a future time period after the current time. For example, if the current time is 13:00 on the same day, the first preset time period can be 12:30-12:59, and the second preset time period can be 13:01-13:30.

[0038] The smoke sensor can monitor the smoke concentration in the air in real time. Once the smoke concentration is detected to be abnormally high, an alarm is immediately sent out. The high-definition visible light camera can collect real-time images, especially for the smoke alarm forest area. The UAV carries a high-resolution camera and a thermal imager to patrol the forest area in real time, collect high-definition images and video data; during the collection process, the high-resolution camera captures multiple images per second to ensure that it can capture the subtle changes of the smoke.

[0039] S120, smoke feature analysis is performed on the multi-frame image sequence of the target inspection area within the first preset time period and the multi-frame image sequence of the target inspection area within the second preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection area.

[0040] The thermal imaging camera can collect thermal imaging data of the scene in real time, especially for the smoke alarm forest area.

[0041] In some embodiments, the smoke feature analysis on the multi-frame image sequence of the target inspection area within the first preset time period and the multi-frame image sequence of the target inspection area within the second preset time period to obtain smoke multi-dimensional features corresponding to the target inspection area includes:

[0042] The multi-frame image sequence of the target inspection area within the first preset time period and the multi-frame image sequence of the target inspection area within the second preset time period are respectively subjected to image enhancement processing; the smoke change degree of the multi-frame image sequence of the target inspection area within the second preset time period corresponding to the multi-frame image sequence within the first preset time period is determined; if the smoke change degree is greater than or equal to a preset degree threshold, smoke feature extraction is performed on the multi-frame image sequence of the target inspection area within the second preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection area; if the smoke change degree is less than the preset degree threshold, smoke feature extraction is performed on the multi-frame image sequence of the target inspection area within the first preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection area.

[0043] The smoke multi-dimensional features corresponding to the target inspection area include smoke color features, smoke texture features, and smoke shape features. The smoke color features are, for example, gray or white; the smoke texture features are, for example, the direction of dispersion; and the smoke shape features are, for example, the shape boundary and the smoke concentration.

[0044] Deep learning algorithms can be used to automatically extract features and accurately identify them by learning a large amount of smoke image data, to perform smoke feature extraction on the multi-frame image sequence of the target inspection area within the first preset time period / second preset time period. Smoke usually appears as a gray or white hazy area, and deep learning algorithms can distinguish smoke from background interference (such as clouds and fog) through convolutional neural networks, to reduce false positives. Image enhancement can include denoising, contrast enhancement, color correction, and other operations, to improve the recognizability of smoke.

[0045] S130, a thermal imaging data sequence of the target inspection area within the second preset time period is collected by the thermal imaging camera, and whether a fire point thermal signal appears in the target inspection area is identified based on the thermal imaging data sequence; if a fire point thermal signal appears in the target inspection area, flame multi-dimensional features corresponding to the target inspection area are extracted from the thermal imaging data sequence.

[0046] The thermal imaging camera can identify fire ignition signals under smoke or low-light conditions. If a fire ignition signal appears in any image of the thermal imaging data sequence, then a fire ignition signal is determined to exist within the target inspection area.

[0047] In some embodiments, multidimensional flame features corresponding to the target inspection area are extracted from the thermal imaging data sequence, including:

[0048] Image enhancement processing is performed on the thermal imaging data sequence, and background images and flame images in the thermal imaging data sequence are identified by a neural network segmentation model; features are extracted from the flame images in the thermal imaging data sequence by a neural network extraction model to obtain multi-dimensional flame features corresponding to the target inspection area.

[0049] The multi-dimensional flame features corresponding to the target inspection area include: flame color features, flame texture features, and flame shape features. Flame color features include red or orange dots; flame shape features include the shape of the flame.

[0050] Deep learning algorithms can be used to learn from a large amount of flame image data, automatically extracting its features and accurately identifying it. The fire source appears as a distinct red or orange dot; deep learning algorithms use convolutional neural networks to distinguish flames from background interference, reducing false alarms. Image enhancement can include operations such as noise reduction, contrast enhancement, and color correction, thereby improving the flame's recognizability.

