Debris flow channel data acquisition method and system based on unmanned aerial vehicle

Abstract

According to the debris flow channel data acquisition method and system based on the unmanned aerial vehicle provided by the invention, the item dimension analysis and the specific local dimension analysis are performed on the target item in the target debris flow channel acquisition data, so that the item debris flow hidden danger point data for the target item can be obtained; according to the item debris flow hidden danger point data of the first item local analysis result of the target item and the local debris flow hidden danger point data of the first item local analysis result of the target item, dimension analysis can be performed on the target item in the target debris flow channel collection data based on the item debris flow hidden danger point data, the local debris flow hidden danger point data and the sample dimension; according to the invention, local hidden danger important feature description contents lacked in the target matter in the target debris flow channel acquisition data can be repaired, the integrity and rationality of the debris flow channel data acquisition result of the finally obtained target matter can be ensured, and the accuracy and reliability of the debris flow channel data acquisition result can be improved.

Description

Technical Field

[0001] This application relates to the technical field of data collection, and more specifically, to a method and system for collecting debris flow channel data based on an unmanned aerial vehicle (UAV). Background Art

[0002] Debris flows often occur in areas with complex geological structures, developed fault folds, intense neotectonic activities, and high seismic intensities, where the surface rocks are broken and there are collapses, landslides, and other bad geological phenomena. The development of these phenomena provides a rich source of solid materials for the formation of debris flows. In addition, areas with loose, soft, easily weathered rock formations, developed joints, or alternating hard and soft layers are also prone to being damaged and can provide a rich source of debris for debris flows. Human engineering activities such as deforestation causing soil erosion, mountain quarrying, and waste slag from stone quarrying often also provide a large amount of material sources for debris flows. Water is not only an important component of debris flows but also the triggering condition and transportation medium (power source). The water sources of debris flows include forms such as heavy rain, snowmelt, and the bursting of reservoirs (ponds). In China, the water sources of most debris flows are heavy rain, continuous rainfall for a long time, etc.

[0003] Currently, the situation of debris flow channels is very complex. Conducting surveys manually requires a large amount of manpower and time, and it is also impossible to accurately survey every location. With the rapid development of UAVs, when UAV technology is specifically applied to the survey of debris flow channels, it can survey locations that cannot be surveyed manually. However, how to accurately collect data through UAVs is a technical problem that is difficult to solve currently. Summary of the Invention

[0004] To improve the technical problems existing in the related art, this application provides a method and system for collecting debris flow channel data based on an unmanned aerial vehicle (UAV).

[0005] In a first aspect, a method for collecting debris flow channel data based on an unmanned aerial vehicle (UAV) is provided, including: Obtaining the data of major debris flow hazard points and local debris flow hazard points corresponding to the target debris flow channel collection data; the data of major debris flow hazard points matches the target matter in the target debris flow channel collection data, and the data of local debris flow hazard points matches the first local analysis result of the target matter; Obtaining a sample dimension that matches the target matter, counting the number of the first important hazard feature description contents corresponding to the sample dimension, and the number of the second important hazard feature description contents corresponding to the data of major debris flow hazard points; When the number of important feature description contents of the first hidden danger exceeds the number of important feature description contents of the second hidden danger, analyze and process the debris flow hidden danger point data of the matter according to the sample dimension to obtain a first pending matter dimension analysis result; Analyze and process the first pending matter dimension analysis result according to the local debris flow hidden danger point data to obtain the debris flow channel data acquisition result corresponding to the target matter.

[0006] In this application, obtaining the debris flow hidden danger point data of the matter corresponding to the target debris flow channel acquisition data and the local debris flow hidden danger point data includes: Obtain the target debris flow channel acquisition data from the channel information dataset, load the target debris flow channel acquisition data into the matter intelligent analysis thread, and output the debris flow hidden danger point data of the matter corresponding to the target debris flow channel acquisition data through the matter intelligent analysis thread; Load the target debris flow channel acquisition data into the local intelligent analysis thread, and output the local debris flow hidden danger point data of the matter corresponding to the target debris flow channel acquisition data through the local intelligent analysis thread.

