A fire point monitoring method and system based on optical signals

Abstract

The present invention relates to the technical field of fire point detection, and specifically discloses a fire point monitoring method and system based on optical signals, wherein the method comprises real-time acquisition of the position of a drone, calculation of real acquisition parameters of the drone according to the position, comparison of the real acquisition parameters of each drone with corresponding standard acquisition parameters, and prediction of the fire point position; based on the fire point position, the acquisition data of all drones are read, and the data are uploaded to a detection platform as monitoring data; wherein, when the drone moves along the acquisition path, the movement speed is adjusted based on the optical signal; the present invention transfers a part of the recognition process to the drone, and during the working process, the drone simultaneously performs low-performance recognition, and adjusts the movement speed according to the recognition result. The main end only needs to analyze the position of the drone to preliminarily determine which position may have an abnormality, and the response speed is fast, and the subsequent image precision recognition has a clear target and is very efficient.

Description

Technical Field

[0001] The invention relates to the technical field of fire point detection, and in particular to a fire point monitoring method and system based on optical signals. Background Art

[0002] For some large areas prone to fire, staff are required to conduct fire inspections regularly. Due to the large area of ​​the area, staff can only conduct rough inspections during inspections. In addition, due to viewing angle problems, it is difficult for staff to cover everything. Therefore, drone-based fire inspection solutions came into being.

[0003] The existing fire inspection scheme of drones is a detection scheme based on optical signals, which acquires images and recognizes them. The process of image recognition is completed by the main end, which needs to recognize all images after one acquisition. The timeliness is not high and the image processing pressure is very high. Summary of the invention

[0004] The object of the present invention is to provide a fire point monitoring method and system based on optical signals to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A fire point monitoring method based on optical signals, the method comprising:

[0007] Acquire road information in the vegetation area based on a map service, and determine the probability of a fire point at each position in the vegetation area according to the road information;

[0008] Divide the vegetation area into blocks according to the probability of fire points, and calculate the comprehensive probability of each small area;

[0009] The standard acquisition parameters of the drone are set according to the comprehensive probability of each small area and its position relationship; the standard acquisition parameters include the acquisition path and the acquisition speed;

[0010] Collect the position of the drone in real time, calculate the real acquisition parameters of the drone based on the position, compare the real acquisition parameters of each drone with the corresponding standard acquisition parameters, and predict the location of the fire point;

[0011] Read all the collected data from drones based on the fire location, upload it to the detection platform as monitoring data;

[0012] Among them, when the drone moves along the collection path, it adjusts the movement speed based on the optical signal.

[0013] As a further solution of the present invention: the step of obtaining road information in the vegetation area based on the map service and determining the probability of fire points at various positions in the vegetation area according to the road information comprises:

[0014] Receive the regional coordinates input by the staff, and locate the vegetation area in the map service according to the regional coordinates;

[0015] Identifying the vegetation area and determining the road area in the vegetation area;

[0016] Display vegetation areas containing road areas, receive additional information input by staff, and update road areas;

[0017] Determine the probability of a fire point at each location based on the updated road area;

[0018] The determination process of the fire point probability is:

[0019] ; In the formula, Indicates location The probability of fire point, is the total number of roads, Indicates location With The distance of the road, is a preset parameter determined by the road width. is the preset adjustment parameter, Indicates Roads and Locations The weight of and is directly proportional to the value of .

[0020] As a further solution of the present invention: the step of dividing the vegetation area into blocks according to the probability of fire points and calculating the comprehensive probability of each small area comprises:

[0021] Traverse each position in the vegetation area and calculate the probability difference at each position;

[0022] When the probability difference at a certain position reaches a preset difference threshold, the position is marked as a contour position;

[0023] Divide the vegetation area into blocks according to the contour position to obtain each small area;

[0024] For each small area, the comprehensive probability of the small area is calculated; the comprehensive probability adopts the average value of the fire point probabilities at all locations.

[0025] As a further solution of the present invention: the step of setting the standard acquisition parameters of the drone according to the comprehensive probability of each small area and its position relationship includes:

[0026] Cluster each small area according to the comprehensive probability;

[0027] Determine the collection path of the drone based on similar small areas, and determine the collection speed based on the average comprehensive probability of similar small areas;

[0028] Taking each small area as a benchmark, obtain the number of corresponding acquisition paths and calculate the exposure;

[0029] For any collection path, obtain the exposure of the small area it passes through, and update the collection speed according to the exposure;

[0030] The acquisition speed is proportional to the average comprehensive probability, and the acquisition speed is updated according to the exposure as follows:

[0031] ; In the formula, is the preset correction factor, is the preset basic exposure. For exposure, is the acquisition speed before the update, is the updated acquisition speed.