[0051] S140. Based on the multi-dimensional smoke and flame characteristics corresponding to the target inspection area, determine the fire warning level corresponding to the target inspection area; if the fire warning level is greater than the preset warning level, locate the fire source in the target inspection area through the navigation and positioning sensor to obtain the fire source navigation and positioning information, and locate the fire source in the target inspection area through the inertial navigation system to obtain the fire source dead reckoning information.

[0052] This involves employing visual saliency detection algorithms to quickly locate potential fire or smoke sources by analyzing salient areas in images, enabling precise detection and localization of smoke and fire sources. While maintaining high accuracy, it also allows for rapid processing of image data, enabling real-time monitoring on the UAV platform. Specifically, it combines the UAV's GNSS (Navigation and Positioning Sensor) and inertial navigation system to fuse positioning data, allowing for precise location of fault points.

[0053] S150, by integrating fire source navigation and positioning information and fire source dead reckoning information, the coordinates of the fire source location are obtained.

[0054] The unmanned aerial vehicle is equipped with a centimeter-level positioning system, adopts a fire point accurate positioning technology, combines GNSS, inertial navigation system (INS) sensors and other multi-source navigation fusion technologies, improves the positioning accuracy of the unmanned aerial vehicle, realizes sub-meter accuracy in a large range and centimeter-level accurate positioning of a fault point in a key area, fuses satellite positioning data and inertial sensor data through a tight coupling algorithm, realizes high-precision positioning in a complex environment, and accurately determines the fault position. Moreover, the collected image can be matched with a pre-constructed map to display the fault position in a visual manner, so as to facilitate forest fire prevention emergency rescue workers to quickly locate and handle the fire.

[0055] In some embodiments, the fire source navigation positioning information and the fire source dead reckoning information are fused to obtain fire source position coordinates, including:

[0056] The initial heading deviation of the inertial navigation system is used to coarsely fuse the fire source navigation positioning information and the fire source dead reckoning information to obtain a first fusion result; the carrier phase measurement value of the navigation positioning sensor is used to accurately fuse the fire source navigation positioning information and the fire source dead reckoning information to obtain a second fusion result; and the fire source position coordinates are determined based on the first fusion result and the second fusion result.

[0057] In the method, the GNSS positioning result (i.e., the fire source navigation positioning information) is loosely combined with the dead reckoning result of the inertial navigation system (MEMS INS) (i.e., the fire source dead reckoning information); the initial heading deviation of the inertial navigation is calculated by comparing the displacement vectors of the GNSS trajectory and the INS trajectory, so that rapid and accurate initial alignment is realized. The carrier phase measurement value of the GNSS receiver and the MEMS INS data are used for deep fusion to realize tight combination, so that even in the case that the number of available satellites is less than 4, the heading initial alignment can be realized through the carrier phase change amount. In the loose combination framework, the initial heading deviation of the inertial navigation is calculated by analyzing the displacement vector during the starting process of the unmanned aerial vehicle, and in the tight combination framework, the carrier phase measurement value is used for rapid alignment; the multi-scene adaptability is improved, and it is ensured that rapid and accurate initial alignment can be realized in open sky and urban environment.

[0058] Therefore, the high-precision positioning of the GNSS and the short-time high-dynamic characteristics of the MEMS INS are combined for multi-source data fusion to make up for the deficiency of the GNSS in signal shielding or interruption; the full-system full-band double-antenna GNSS module is used to enhance the anti-interference ability and ensure the positioning accuracy in a complex environment.

[0059] It should be noted that the embodiment can also perform multi-modal data fusion, combine multi-modal data such as visible light images and infrared images, and improve the detection capability; considering that single visual information may not be sufficient in a complex environment, the infrared data can provide additional temperature information.