[0007] In this application, loading the target debris flow channel acquisition data into the matter intelligent analysis thread and outputting the debris flow hidden danger point data of the matter corresponding to the target debris flow channel acquisition data through the matter intelligent analysis thread includes: Load the target debris flow channel acquisition data into the matter intelligent analysis thread, and obtain the matter dimension features corresponding to the target matter in the target debris flow channel acquisition data through the matter intelligent analysis thread, and read the first classification result corresponding to the matter dimension features; the first classification result is used to represent the type of matter local analysis result corresponding to the important feature description content of the hidden danger of the target matter; Generate a first data intersection description set according to the first classification result and the matter key features of the target debris flow channel acquisition data; Obtain the mean value of the description attributes corresponding to the first data intersection description set, and determine the position description content of the important feature description content of the hidden danger in the target matter in the target debris flow channel acquisition data according to the mean value of the description attributes; Determine the debris flow hidden danger point data of the matter corresponding to the target debris flow channel acquisition data according to the type of matter local analysis result and the position description content.

[0008] In this application, obtaining the matter dimension features corresponding to the target matter in the target debris flow channel acquisition data through the matter intelligent analysis thread includes: Obtain the matter description features corresponding to the target matter in the collected data of the target debris flow gully in the intelligent analysis thread of the matter, and output the second classification result corresponding to the matter description features according to the classification unit in the intelligent analysis thread of the matter; Obtain the key matter features of the collected data of the target debris flow gully output by the target feature extraction unit in the intelligent analysis thread of the matter, and perform a function process on the second classification result and the key matter features to obtain the second data intersection description set corresponding to the collected data of the target debris flow gully; Perform bucketing processing on the collected data of the target debris flow gully according to the second data intersection description set to obtain Z geological state description contents of the matter local analysis result range, and obtain the local description features corresponding to the Z geological state description contents of the matter local analysis result range according to the intelligent analysis thread of the matter; Z is a positive integer; Integrate the matter description features and the local description features corresponding to the Z geological state description contents of the matter local analysis result range into the matter dimension features.

[0009] In this application, the loading of the collected data of the target debris flow gully into the local intelligent analysis thread and outputting the local debris flow hidden danger point data corresponding to the collected data of the target debris flow gully by the local intelligent analysis thread includes: Load the collected data of the target debris flow gully into the local intelligent analysis thread, and analyze the first matter local analysis result of the target matter in the collected data of the target debris flow gully in the local intelligent analysis thread; If the first matter local analysis result is analyzed in the collected data of the target debris flow gully, obtain the range geological state description content covering the first matter local analysis result from the collected data of the target debris flow gully, obtain the position of the local hidden danger important feature description content corresponding to the first matter local analysis result according to the range geological state description content, and determine the local debris flow hidden danger point data corresponding to the collected data of the target debris flow gully based on the position of the local hidden danger important feature description content; If the first matter local analysis result is not analyzed in the collected data of the target debris flow gully, determine that the local debris flow hidden danger point data corresponding to the collected data of the target debris flow gully is a null value.

[0010] In this application, if the local analysis result of the first matter is analyzed from the data collected from the target debris flow channel, the description content of the range geological state covering the local analysis result of the first matter is obtained from the data collected from the target debris flow channel. Based on the description content of the range geological state, the position of the description content of the important features of the local hidden danger corresponding to the local analysis result of the first matter is obtained. Based on the position of the description content of the important features of the local hidden danger, the local debris flow hidden danger point data corresponding to the data collected from the target debris flow channel is determined, including: If the local analysis result of the first matter is analyzed from the data collected from the target debris flow channel, the data collected from the target debris flow channel is filtered to obtain the description content of the range geological state covering the local analysis result of the first matter; The local edge features corresponding to the description content of the range geological state are obtained, and the position of the description content of the important features of the local hidden danger corresponding to the local analysis result of the first matter is regression-analyzed based on the local edge features; Based on the position of the description content of the important features of the local hidden danger, the description content of the important features of the hidden danger of the local analysis result of the first matter is associated to obtain the local debris flow hidden danger point data corresponding to the data collected from the target debris flow channel.

[0011] In this application, the analysis and processing of the first undetermined matter dimension analysis result according to the local debris flow hidden danger point data to obtain the debris flow channel data collection result corresponding to the target matter includes: If the matter debris flow hidden danger point data covers the dimension of the local analysis result of the second matter and does not cover the dimension of the local analysis result of the third matter, the first local dimension corresponding to the local analysis result of the third matter is determined according to the position of the description content of the important features of the hidden danger of the local analysis result of the first matter covered in the local debris flow hidden danger point data; the local analysis result of the second matter and the local analysis result of the third matter are in a mirror image relationship; The first local data volume of the local analysis result of the second matter in the first undetermined matter dimension analysis result is obtained, and the position of the description content of the important features of the hidden danger of the local analysis result of the third matter is determined according to the first local data volume and the first local dimension; The position of the description content of the important features of the hidden danger of the local analysis result of the third matter is loaded into the first undetermined matter dimension analysis result to obtain the debris flow channel data collection result corresponding to the target matter in the data collected from the target debris flow channel.