[0032] As a further solution of the present invention, the step of collecting the position of the drone in real time, calculating the real collection parameters of the drone according to the position, comparing the real collection parameters of each drone with the corresponding standard collection parameters, and predicting the fire point position includes:

[0033] Establish a data transmission channel with the positioning module of each drone to obtain the location of the drone in real time;

[0034] Performing statistics on the positions to determine the true path and true speed;

[0035] Align the real path with the collected path, compare the collected speed and the real speed at each position on the path, and calculate the speed difference;

[0036] When the speed difference reaches a preset speed threshold, the position on the path is marked as a fire point position.

[0037] As a further solution of the present invention: when the drone moves along the acquisition path, the process of adjusting the movement speed based on the optical signal is as follows:

[0038] Obtaining collected images within a preset band range;

[0039] Output the collected image to the trained convolution recognition model to locate abnormal objects and abnormality levels in the collected image;

[0040] The movement speed is adjusted according to the abnormal objects and the abnormality level; the adjusted movement speed is inversely proportional to the number of abnormal objects, and the adjusted movement speed is inversely proportional to the abnormality level of the abnormal objects.

[0041] The technical solution of the present invention also provides a fire point monitoring system based on optical signals, the system comprising:

[0042] A fire point probability calculation module, used to obtain road information in the vegetation area based on a map service, and determine the fire point probability at each position in the vegetation area according to the road information;

[0043] The comprehensive probability calculation module is used to divide the vegetation area into blocks according to the probability of fire points and calculate the comprehensive probability of each small area;

[0044] A standard parameter setting module is used to set standard acquisition parameters of the drone according to the comprehensive probability of each small area and its position relationship; the standard acquisition parameters include acquisition path and acquisition speed;

[0045] The fire point location prediction module is used to collect the position of the drone in real time, calculate the real acquisition parameters of the drone according to the position, compare the real acquisition parameters of each drone with the corresponding standard acquisition parameters, and predict the fire point location;

[0046] The data upload module is used to read the collected data of all drones based on the fire point location and upload it to the detection platform as monitoring data;

[0047] Among them, when the drone moves along the collection path, it adjusts the movement speed based on the optical signal.

[0048] As a further solution of the present invention: the fire point probability calculation module includes:

[0049] A vegetation area positioning unit is used to receive the area coordinates input by the staff and locate the vegetation area in the map service according to the area coordinates;

[0050] A vegetation area identification unit, used to identify the vegetation area and determine the road area in the vegetation area;

[0051] A road area updating unit is used to display the vegetation area containing the road area, receive supplementary information input by the staff, and update the road area;

[0052] A probability determination unit, used to determine the probability of a fire point at each location according to the updated road area;

[0053] The determination process of the fire point probability is:

[0054] ; In the formula, Indicates location The probability of fire point, is the total number of roads, Indicates location With The distance of the road, is a preset parameter determined by the road width. is the preset adjustment parameter, Indicates Roads and Locations The weight of and is directly proportional to the value of .

[0055] As a further solution of the present invention: the comprehensive probability calculation module includes:

[0056] A difference calculation unit, used for traversing each position in the vegetation area and calculating the probability difference at each position;

[0057] A contour marking unit, used to mark a position as a contour position when the probability difference at a certain position reaches a preset difference threshold;

[0058] A block division unit, used for dividing the vegetation area into blocks according to the contour position to obtain each small area;

[0059] The calculation execution unit is used to calculate the comprehensive probability of each small area; the comprehensive probability adopts the average value of the fire point probabilities at all positions.

[0060] As a further solution of the present invention: the standard parameter setting module includes:

[0061] The clustering unit is used to cluster each small area according to the comprehensive probability;

[0062] A collection speed determination unit, used to determine the collection path of the drone according to the same type of small areas, and to determine the collection speed according to the average comprehensive probability of the same type of small areas;

[0063] An exposure calculation unit, used to sequentially use each small area as a reference, obtain the number of corresponding acquisition paths, and calculate the exposure;

[0064] An acquisition speed updating unit, configured to obtain, for any acquisition path, an exposure of a small area passed by the acquisition path, and update the acquisition speed according to the exposure;

[0065] The acquisition speed is proportional to the average comprehensive probability, and the acquisition speed is updated according to the exposure as follows:

[0066] ; In the formula, is the preset correction factor, is the preset basic exposure. For exposure, is the acquisition speed before the update, is the updated acquisition speed.