[0060] For example, data of different modalities are acquired from various sensors, including visible light images, infrared thermal imaging images, meteorological data, etc.; data preprocessing can standardize and align data of different modalities, and image data is resized and normalized, and time series data needs to be denoised and normalized. Multi-modal fusion of three levels of data level fusion, feature level fusion and decision level fusion is adopted. For data level fusion: the original data is directly processed, and the original data of the sensor is filtered and denoised; by using Kalman filter smoothing, the data of visible light and infrared sensors can be reduced. By using weighted average method, the data of different sensors is weighted and summed, and the weight is adjusted according to the signal-to-noise ratio or historical data of the sensor. For feature level fusion: feature extraction is performed to extract key features such as image edges from sensor data; and feature combination is performed to reduce the redundancy of multi-dimensional sensor data and improve the calculation efficiency by using principal component analysis or linear discriminant analysis; a deep learning model is used to extract color, texture, shape and other features of smoke and flame in the image by using convolutional neural network, and to analyze time series data by using long short-term memory network to analyze the features of smoke such as dispersion direction, expansion range, color concentration and shape boundary, to help determine whether a fire occurs, and to realize more intelligent fire point detection. For decision level fusion: the final decision is made based on the independent detection results of each sensor, and the joint probability of the detection results of different sensors is calculated by Bayesian inference to determine whether there is an anomaly; D-S evidence theory can be used in high-uncertainty environment, and the comprehensive confidence is calculated by confidence distribution of different sensors. By fusing visible light images, thermal infrared images and other data, the shortcomings of single sensor data can be compensated, and the accuracy of fire detection can be improved.

[0061] S160, generate a fire prevention warning prompt based on the fire source position coordinates, and send the fire prevention warning prompt to the ground control center, so that the ground control center performs fire prevention treatment on the fire source position coordinates through a fire prevention warning device.

[0062] The unmanned aerial vehicle device can also be built with a fire warning system to generate a fire prevention warning prompt based on the coordinates of the fire source and send the fire prevention warning prompt to the ground control center. In addition, the unmanned aerial vehicle device can also be built with a data transmission communication and control system, which uses edge computing technology for data preprocessing and realizes real-time processing and analysis of the collected data, and transmits the collected data and identification results to the ground control center in real time for data storage and checking analysis. The 4G / 5G communication technology, wireless ad hoc network technology, etc. can be used to transmit the images, videos and sensor data collected by the unmanned aerial vehicle to the ground control center in real time. The low delay characteristics of the 4G / 5G network can ensure the real-time and stability of data transmission. The ad hoc network technology can form a wider monitoring network through the cooperation of multiple unmanned aerial vehicles in an area without mobile communication network, realizing efficient and reliable communication of data. Once smoke or fire is detected, the data can be transmitted to the ground control center in time. At the same time, after sending the warning, the embodiment will also automatically send a short message and a screenshot to the mobile device of the responsible personnel, facilitating the responsible personnel to check in time.

[0063] In the present embodiment, during the UAV inspection of the forest inspection area, the smoke concentration value in the target inspection area at the current time is detected by the smoke sensor, and when the smoke concentration value is greater than the preset concentration value, a plurality of image sequences of the target inspection area in a first preset time period are acquired by the high-definition visible light camera, and a plurality of image sequences of the target inspection area in a second preset time period are acquired by the high-definition visible light camera, the first preset time period being a time period before the current time, and the second preset time period being a time period after the current time; smoke feature analysis is performed on the plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period to obtain the smoke multi-dimensional feature corresponding to the target inspection area; a thermal imaging data sequence of the target inspection area in the second preset time period is acquired by the thermal imaging camera, and whether a fire point thermal signal appears in the target inspection area is identified based on the thermal imaging data sequence; if the fire point thermal signal appears in the target inspection area, the flame multi-dimensional feature corresponding to the target inspection area is extracted from the thermal imaging data sequence; the fire warning level corresponding to the target inspection area is determined based on the smoke multi-dimensional feature and the flame multi-dimensional feature corresponding to the target inspection area; if the fire warning level is greater than the preset warning level, the fire source in the target inspection area is positioned by the navigation positioning sensor to obtain fire source navigation positioning information, and the fire source in the target inspection area is positioned by the inertial navigation system to obtain fire source dead reckoning information; the fire source position coordinates are obtained by fusing the fire source navigation positioning information and the fire source dead reckoning information; the fire warning prompt is generated based on the fire source position coordinates, and the fire warning prompt is sent to the ground control center, so that the ground control center performs fire prevention treatment at the fire source position coordinates by the fire warning device. In this way, by using the UAV to carry the high-definition camera and various advanced sensors, combined with feature recognition, the smoke and fire source can be quickly found, early warning and rapid response can be realized, and the efficiency and accuracy of forest fire prevention can be effectively improved.