[0012] In this application, loading the position of the hidden danger important feature description content of the third matter partial analysis result into the first undetermined matter dimension analysis result to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel collection data includes: Loading the position of the hidden danger important feature description content of the third matter partial analysis result into the first undetermined matter dimension analysis result to obtain the second undetermined matter dimension analysis result corresponding to the target matter in the target debris flow channel collection data, and obtaining the dimension difference value between the sample dimension and the second undetermined matter dimension analysis result; If the dimension difference value exceeds the target difference value, then perform debugging on the hidden danger important feature description content of the second undetermined matter dimension analysis result based on the sample dimension to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel collection data.

[0013] In this application, the target debris flow channel collection data is the a-th debris flow channel collection data in the channel information dataset, where a is a positive integer; analyzing and processing the first undetermined matter dimension analysis result according to the local debris flow hidden danger point data to obtain the debris flow channel data collection result corresponding to the target matter includes: If the matter debris flow hidden danger point data does not include the dimensions of the second matter partial analysis result and the third matter partial analysis result, then determine the second local dimension corresponding to the second matter partial analysis result and the third local dimension corresponding to the third matter partial analysis result according to the position of the hidden danger important feature description content of the first matter partial analysis result covered in the local debris flow hidden danger point data; the second matter partial analysis result and the third matter partial analysis result are in a mirror image relationship; In the (a - 1)-th debris flow channel collection data, obtain the second local data volume corresponding to the second matter partial analysis result and the third local data volume corresponding to the third matter partial analysis result, and determine the position of the hidden danger important feature description content of the second matter partial analysis result according to the second local data volume and the second local dimension; Determine the position of the hidden danger important feature description content of the third matter partial analysis result according to the third local data volume and the third local dimension, and load the position of the hidden danger important feature description content of the second matter partial analysis result and the position of the hidden danger important feature description content of the third matter partial analysis result into the first undetermined matter dimension analysis result to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel collection data.

[0014] In a second aspect, a debris flow channel data acquisition system based on an unmanned aerial vehicle is provided, including a processor and a memory that communicate with each other. The processor is configured to read and execute a computer program from the memory to implement the above method.

[0015] For a debris flow channel data acquisition method and system provided in an embodiment of the present application, by respectively performing item dimension analysis and specific local dimension analysis on target items in the data collected from a target debris flow channel, debris flow hidden danger point data for the target item and local debris flow hidden danger point data for the first item local analysis result of the target item can be obtained. Furthermore, based on the debris flow hidden danger point data for the item, the local debris flow hidden danger point data, and the sample dimension, dimension analysis can be performed on the target item in the data collected from the target debris flow channel, the content of the missing local hidden danger important feature description for the target item in the data collected from the target debris flow channel can be repaired, and the integrity and rationality of the debris flow channel data acquisition result of the target item finally obtained can be ensured. Furthermore, the accuracy and reliability of the debris flow channel data acquisition result can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present application, the accompanying drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application and should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 It is a flowchart of a debris flow channel data acquisition method based on an unmanned aerial vehicle provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To better understand the above technical solutions, the technical solutions of the present application will be described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solutions of the present application, rather than limitations on the technical solutions of the present application. Without conflict, the technical features in the embodiments of the present application and the embodiments can be combined with each other.

[0019] Please refer to Figure 1 , which shows a debris flow channel data acquisition method based on an unmanned aerial vehicle. The method may include the technical solutions described in the following steps S101 - step S104.

[0020] Step S101: Obtain the data of the potential debris flow hazard points corresponding to the collected data of the target debris flow gully and the data of the local potential debris flow hazard points; the data of the potential debris flow hazard points corresponding to the events match the target events in the collected data of the target debris flow gully, and the data of the local potential debris flow hazard points match the first local analysis results of the target events.