[0067] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention transfers part of the recognition process to the drone. During the operation, the drone simultaneously performs low-performance recognition and adjusts the movement speed according to the recognition result. The main end only needs to analyze the position of the drone to preliminarily determine which position may have an abnormality. The response speed is fast, and the subsequent image precision recognition has a clear goal and is very efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0069] Figure 1 The flowchart of the fire point monitoring method based on optical signals is shown in FIG.

[0070] Figure 2 This is a structural block diagram of the fire point monitoring system based on optical signals. DETAILED DESCRIPTION

[0071] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0072] Figure 1 The flowchart of the fire point monitoring method based on optical signals is shown in the figure. In an embodiment of the present invention, a fire point monitoring method based on optical signals includes:

[0073] Step S100: Obtaining road information in the vegetation area based on a map service, and determining the probability of a fire point at each position in the vegetation area according to the road information;

[0074] Map service refers to some existing map software. If the authority is sufficient, more detailed maps can be obtained. Roads in vegetation areas are obtained based on map services. The vegetation area is the area where fire points need to be monitored, and the road is the area where traffic can be passed. Pedestrians can walk on the road and enter the vegetation area. Generally, the vegetation area can be compared to a forest park. The probability of fire points at various locations in the vegetation area can be determined based on road information. Generally, the probability of fire points will be smaller if it is too close or too far from the road area. When it is too close, there are fewer combustible materials, and when it is too far, there are fewer ignition sources. Within a certain range, the probability of fire points gradually decreases with increasing distance.

[0075] Step S200: Divide the vegetation area into blocks according to the probability of fire points, and calculate the comprehensive probability of each small area;

[0076] After determining the probability of fire at each location, when the fire probabilities of two adjacent locations are similar, they are grouped into a small area to achieve block division. After the block division is completed, the comprehensive probability of each small area is calculated at the same time. Generally, the average vegetation probability of all locations in the small area is used.

[0077] It should be noted that the fire point probability at each location is independent, and the fire point probability represents the probability of a fire occurring, and the sum of the probability of a fire occurring at each location and the probability of no fire occurring is one.

[0078] Step S300: setting standard acquisition parameters of the drone according to the comprehensive probability of each small area and its position relationship; the standard acquisition parameters include acquisition path and acquisition speed;

[0079] The comprehensive probability of each small area represents the probability of a fire occurring in each small area. Small areas with similar comprehensive probabilities are classified into one category. For small areas of the same type, a collection path and the collection speed at each position in the motion path can be determined according to their positions. Since the main body of the fire point detection in this application is a drone, the collection path is the motion path of the drone, and the collection speed is the motion speed of the drone.

[0080] It should be noted that the process of determining a collection path according to its position can be done by using an existing path planning algorithm.

[0081] Step S400: collecting the position of the drone in real time, calculating the real acquisition parameters of the drone according to the position, comparing the real acquisition parameters of each drone with the corresponding standard acquisition parameters, and predicting the fire point position;

[0082] The core of the technical solution of the present invention is that the independence between the UAV and the data management platform is relatively high (not completely independent, there will be a small amount of data transmission). When the UAV is working, it autonomously acquires images and identifies fire points. When a fire point is identified, it will autonomously adjust its speed. For the data management platform, the position of the UAV is collected in real time (with the help of a positioning module), and the real collection parameters of the UAV can be calculated based on the position. The real collection parameters are the actual movement path of the UAV and the speed at each position on the path. The real collection parameters of each UAV are compared with the corresponding standard collection parameters. The slower the speed, the higher the possibility of the existence of a fire point. Therefore, when the UAV detects a suspected target, it will autonomously adjust its speed.

[0083] In short, the data management platform only needs to obtain the location of the drone to determine which areas may have abnormalities and then obtain images of the corresponding areas. This means that the drone does not need to provide real-time feedback of images during operation, greatly alleviating the pressure of data transmission. The link between the drone and the data management platform is location. The drone can autonomously identify and convert the identification results into position changes (speed). The data management platform can use positioning equipment to directly obtain position changes, thereby reversely judging the autonomous identification results, and then selecting the corresponding images for further identification. Other images can even be deleted directly, which is extremely efficient.

[0084] The premise of the above content is that when the drone moves along the acquisition path, the movement speed is adjusted based on the optical signal, and the optical signal is the acquired image.