[0064] In some embodiments, the method of the present embodiment further comprises:

[0065] The vegetation humidity at the fire source position coordinates and the meteorological data at the current time are acquired; the fire spread information at the fire source position coordinates is predicted based on the vegetation humidity at the fire source position coordinates and the meteorological data at the current time; and the fire spread information at the fire source position coordinates is sent to the ground control center.

[0066] The meteorological data can be, for example, temperature, wind speed, etc. The fire spread information can be, for example, fire spread direction, fire spread speed, etc. By predicting the fire spread information at the fire source position coordinates based on the vegetation humidity at the fire source position coordinates and the meteorological data at the current time, and sending it to the ground control center, the ground control center can take appropriate measures.

[0067] In some embodiments, the method of the present embodiment further comprises:

[0068] Visible light image data in the forest inspection area is collected by the high-definition visible light camera, and whether there is a preset sensitive behavior in the forest inspection area is determined based on the visible light image data in the forest inspection area; if it is determined based on the visible light image data in the forest inspection area that there is a preset sensitive behavior in the forest inspection area, a warning prompt voice is generated, and the warning prompt voice is played in the corresponding airspace of the preset sensitive behavior.

[0069] The preset sensitive behavior can be, for example, a rule-violating fire use behavior. In this embodiment, the rule-violating fire use behavior is discouraged by high-altitude shouting, and an electronic file of the whole process of fire disposition can be formed to realize precise tracing and provide a scientific basis for fire research and disposition.

[0070] In addition, a real-time monitoring control system can also be built in the unmanned aerial vehicle device to realize data reception, processing, analysis and storage, and the state of the unmanned aerial vehicle and the inspection data of forest fire prevention can be monitored in real time, AI image recognition fire diagnosis and analysis functions are provided to determine the fire location and smoke and fire state. At the same time, the ground control center can also control the flight and shooting of the unmanned aerial vehicle through remote instructions, and receive and process the data information transmitted by the unmanned aerial vehicle forest fire prevention smoke intelligent early warning platform.

[0071] In some embodiments, the method of the present embodiment further comprises:

[0072] The flight route in the forest inspection area is planned based on the inspection requirements, and the forest inspection area is inspected by the unmanned aerial vehicle according to the flight route in the forest inspection area. After the unmanned aerial vehicle inspection of the forest inspection area is completed, a regional inspection report corresponding to the forest inspection area is generated, and the regional inspection report corresponding to the forest inspection area is sent to the ground control center.

[0073] The unmanned aerial vehicle device can also be built with a patrol task management system, which can remotely plan and real-time control the cruise task through the background, and the patrol task management system can store and analyze the fire information in real time and trigger an alarm. Through the PC platform client or mobile terminal application of the ground control center, the creation of the patrol task, the flight path planning, the task allocation and the execution monitoring can be realized; the automatic flight path planning and the manual flight path adjustment can be performed. After the completion of the patrol task, the system automatically generates a patrol report containing the fire information, the smoke type, the fire point position information, the weather state and the like. Specifically, according to the geographical information and the patrol requirements of the forest fire patrol area, the flight path of the unmanned aerial vehicle is automatically or manually planned by the system. Automatic patrol: the unmanned aerial vehicle automatically flies according to the preset flight path, and collects the images and sensor data collected in the forest fire patrol process in real time; during the flight, the system monitors the state of the unmanned aerial vehicle in real time, such as the battery capacity, the flight position, the flight speed and the meteorological environmental parameters such as the temperature, the humidity and the wind direction. Data collection and transmission: the images and sensor data collected by the unmanned aerial vehicle are transmitted to the ground control center in real time through the 4G / 5G network or the wireless ad hoc network communication technology; at the same time, the data can also be stored in the local storage device of the unmanned aerial vehicle, and uploaded after the task is completed. Task management: the PC platform client or the mobile terminal application, the operation and maintenance personnel can create the patrol task, plan the flight path, allocate the task to the unmanned aerial vehicle, and monitor the task execution in real time, and once the smoke or flame is detected, the system immediately issues an alarm and notifies the relevant personnel through the short message, the pop-up window and the like. Report generation: after the completion of the patrol task, the system generates a patrol report according to the AI image recognition result and the positioning data, and the report content includes the fire point image position and information, the detailed defect list, the maintenance suggestion and the like, which provides a basis for the emergency team to plan the optimal extinguishing path.