[0021] For debris flow to form, the following three conditions must be met simultaneously, namely, steep terrain and landforms that are convenient for water and debris collection; abundant loose materials; and a large amount of water source in a short period of time. Debris flow landforms can generally be divided into three parts: the formation area, the flow area, and the deposition area. The terrain in the upstream formation area is mostly in the shape of a ladle or funnel with mountains on three sides and an outlet on one side. The terrain is relatively open, with high mountains and steep slopes, broken mountain bodies, and poor vegetation growth around, which is conducive to the concentration of water and debris. The terrain in the middle flow area is mostly narrow, steep, and deep canyons, with a large longitudinal slope of the valley bed, enabling the debris flow to rush straight down rapidly. The terrain in the downstream deposition area is usually an open and flat piedmont plain or river valley terrace, providing a place for the accumulation of deposits.

[0022] Exemplarily, the data of potential debris flow hazard points are such as: source points, soil softness, and water sources, etc. The data of the potential debris flow hazard points corresponding to the events and the data of the local potential debris flow hazard points are collected by drones. The drones are not only equipped with devices with camera functions but also have geophysical exploration-related devices, capable of collecting ground and underground data.

[0023] For example, it is possible to obtain a dataset of gully information of the target event or data of the description of the geological state captured in the mobile scenario; when performing dimensional analysis on the dataset of gully information or the data of the description of the geological state, the collected data of the debris flow gully in the data of the description of the geological state or the dataset of gully information can be determined as the target collected data of the debris flow gully. By performing event analysis on the target collected data of the debris flow gully, the data of the potential debris flow hazard points corresponding to the target event can be obtained; meanwhile, local analysis can also be performed on the target collected data of the debris flow gully to obtain the data of the local potential debris flow hazard points of the first local analysis result of the target event.

[0024] Among them, it is possible to obtain an event intelligent analysis thread and a local intelligent analysis thread, load the target collected data of the debris flow gully into the event intelligent analysis thread, and through this event intelligent analysis thread, the data of the potential debris flow hazard points corresponding to the target collected data of the debris flow gully can be output; meanwhile, the target collected data of the debris flow gully can also be loaded into the local intelligent analysis thread, and through this local intelligent analysis thread, the data of the local potential debris flow hazard points corresponding to the target collected data of the debris flow gully can be output.

[0025] Optionally, the matter intelligent analysis thread used in the embodiments of the present application may be a decision-making thread with a credibility factor. For example, through the matter intelligent analysis thread, the important feature description content of the matter hidden danger of the target matter in the collected data of the target debris flow channel can be regressed and analyzed. Each important feature description content of the matter hidden danger obtained by regression analysis can correspond to a first credibility factor. The first credibility factor can be used to represent the regression analysis accuracy of each important feature description content of the matter hidden danger obtained by regression analysis. The important feature description content of the matter hidden danger obtained by regression analysis and the corresponding first credibility factor can be called the matter debris flow hidden danger point data corresponding to the collected data of the target debris flow channel.

[0026] Step S102: Obtain a sample dimension that matches the target matter, and count the number of first important feature description contents corresponding to the sample dimension, and the number of second important feature description contents corresponding to the matter debris flow hidden danger point data.

[0027] Exemplarily, the direction of the historical data saved in the database of the sample dimension.

[0028] Among them, the dimension parsing is performed through different directions, which can analyze the data more comprehensively.

[0029] The first important feature description content can be understood as the current important feature description content of the hidden danger, and the second important feature description content can be understood as the historical important feature description content of the hidden danger.

[0030] Step S103: When the number of the first important feature description contents exceeds the number of the second important feature description contents, analyze and process the matter debris flow hidden danger point data according to the sample dimension to obtain a first pending matter dimension analysis result.

[0031] Step S104: Analyze and process the first pending matter dimension analysis result according to the local debris flow hidden danger point data to obtain the debris flow channel data collection result corresponding to the target matter.

[0032] The data collected from the target debris flow gully may not cover all the target matters completely. For example, if the local hidden danger range of the target matter is not included in the data collected from the target debris flow gully, then some important feature descriptions of the matter hidden danger are missing in the data of the matter debris flow hidden danger point corresponding to the data collected from the target debris flow gully. The important feature descriptions of the hidden danger corresponding to the target matter can be patched through the sample dimension to complete the missing important feature descriptions of the matter hidden danger, so as to obtain the first undetermined matter dimension analysis result corresponding to the target matter. When the local debris flow hidden danger point data includes the important feature descriptions of the local hidden danger in the first matter local analysis result, the important feature descriptions of the local hidden danger in the local debris flow hidden danger point data and the important feature descriptions of the matter hidden danger in the matter debris flow hidden danger point data can be combined to adjust the first undetermined matter dimension analysis result, so as to obtain the debris flow gully data collection result of the target matter in the data collected from the target debris flow gully. After obtaining the debris flow gully data collection result corresponding to the current data collected from the target debris flow gully, the matter dimension analysis can be continued for the next data collected from the debris flow gully in the gully information data set, so as to obtain the debris flow gully data collection results of the target matter in the data collected from each debris flow gully in the gully information data set.