[0085] Step S500: Read the collected data of all drones based on the fire point location, upload it to the detection platform as monitoring data;

[0086] The actual acquisition parameters of each drone are compared with the corresponding standard acquisition parameters, and the result is the location of the fire point. The fire point location is a suspicious location in the vegetation area. The drone's acquired image contains a time tag and a location tag (basic attribute). The acquisition data at the suspicious location is extracted based on the location tag and is sent to the detection platform as the final data that needs further analysis. The function of the detection platform is to perform more detailed detection of the image, which is not specifically limited. The detection platform can be a manual end, an intelligent recognition end with a built-in recognition algorithm, or even an AI image recognition device.

[0087] Regarding step S100, the step of obtaining road information in the vegetation area based on the map service and determining the probability of fire points at various positions in the vegetation area according to the road information includes:

[0088] Receive the regional coordinates input by the staff, and locate the vegetation area in the map service according to the regional coordinates;

[0089] Identifying the vegetation area and determining the road area in the vegetation area;

[0090] Display vegetation areas containing road areas, receive additional information input by staff, and update road areas;

[0091] The fire probability at each location is determined based on the updated road area.

[0092] In an example of the technical solution of the present invention, the process of determining the probability of a fire point is described in detail. The regional coordinates input by the staff are received, and the vegetation area is located in the map service according to the regional coordinates. Then, the vegetation area is identified, and the road area can be determined. Compared with the area covered with vegetation, the road area has a very obvious color value difference, and the identification and determination process is very simple.

[0093] Taking into account the problem of recognition errors, this application also introduces a correction process, which displays the vegetation area containing the road area, receives supplementary information input by the staff, and updates the road area. The supplementary information is a coordinate set, and the staff independently adds roads in the vegetation area based on actual conditions.

[0094] Finally, the probability of fire occurring at each location in the entire vegetation area is determined based on the road area, which is called the fire point probability. The determination process of the fire point probability is as follows:

[0095] ; In the formula, Indicates location The probability of fire point, is the total number of roads, Indicates location With The distance of the road, is a preset parameter determined by the road width. is the preset adjustment parameter, Indicates Roads and Locations The weight of and is directly proportional to the value of .

[0096] The above-mentioned determination process of the fire point probability is only one of the schemes. It is a composite function based on the Gaussian function. For a location and a road, the closer their locations are to the preset threshold, the higher the probability of the fire point is. The closer to the road or the farther away from the road, the lower the probability of the fire point is. The threshold is determined by the width of the road, which is generally several times the width of the road. In general, pedestrians throwing flaming objects are near the road. For a location, the final probability of the fire point can be determined by accumulating the analysis results of it and all roads. The term is used to ensure that the final fire point probability is in the range of zero to one.

[0097] Regarding step S200, the step of dividing the vegetation area into blocks according to the fire point probability and calculating the comprehensive probability of each small area includes:

[0098] Traverse each position in the vegetation area and calculate the probability difference at each position;

[0099] When the probability difference at a certain position reaches a preset difference threshold, the position is marked as a contour position;

[0100] Divide the vegetation area into blocks according to the contour position to obtain each small area;

[0101] For each small area, the comprehensive probability of the small area is calculated; the comprehensive probability adopts the average value of the fire point probabilities at all locations.

[0102] Each position in the vegetation area is traversed in a preset order, and the probability difference at each position is calculated, and the probability difference is used to characterize the difference between the fire point probability at this position and the fire point probability at the surrounding positions; when the probability difference at a certain position is large enough (reaching a preset difference threshold), it is used as the contour position, and multiple contour positions are connected to obtain small areas, and the vegetation area is divided into blocks; for each small area after blocking, it is necessary to calculate a comprehensive probability, and the simplest way to calculate the comprehensive probability is to calculate the mean of all fire point probabilities.

[0103] Regarding step S300, the step of setting the standard acquisition parameters of the drone according to the comprehensive probability of each small area and its position relationship includes:

[0104] Cluster each small area according to the comprehensive probability;

[0105] Determine the collection path of the drone based on similar small areas, and determine the collection speed based on the average comprehensive probability of similar small areas;

[0106] Taking each small area as a benchmark, obtain the number of corresponding acquisition paths and calculate the exposure;

[0107] For any collection path, the exposure of the small area it passes through is obtained, and the collection speed is updated according to the exposure.