[0074] In summary, the unmanned aerial vehicle forest fire smoke intelligent early warning method provided by the embodiment can quickly discover smoke and fire sources, prevent the spread of fire, and is particularly suitable for high-frequency monitoring in high-incidence areas of forest fires, such as forests in dry seasons and forests near residential areas, which is conducive to early warning of forest fires; after the occurrence of a fire, the system can monitor the changes in forest fire prevention, assist in the development of emergency rescue and fire extinguishing work, dispatch the unmanned aerial vehicle to carry out stereoscopic reconnaissance, use high-altitude shouting to discourage illegal use of fire, form an electronic file of the whole process of fire disposal, realize accurate tracing, and provide a scientific basis for fire research and disposal. Through the self-organizing network communication of multiple unmanned aerial vehicles, multiple unmanned aerial vehicles can work together to form a wider monitoring network, and by reasonably planning the fire extinguishing path, the unmanned aerial vehicle can carry fire extinguishing bombs to extinguish the initial fire, realize the effect of "early and small" of early discovery and early elimination of fire, and compared with the traditional monitoring means, the deployment and maintenance cost of the unmanned aerial vehicle and the AI system is lower.

[0075] Figure 2 A structure diagram of a forest fire prevention smoke intelligent early warning device of a UAV is provided for the present embodiment. The forest fire prevention smoke intelligent early warning device is applied to a UAV device. The UAV device is provided with a smoke sensor, a high-definition visible light camera, a thermal imaging camera, a navigation positioning sensor and an inertial navigation system. The forest fire prevention smoke intelligent early warning device can include:

[0076] The detection module 210 is configured to detect, during UAV inspection of a forest inspection area, a smoke concentration value in a target inspection area at a current time through the smoke sensor, and when the smoke concentration value is greater than a preset concentration value, acquire a plurality of image sequences of the target inspection area in a first preset time period collected by the high-definition visible light camera, and acquire a plurality of image sequences of the target inspection area in a second preset time period collected by the high-definition visible light camera. The first preset time period is a time period before the current time, and the second preset time period is a time period after the current time.

[0077] The analysis module 220 is configured to perform smoke feature analysis on the plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection area.

[0078] The acquisition module 230 is configured to acquire a thermal imaging data sequence of the target inspection area in the second preset time period through the thermal imaging camera, and identify whether a fire point thermal signal appears in the target inspection area based on the thermal imaging data sequence; if the fire point thermal signal appears in the target inspection area, extract flame multi-dimensional features corresponding to the target inspection area from the thermal imaging data sequence.

[0079] The judgment module 240 is configured to judge a fire prevention warning level corresponding to the target inspection area based on the smoke multi-dimensional features and the flame multi-dimensional features corresponding to the target inspection area; if the fire prevention warning level is greater than a preset warning level, locate a fire source in the target inspection area through the navigation positioning sensor to obtain fire source navigation positioning information, and locate the fire source in the target inspection area through the inertial navigation system to obtain fire source dead reckoning information.

[0080] The fusion module 250 is configured to fuse the fire source navigation positioning information and the fire source dead reckoning information to obtain fire source position coordinates.

[0081] The generation module 260 is configured to generate a fire prevention warning prompt based on the fire source position coordinates, and send the fire prevention warning prompt to a ground control center, so that the ground control center performs fire prevention treatment at the fire source position coordinates through a fire prevention warning device.

[0082] In the present embodiment, the analysis module 220 can be configured to:

[0083] respectively, on the multi-frame image sequence of the target inspection region in the first preset time period and the multi-frame image sequence of the target inspection region in the second preset time period are subjected to image enhancement processing; and a smoke change degree of the multi-frame image sequence of the target inspection region in the second preset time period corresponding to the multi-frame image sequence in the first preset time period is determined; if the smoke change degree is greater than or equal to a preset degree threshold, smoke feature extraction is performed on the multi-frame image sequence of the target inspection region in the second preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection region; if the smoke change degree is less than the preset degree threshold, smoke feature extraction is performed on the multi-frame image sequence of the target inspection region in the first preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection region; wherein the smoke multi-dimensional features corresponding to the target inspection region include smoke color features, smoke texture features and smoke shape features.