[0033] In the embodiment of the present application, by respectively performing matter dimension analysis and specific local dimension analysis on the target matter in the data collected from the target debris flow gully, the data of the matter debris flow hidden danger point corresponding to the target matter and the local debris flow hidden danger point data of the first matter local analysis result corresponding to the target matter can be obtained. Furthermore, based on the data of the matter debris flow hidden danger point, the local debris flow hidden danger point data and the sample dimension, the dimension analysis of the target matter in the data collected from the target debris flow gully can be carried out, and the missing important feature descriptions of the local hidden danger of the target matter in the data collected from the target debris flow gully can be patched, which can ensure the integrity and rationality of the finally obtained debris flow gully data collection result of the target matter, and further improve the accuracy and reliability of the debris flow gully data collection result.

[0034] Another method for collecting debris flow gully data based on an unmanned aerial vehicle provided by the embodiment of the present application may include the following steps S201-S208: Step S201, load the data collected from the target debris flow gully into the matter intelligent analysis thread, obtain the matter dimension features corresponding to the target matter in the data collected from the target debris flow gully through the matter intelligent analysis thread, and read the first classification result corresponding to the matter dimension features; the first classification result is used to represent the type of the matter local analysis result corresponding to the important feature description content of the hidden danger of the target matter.

[0035] Exemplarily, the matter intelligent analysis thread can be understood as an artificial intelligence analysis thread, specifically including CNN, etc.

[0036] For example, after obtaining the channel information dataset captured in the mobile scenario, one can select a debris flow channel collection data from the channel information dataset as the target debris flow channel collection data, load the target debris flow channel collection data into the trained matter intelligent analysis thread. Through the matter intelligent analysis thread, the matter dimension features corresponding to the target matters in the target debris flow channel collection data can be obtained. Through the classification unit of the matter intelligent analysis thread, the first classification result corresponding to the matter dimension features can be output. This first classification result can be used to represent the type of the matter local analysis result corresponding to the description content of the hidden danger important features of the target matter.

[0037] Optionally, during the process of using the matter intelligent analysis thread to extract features from the target debris flow channel collection data, the target debris flow channel collection data can be loaded into the matter intelligent analysis thread. In the matter intelligent analysis thread, the matter description features corresponding to the target matters in the target debris flow channel collection data can be obtained. According to the classification unit in the matter intelligent analysis thread, the second classification result corresponding to the matter description features can be output; the matter key features for the target debris flow channel collection data output by the target feature extraction unit in the matter intelligent analysis thread are obtained, and the second classification result and the matter key features are processed by a function to obtain the second data intersection description set corresponding to the target debris flow channel collection data; the target debris flow channel collection data is bucketed according to the second data intersection description set to obtain Z matter local analysis result range geological state description contents. According to the matter intelligent analysis thread, the local description features corresponding to the Z matter local analysis result range geological state description contents are obtained, where Z is a positive integer; the matter description features and the local description features corresponding to the Z matter local analysis result range geological state description contents are integrated into matter dimension features.

[0038] Among them, the matter description feature can be considered as the feature representation extracted from the data collected from the target debris flow channel for representing the target matter; the second classification result can also be used to represent the type of local analysis result of the matter corresponding to the important feature description content of the matter hidden danger covered in the data collected from the target debris flow channel; the target feature extraction unit can refer to the last feature extraction unit in the matter intelligent analysis thread, and the matter key feature can represent the key feature of the data collected from the target debris flow channel output by the last feature extraction unit in the matter intelligent analysis thread; the second data intersection description set can be the class data intersection description set corresponding to the data collected from the target debris flow channel. By weighting the matter key feature output by the last feature extraction unit in the matter intelligent analysis thread and the second classification result (the second classification result can be considered as the weight corresponding to the matter key feature), the second data intersection description set can be obtained. This second data intersection description set can be considered as the result of visualizing the matter key feature output by the target feature extraction unit, and can be used to represent the range of the description attribute points of the geological state description content concerned by the matter intelligent analysis thread.