[0108] In an example of the technical solution of the present invention, the process of limiting the standard acquisition parameters is described. There are mainly two standard acquisition parameters, one is the acquisition path, and the other is the acquisition speed at each position in the acquisition path. First, the small areas are clustered according to the comprehensive probability, and the small areas with similar (small probability difference) comprehensive probabilities are classified into one category. Then, for each type of small area, the acquisition path is determined separately and an independent drone is equipped. The average comprehensive probability of the same type of small areas is calculated. This value is actually close to the comprehensive probability. The acquisition speed is determined based on the average comprehensive probability.

[0109] After the above processing, the entire vegetation area is divided into multiple areas, and each area is collected and identified by a separate drone. The movement of the drone is continuous, and it will not suddenly appear from one small area to another. When it moves along the collection path, it will definitely "incidentally" collect data and identify other small areas. If we judge which small areas it has identified from the perspective of the drone, it will be a bit cumbersome. Therefore, this application starts from the perspective of small areas, obtains which drones will collect and identify each small area, and then calculates the exposure. The greater the exposure, the more drones will collect and identify the small area. Correspondingly, when the drone moves to the small area, the collection speed can be appropriately increased.

[0110] Specifically, the acquisition speed is proportional to the average comprehensive probability, and the acquisition speed is updated according to the exposure as follows:

[0111] ; In the formula, is the preset correction factor, is the preset basic exposure. For exposure, is the acquisition speed before the update, is the updated acquisition speed.

[0112] Regarding step S400, the real-time acquisition of the position of the drone, the calculation of the real acquisition parameters of the drone according to the position, and the comparison of the real acquisition parameters of each drone with the corresponding standard acquisition parameters, the step of predicting the fire point position includes:

[0113] Establish a data transmission channel with the positioning module of each drone to obtain the location of the drone in real time;

[0114] Performing statistics on the positions to determine the true path and true speed;

[0115] Align the real path with the collected path, compare the collected speed and the real speed at each position on the path, and calculate the speed difference;

[0116] When the speed difference reaches a preset speed threshold, the position on the path is marked as a fire point position.

[0117] The data management platform remotely obtains the location of the UAV and counts the location to obtain the actual movement path and true speed of the UAV. The actual movement path is point-to-point aligned with the pre-planned collection path, which is called registration. For the corresponding position, the true speed is compared with the pre-planned collection speed. If the true speed is less than the collection speed and the difference is large (reaching the preset speed threshold), it means that the UAV has recognized the abnormal phenomenon at the corresponding position. At this time, the corresponding position is marked as the fire point.

[0118] As a preferred embodiment of the technical solution of the present invention, when the drone moves along the acquisition path, the process of adjusting the movement speed based on the optical signal is as follows:

[0119] Obtaining collected images within a preset band range;

[0120] Output the collected image to the trained convolution recognition model to locate abnormal objects and abnormality levels in the collected image;

[0121] The movement speed is adjusted according to the abnormal objects and the abnormality level; the adjusted movement speed is inversely proportional to the number of abnormal objects, and the adjusted movement speed is inversely proportional to the abnormality level of the abnormal objects.

[0122] In an example of the technical solution of the present invention, the working process of the drone is specifically defined. The image acquisition function of the drone is a basic function, that is, the acquired image is acquired within a preset band range; then, a convolution recognition model is pre-built into the drone to identify abnormal objects, and then the corresponding preset abnormal level is queried; the number of abnormal objects and the abnormal level are adjusted by the active motion speed; the simplest way is to determine an abnormal score for different abnormal levels, and then accumulate the abnormal scores of all abnormal objects, and the total score obtained is used to evaluate the degree of abnormality; finally, a subtraction function in the range of zero to one is compounded on the total score, and multiplied by the original motion speed to achieve the adjustment function.

[0123] Figure 2 FIG. 1 is a structural block diagram of a fire point monitoring system based on optical signals. In an embodiment of the present invention, a fire point monitoring system based on optical signals, the system 10 includes:

[0124] A fire point probability calculation module 11 is used to obtain road information in the vegetation area based on a map service, and determine the fire point probability at each position in the vegetation area according to the road information;

[0125] The comprehensive probability calculation module 12 is used to divide the vegetation area into blocks according to the probability of the fire point, and calculate the comprehensive probability of each small area;

[0126] The standard parameter setting module 13 is used to set the standard acquisition parameters of the drone according to the comprehensive probability of each small area and its position relationship; the standard acquisition parameters include the acquisition path and the acquisition speed;

[0127] The fire point location prediction module 14 is used to collect the position of the drone in real time, calculate the real acquisition parameters of the drone according to the position, compare the real acquisition parameters of each drone with the corresponding standard acquisition parameters, and predict the fire point location;

[0128] The data uploading module 15 is used to read the collected data of all drones based on the fire point location and upload it to the detection platform as monitoring data;

[0129] Among them, when the drone moves along the collection path, it adjusts the movement speed based on the optical signal.