[0084] In this embodiment, optionally, the collection module 230 is specifically configured to:

[0085] The thermal imaging data sequence is subjected to image enhancement processing, and the background image and the flame image in the thermal imaging data sequence are identified through the neural network segmentation model; the flame multi-dimensional features corresponding to the target inspection region are obtained by performing feature extraction on the flame image in the thermal imaging data sequence through the neural network extraction model; wherein the flame multi-dimensional features corresponding to the target inspection region include flame color features, flame texture features and flame shape features.

[0086] In this embodiment, optionally, the fusion module 250 is specifically configured to:

[0087] The initial heading deviation of the inertial navigation system is used to coarsely fuse the fire source navigation positioning information and the fire source dead reckoning information, to obtain a first fusion result; the carrier phase measurement value of the navigation positioning sensor is used to accurately fuse the fire source navigation positioning information and the fire source dead reckoning information, to obtain a second fusion result; and the fire source position coordinates are determined based on the first fusion result and the second fusion result.

[0088] In this embodiment, optionally, the first processing module is further included.

[0089] The first processing module is configured to acquire the vegetation humidity at the fire source position coordinates and the meteorological data at the current moment; predict the fire spread information at the fire source position coordinates based on the vegetation humidity at the fire source position coordinates and the meteorological data at the current moment; and send the fire spread information at the fire source position coordinates to the ground control center.

[0090] In this embodiment, optionally, the second processing module is further included.

[0091] The second processing module is configured to collect visible light image data in the forest inspection area by the high-definition visible light camera, and determine whether a preset sensitive behavior exists in the forest inspection area based on the visible light image data in the forest inspection area; if it is determined that the preset sensitive behavior exists in the forest inspection area based on the visible light image data in the forest inspection area, a warning prompt voice is generated, and the warning prompt voice is played in a corresponding airspace of the preset sensitive behavior.

[0092] In the embodiment, the method further comprises a third processing module.

[0093] The flight route in the forest inspection area is planned based on the inspection requirement, the forest inspection area is inspected by the unmanned aerial vehicle according to the flight route in the forest inspection area, the corresponding regional inspection report of the forest inspection area is generated after the unmanned aerial vehicle inspection of the forest inspection area is completed, and the corresponding regional inspection report of the forest inspection area is sent to the ground control center.

[0094] The unmanned aerial vehicle forest fire prevention smoke intelligent early warning device provided by the present disclosure can execute the above method embodiments, and the specific implementation principles and technical effects can be referred to the above method embodiments, which will not be described here again.

[0095] The present application also provides a computer device. For details, please refer to Figure 3 , Figure 3 The present application also provides a computer device. For details, please refer to

[0096] The computer device includes a memory 310 and a processor 320 which are connected to each other for communication through a system bus. It should be noted that only the computer device with the memory 310 and the processor 320 is shown in the figure, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented. Among them, those skilled in the art can understand that the computer device herein is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0097] The computer device can be a desktop computer, a notebook computer, a palm computer, and a cloud server, etc. The computer device can interact with the user through a keyboard, a mouse, a remote controller, a touchpad, a voice control device, etc.

[0098] The memory 310 includes at least one type of readable storage medium, including non-volatile memory or volatile memory, for example, flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. The RAM can include static RAM or dynamic RAM. In some embodiments, the memory 310 can be an internal storage unit of the computer device, for example, a hard disk or a memory of the computer device. In other embodiments, the memory 310 can also be an external storage device of the computer device, for example, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, or a flash card, etc. equipped on the computer device. Of course, the memory 310 can include both an internal storage unit and an external storage device of the computer device. In the present embodiment, the memory 310 is generally used to store an operating system and various application software installed on the computer device, for example, program codes of the above-described method, etc. In addition, the memory 310 can also be used to temporarily store various data that has been output or will be output.

[0099] The processor 320 is generally used to perform the overall operation of the computer device. In the present embodiment, the memory 310 is used to store program codes or instructions, which include computer operation instructions, and the processor 320 is used to execute the program codes or instructions stored in the memory 310 or process data, for example, run the program codes of the above-described method.

[0100] In this article, the bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus system can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or only one type of bus.