[0039] The calculation can use the class data intersection description set (the second data intersection description set) of the important feature description content of each matter hidden danger in the data collected from the target debris flow channel as the prior information of the range position, and perform bucketing processing on the data collected from the target debris flow channel, that is, filter the data collected from the target debris flow channel according to the second data intersection description set to obtain the geological state description content of the range of the local analysis result of the matter covering a single part; furthermore, the matter intelligent analysis thread can perform feature extraction on the geological state description content of the range of each local analysis result of the matter to obtain the local description features corresponding to the geological state description content of the range of each local analysis result of the matter respectively. The foregoing matter description feature and the local description features corresponding to the geological state description content of the range of each local analysis result of the matter can be integrated into the matter dimension feature for the target matter; the local description feature can be considered as the feature representation extracted from the geological state description content of the range of the local analysis result of the matter for representing the local analysis result of the matter.

[0040] Step S202, generate a first data intersection description set according to the first classification result and the matter key feature of the data collected from the target debris flow channel.

[0041] Exemplarily, the matter key feature can be understood as the soil dryness degree, the vibration situation, etc.

[0042] For example, after obtaining the first classification result, the first classification result can be multiplied by the key feature of the matter for collecting data of the target debris flow channel to generate the first data intersection description set. Among them, both the first data intersection description set and the second data intersection description set are class data intersection description sets for collecting data of the target debris flow channel. Only the first data intersection description set uses the first classification result as the weight of the key feature of the matter output by the target feature extraction unit (here it is defaulted that the first classification result combines the matter description feature and the local description feature), while the second data intersection description set uses the second classification result as the weight of the key feature of the matter output by the target feature extraction unit, and the second classification result is only related to the matter description feature.

[0043] Step S203: Obtain the mean value of the description attributes corresponding to the first data intersection description set, determine the position description content of the important hidden danger feature description in the target matter in the data collected from the target debris flow channel according to the mean value of the description attributes, and determine the matter debris flow hidden danger point data corresponding to the data collected from the target debris flow channel according to the type of the matter local analysis result and the position description content.

[0044] Exemplarily, the mean value of the description attribute is understood as the value of the factor affecting the occurrence of debris flow. The larger the value, the greater the possibility of debris flow occurrence.

[0045] For example, the mean value of the description attribute can be obtained from the first data intersection description set, and the mean value of the description attribute can be determined as the position description content of the important hidden danger feature description in the target matter in the data collected from the target debris flow channel. According to the type of the matter local analysis result and the position description content, the matter structure of the target matter in the data collected from the target debris flow channel can be determined, and this matter structure can be used as the matter debris flow hidden danger point data corresponding to the target matter in the data collected from the target debris flow channel.

[0046] Step S204: Load the data collected from the target debris flow channel into the local intelligent analysis thread, and analyze the first matter local analysis result of the target matter in the data collected from the target debris flow channel in the local intelligent analysis thread.

[0047] For example, the meter can also load the collected data of the target debris flow channel into a local intelligent analysis thread, and first analyze whether the first matter partial analysis result of the target matter is included in the collected data of the target debris flow channel in the local intelligent analysis thread. Among them, the local intelligent analysis thread can be used to analyze the hidden danger important feature description content of the first matter partial analysis result. Therefore, it is necessary to analyze the first matter partial analysis result in the collected data of the target debris flow channel. If the first matter partial analysis result of the target matter is not analyzed in the collected data of the target debris flow channel, it can be directly determined that the local debris flow hidden danger point data corresponding to the collected data of the target debris flow channel is a null value, and there is no need to execute the subsequent step of analyzing the hidden danger important feature description content of the first matter partial analysis result.

[0048] Step S205, if the first matter partial analysis result is analyzed in the collected data of the target debris flow channel, obtain the range geological state description content covering the first matter partial analysis result from the collected data of the target debris flow channel, obtain the position of the local hidden danger important feature description content corresponding to the first matter partial analysis result according to the range geological state description content, and determine the local debris flow hidden danger point data corresponding to the collected data of the target debris flow channel based on the position of the local hidden danger important feature description content.