[0130] Furthermore, the fire point probability calculation module 11 includes:

[0131] A vegetation area positioning unit is used to receive the area coordinates input by the staff and locate the vegetation area in the map service according to the area coordinates;

[0132] A vegetation area identification unit, used to identify the vegetation area and determine the road area in the vegetation area;

[0133] A road area updating unit is used to display the vegetation area containing the road area, receive supplementary information input by the staff, and update the road area;

[0134] A probability determination unit, used to determine the probability of a fire point at each location according to the updated road area;

[0135] The determination process of the fire point probability is:

[0136] ; In the formula, Indicates location The probability of fire point, is the total number of roads, Indicates location With The distance of the road, is a preset parameter determined by the road width. is the preset adjustment parameter, Indicates Roads and Locations The weight of and is directly proportional to the value of .

[0137] Specifically, the comprehensive probability calculation module 12 includes:

[0138] A difference calculation unit, used for traversing each position in the vegetation area and calculating the probability difference at each position;

[0139] A contour marking unit, used to mark a position as a contour position when the probability difference at a certain position reaches a preset difference threshold;

[0140] A block division unit, used for dividing the vegetation area into blocks according to the contour position to obtain each small area;

[0141] The calculation execution unit is used to calculate the comprehensive probability of each small area; the comprehensive probability adopts the average value of the fire point probabilities at all positions.

[0142] In addition, the standard parameter setting module 13 includes:

[0143] The clustering unit is used to cluster each small area according to the comprehensive probability;

[0144] A collection speed determination unit, used to determine the collection path of the drone according to the same type of small areas, and to determine the collection speed according to the average comprehensive probability of the same type of small areas;

[0145] An exposure calculation unit, used to sequentially use each small area as a reference, obtain the number of corresponding acquisition paths, and calculate the exposure;

[0146] An acquisition speed updating unit, configured to obtain, for any acquisition path, an exposure of a small area passed by the acquisition path, and update the acquisition speed according to the exposure;

[0147] The acquisition speed is proportional to the average comprehensive probability, and the acquisition speed is updated according to the exposure as follows:

[0148] ; In the formula, is the preset correction factor, is the preset basic exposure. For exposure, is the acquisition speed before the update, is the updated acquisition speed.

[0149] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A fire point monitoring method based on optical signals, characterized in that: The method comprises: Acquire road information in the vegetation area based on a map service, and determine the probability of a fire point at each position in the vegetation area according to the road information; Divide the vegetation area into blocks according to the probability of fire points, and calculate the comprehensive probability of each small area; The standard acquisition parameters of the drone are set according to the comprehensive probability of each small area and its position relationship; the standard acquisition parameters include the acquisition path and the acquisition speed; Collect the position of the drone in real time, calculate the real acquisition parameters of the drone based on the position, compare the real acquisition parameters of each drone with the corresponding standard acquisition parameters, and predict the location of the fire point; Read all the collected data from drones based on the fire location, upload it to the detection platform as monitoring data; Among them, the UAV adjusts its movement speed based on the optical signal when moving along the collection path; The step of acquiring road information in the vegetation area based on the map service and determining the probability of fire points at various positions in the vegetation area according to the road information comprises: Receive the regional coordinates input by the staff, and locate the vegetation area in the map service according to the regional coordinates; Identifying the vegetation area and determining the road area in the vegetation area; Display vegetation areas containing road areas, receive additional information input by staff, and update road areas; Determine the probability of a fire point at each location based on the updated road area; The determination process of the fire point probability is: ; In the formula, Indicates location The probability of fire point, is the total number of roads, Indicates location The distance to the first road, is a preset parameter determined by the road width. is the preset adjustment parameter, Indicates Roads and Locations The weight of and The value of is directly proportional to; The step of dividing the vegetation area into blocks according to the probability of the fire point and calculating the comprehensive probability of each small area comprises: Traverse each position in the vegetation area and calculate the probability difference at each position; When the probability difference at a certain position reaches a preset difference threshold, the position is marked as a contour position; Divide the vegetation area into blocks according to the contour position to obtain each small area; For each small area, the comprehensive probability of the small area is calculated; the comprehensive probability adopts the average value of the fire point probabilities at all locations.

Images