[0101] Another embodiment of the present application also provides a computer readable medium, which can be a computer readable signal medium or a computer readable medium. The processor in the computer reads the computer readable program code stored in the computer readable medium, so that the processor can perform the function actions specified in each step or combination of steps in the above method; generate the device implementing the function actions specified in each block or combination of blocks in the block diagram.

[0102] The computer readable medium includes but is not limited to electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any appropriate combination of the foregoing, for storing program codes or instructions, which include computer operation instructions, and processors for executing the program codes or instructions of the above method stored in the memory.

[0103] The definition of the memory and the processor can refer to the description of the foregoing computer device embodiment, which will not be repeated here.

[0104] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiment described above is only schematic, for example, the division of modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0105] The function units or modules in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software function unit.

[0106] If the integrated unit is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0107] In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The application can be implemented by means of both hardware and software, and any combination thereof. In the device claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The mere fact that certain measures are recited in mutually different claims does not indicate that a combination of these measures cannot be used to advantage. The use of relative terms such as "first", "second" and "third", etc. does not connote any prioritization, but such terms are used to distinguish a certain feature from another feature with the same name. The steps of the above-described methods shall not be understood as necessarily limited to the order in which they are presented.

[0108] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; even though the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still make modifications to the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A forest fire prevention smoke intelligent early warning method for unmanned aerial vehicles, characterized in that, The method is applied to a UAV device, and the UAV device is provided with a smoke sensor, a high-definition visible light camera, a thermal imaging camera, a navigation positioning sensor and an inertial navigation system. In the process of UAV inspection of a forest inspection area, the smoke concentration value of a target inspection area at the current time is detected by the smoke sensor, and when the smoke concentration value is greater than a preset concentration value, a plurality of image sequences of the target inspection area in a first preset time period are acquired by the high-definition visible light camera, and a plurality of image sequences of the target inspection area in a second preset time period are acquired by the high-definition visible light camera, the first preset time period being a time period before the current time, and the second preset time period being a time period after the current time. Smoke feature analysis is performed on the plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period to obtain smoke multi-dimensional features corresponding to the target inspection area. Thermal imaging data sequences of the target inspection area in the second preset time period are acquired by the thermal imaging camera, and whether a fire point thermal signal appears in the target inspection area is identified based on the thermal imaging data sequences; if a fire point thermal signal appears in the target inspection area, flame multi-dimensional features corresponding to the target inspection area are extracted from the thermal imaging data sequences. Based on the smoke multi-dimensional features and the flame multi-dimensional features corresponding to the target inspection area, a fire prevention warning level corresponding to the target inspection area is determined; if the fire prevention warning level is greater than a preset warning level, a fire source in the target inspection area is located by the navigation positioning sensor to obtain fire source navigation positioning information, and the fire source in the target inspection area is located by the inertial navigation system to obtain fire source dead reckoning information. The fire source navigation positioning information and the fire source dead reckoning information are fused to obtain fire source position coordinates. A fire prevention warning prompt is generated based on the fire source position coordinates, and the fire prevention warning prompt is sent to a ground control center, so that the ground control center performs fire prevention treatment at the fire source position coordinates by a fire prevention warning device.

2. The method of claim 1, wherein, The smoke feature analysis on the plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period to obtain the smoke multi-dimensional features corresponding to the target inspection area comprises: Image enhancement processing is respectively performed on the plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period; and the smoke change degree of the plurality of image sequences of the target inspection area in the second preset time period corresponding to the plurality of image sequences of the target inspection area in the first preset time period is determined. If the degree of smoke change is greater than or equal to a preset degree threshold, smoke features are extracted from the multi-frame image sequence of the target inspection area within the second preset time period to obtain the multi-dimensional smoke features corresponding to the target inspection area; if the degree of smoke change is less than the preset degree threshold, smoke features are extracted from the multi-frame image sequence of the target inspection area within the first preset time period to obtain the multi-dimensional smoke features corresponding to the target inspection area. The multidimensional features of the smoke corresponding to the target inspection area include: smoke color features, smoke texture features, and smoke morphology features.

3. The method of claim 1, wherein, Extracting the multidimensional flame features corresponding to the target inspection area from the thermal imaging data sequence includes: The thermal imaging data sequence is subjected to image enhancement processing, and the background image and flame image in the thermal imaging data sequence are identified by a neural network segmentation model. The flame images in the thermal imaging data sequence are extracted using a neural network extraction model to obtain the multidimensional flame features corresponding to the target inspection area. The multidimensional flame features corresponding to the target inspection area include: flame color features, flame texture features, and flame shape features.