[0049] For example, if the first matter partial analysis result is analyzed in the collected data of the target debris flow channel, the position range of the first matter partial analysis result in the collected data of the target debris flow channel can be determined. Based on the position range of the first matter partial analysis result in the collected data of the target debris flow channel, the collected data of the target debris flow channel is filtered to obtain the range geological state description content covering the first matter partial analysis result. In the local intelligent analysis thread, feature extraction can be performed on the range geological state description content to obtain the local edge features corresponding to the first matter partial analysis result in the range geological state description content. According to the local edge features, the position of the local hidden danger important feature description content corresponding to the first matter partial analysis result can be obtained through regression analysis; based on the position of the local hidden danger important feature description content, the hidden danger important feature description content of the first matter partial analysis result can be associated to obtain the local debris flow hidden danger point data corresponding to the collected data of the target debris flow channel.

[0050] Step S206, obtain the sample dimension matching the target matter, and count the number of the first hidden danger important feature description contents corresponding to the sample dimension, and the number of the second hidden danger important feature description contents corresponding to the matter debris flow hidden danger point data.

[0051] For example, the meter can obtain the sample dimension corresponding to the target matter, and count the number of the first important feature description contents of the matter hidden danger important feature description contents covered in the sample dimension, and the number of the second important feature description contents of the matter hidden danger important feature description contents covered in the matter debris flow hidden danger point data. Among them, the number of the first important feature description contents is known when building the sample dimension, and the number of the second important feature description contents is the number of the matter hidden danger important feature description contents obtained by the regression analysis of the matter intelligent analysis thread.

[0052] Step S207, when the number of the first important feature description contents exceeds the number of the second important feature description contents, analyze and process the matter debris flow hidden danger point data according to the sample dimension to obtain the first pending matter dimension analysis result.

[0053] Exemplarily, the first pending matter dimension analysis result For example, when the number of the first important feature description contents exceeds the number of the second important feature description contents, it means that there are missing matter hidden danger important feature description contents in the matter debris flow hidden danger point data. The matter debris flow hidden danger point data can be repaired for the important feature description contents of the hidden danger (repairing means completing the defective data) through the sample dimension to improve the missing matter hidden danger important feature description contents, so as to obtain the first pending matter dimension analysis result corresponding to the target matter.

[0054] Step S208, analyze and process the first pending matter dimension analysis result according to the local debris flow hidden danger point data to obtain the debris flow channel data acquisition result corresponding to the target matter.

[0055] For example, based on the local debris flow hidden danger point data, the matter local analysis result matching the first matter local analysis result in the first pending matter dimension analysis result can be analyzed and processed to obtain the debris flow channel data acquisition result corresponding to the target matter. Optionally, if the local debris flow hidden danger point data is a null value (that is, the first matter local analysis result is not covered in the target debris flow channel acquisition data), the first pending matter dimension analysis result can be directly determined as the debris flow channel data acquisition result corresponding to the target matter.

[0056] If the debris flow hidden danger point data of the matter does not include the dimensions of the local analysis results of the second matter and the third matter, that is, the data collected from the target debris flow channel does not cover the local analysis results of the second matter nor the local analysis results of the third matter, then the second local dimension corresponding to the local analysis results of the second matter and the third local dimension corresponding to the local analysis results of the third matter can be determined according to the position of the hidden danger important feature description content of the local analysis results of the first matter covered in the local debris flow hidden danger point data; furthermore, in the data collected from the (a - 1)th debris flow channel, the second local data volume corresponding to the local analysis results of the second matter and the third local data volume corresponding to the local analysis results of the third matter can be obtained; in other words, the data volume of the local analysis results of the second matter in the previous debris flow channel data collection can be used as the data volume of the local analysis results of the second matter in the target debris flow channel data collection, and the geological state description content of the local analysis results of the third matter in the previous debris flow channel data collection can be used as the data volume of the local analysis results of the third matter in the target debris flow channel data collection. Then, according to the second local data volume and the second local dimension, the position of the hidden danger important feature description content of the local analysis results of the second matter can be determined; according to the third local data volume and the third local dimension, the position of the hidden danger important feature description content of the local analysis results of the third matter can be determined, and the position of the hidden danger important feature description content of the local analysis results of the second matter and the position of the hidden danger important feature description content of the local analysis results of the third matter are loaded into the analysis results of the first undetermined matter dimension to obtain the debris flow channel data collection results corresponding to the target matter in the target debris flow channel data collection. If the data collected from the (a - 1)th debris flow channel also covers the local analysis results of the second matter and the third matter, then it is possible to continue to trace back to obtain the data volumes of the local analysis results of the second matter and the third matter in the data collected from the (a - 2)th debris flow channel respectively, so as to determine the positions of the hidden danger important feature description contents of the local analysis results of the second matter and the third matter in the target debris flow channel data collection respectively. If the local analysis results of the second matter and the third matter are not analyzed in the debris flow channel data collection before the target debris flow channel data collection, then an approximate data volume can be set for the local analysis results of the second matter and the third matter respectively according to the analysis results of the first undetermined matter dimension, so as to determine the positions of the hidden danger important feature description contents of the local analysis results of the second matter and the third matter in the target debris flow channel data collection respectively.