4. The method of claim 1, wherein, The process of fusing the fire source navigation and positioning information and the fire source dead reckoning information to obtain the fire source location coordinates includes: The initial heading deviation of the inertial navigation system is used to coarsely fuse the fire source navigation and positioning information and the fire source dead reckoning information to obtain a first fusion result; The fire source navigation and positioning information and the fire source dead reckoning information are accurately fused using the carrier phase measurement values ​​of the navigation and positioning sensor to obtain a second fusion result; The coordinates of the fire source location are determined based on the first fusion result and the second fusion result.

5. The method of claim 1, wherein, The method further includes: Obtain the vegetation humidity at the coordinates of the fire source location and the meteorological data at the current moment; Predict the fire spread information at the coordinates of the fire source location based on the vegetation humidity at the fire source location coordinates and the meteorological data at the current moment. Send information on the fire spread at the coordinates of the fire source to the ground control center.

6. The method of claim 1, wherein, The method further includes: The high-definition visible light camera collects visible light image data within the forest patrol area, and determines whether there are any preset sensitive behaviors within the forest patrol area based on the visible light image data within the forest patrol area. If a preset sensitive behavior is determined to exist within the forest patrol area based on visible light image data of the forest patrol area, a warning prompt voice is generated and played in the corresponding airspace of the preset sensitive behavior.

7. The method of claim 1, wherein, The method further includes: Plan flight routes within the forest inspection area based on inspection needs; and conduct drone inspections of the forest inspection area according to the flight routes within the forest inspection area. After the drone inspection of the forest inspection area is completed, a regional inspection report corresponding to the forest inspection area is generated. Send the area inspection report corresponding to the forest inspection area to the ground control center.

8. A forest fire prevention smoke intelligent early warning device for unmanned aerial vehicles, characterized in that, The application is applied to a UAV device, wherein the UAV device is provided with a smoke sensor, a high-definition visible light camera, a thermal imaging camera, a navigation positioning sensor and an inertial navigation system; the device comprises: a detection module, configured to detect a smoke concentration value in a target inspection area at a current time through the smoke sensor during UAV inspection of a forest inspection area, and acquire a plurality of image sequences of the target inspection area in a first preset time period collected by the high-definition visible light camera when the smoke concentration value is greater than a preset concentration value, and acquire a plurality of image sequences of the target inspection area in a second preset time period through the high-definition visible light camera, wherein the first preset time period is a time period before the current time, and the second preset time period is a time period after the current time; an analysis module, configured to perform smoke feature analysis on the plurality of image sequences of the target inspection area in the first preset time period and the plurality of image sequences of the target inspection area in the second preset time period, to obtain smoke multi-dimensional features corresponding to the target inspection area; a collection module, configured to collect a thermal imaging data sequence of the target inspection area in the second preset time period through the thermal imaging camera, and identify whether a fire point thermal signal appears in the target inspection area based on the thermal imaging data sequence; if the fire point thermal signal appears in the target inspection area, extract flame multi-dimensional features corresponding to the target inspection area from the thermal imaging data sequence; a judgment module, configured to judge a fire prevention warning level corresponding to the target inspection area based on the smoke multi-dimensional features and the flame multi-dimensional features corresponding to the target inspection area; if the fire prevention warning level is greater than a preset warning level, locate a fire source in the target inspection area through the navigation positioning sensor to obtain fire source navigation positioning information, and locate the fire source in the target inspection area through the inertial navigation system to obtain fire source dead reckoning information; a fusion module, configured to fuse the fire source navigation positioning information and the fire source dead reckoning information to obtain fire source position coordinates; a generation module, configured to generate a fire prevention warning prompt based on the fire source position coordinates, and send the fire prevention warning prompt to a ground control center, so that the ground control center performs fire prevention treatment at the fire source position coordinates through a fire prevention warning device.

9. A computer device, comprising: The application comprises a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the UAV forest fire smoke intelligent warning method in any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the UAV forest fire smoke intelligent warning method in any one of claims 1-7.

Images