[0057] Optionally, there may be some unreasonable important feature descriptions of matter hazards in the dimension obtained after analyzing and processing the first undetermined matter dimension analysis result based on the local debris flow hazard point data. Therefore, the unreasonable important feature descriptions of matter hazards can be adjusted in combination with the sample dimension to obtain the final debris flow channel data collection result of the target matter. For example, assuming that the local analysis result of the third matter is not analyzed in the data collection of the target debris flow channel, the first undetermined matter dimension analysis result at the position of the important feature description of the hazard with the local analysis result of the third matter added can be determined as the second undetermined matter dimension analysis result; furthermore, the dimension difference value between the sample dimension and the second undetermined matter dimension analysis result can be obtained. If the dimension difference value exceeds the target difference value (which can be understood as a large difference value of the target matter under normal circumstances), the important feature description of the hazard in the second undetermined matter dimension analysis result is adjusted based on the sample dimension to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel data collection.

[0058] In the mobile scenario, the channel information dataset captured usually cannot cover the entire target matter, and the dimension of the target matter obtained through the regression analysis of the matter intelligent analysis thread is incomplete. Through processing such as the analysis of the important feature description of the hazard and the adjustment of the important feature description of the hazard, the rationality of the debris flow channel data collection result can be improved; through the local debris flow hazard point data, the position of the important feature description of the hazard matching the local analysis result of the first matter can be calculated, and the accuracy of the debris flow channel data collection result can be improved.

[0059] In the embodiment of the present application, by respectively performing matter dimension analysis and specific local dimension analysis on the target matter in the target debris flow channel data collection, the matter debris flow hazard point data for the target matter and the local debris flow hazard point data for the local analysis result of the first matter for the target matter can be obtained. Furthermore, based on the matter debris flow hazard point data, the local debris flow hazard point data, and the sample dimension, the dimension analysis of the target matter in the target debris flow channel data collection can be performed, the missing local important feature descriptions of the hazard in the target matter in the target debris flow channel data collection can be repaired, and the important feature descriptions of the hazard that do not conform to the sample dimension can be adjusted, which can ensure the integrity and rationality of the final debris flow channel data collection result of the target matter, and further improve the accuracy and reliability of the debris flow channel data collection result.

[0060] On this basis, a device for collecting debris flow channel data based on a drone is provided. The device includes: A data acquisition module, configured to acquire data of potential debris flow hazard points corresponding to the target debris flow channel acquisition data and data of local potential debris flow hazard points; the data of potential debris flow hazard points corresponding to the matter is matched with the target matter in the target debris flow channel acquisition data, and the data of local potential debris flow hazard points is matched with the first matter local analysis result of the target matter; A number acquisition module, configured to acquire a sample dimension matched with the target matter, count the number of contents of the first important hazard feature descriptions corresponding to the sample dimension, and the number of contents of the second important hazard feature descriptions corresponding to the data of potential debris flow hazard points corresponding to the matter; A result analysis module, configured to, when the number of contents of the first important hazard feature descriptions exceeds the number of contents of the second important hazard feature descriptions, analyze and process the data of potential debris flow hazard points corresponding to the matter according to the sample dimension to obtain a first pending matter dimension analysis result; A result acquisition module, configured to analyze and process the first pending matter dimension analysis result according to the data of local potential debris flow hazard points to obtain a debris flow channel data acquisition result corresponding to the target matter.

[0061] On the basis of the above, a debris flow channel data acquisition system based on an unmanned aerial vehicle is shown, including a processor and a memory that communicate with each other, and the processor is configured to read and execute a computer program from the memory to implement the above method.

[0062] On the basis of the above, a computer-readable storage medium is further provided, and a computer program stored thereon implements the above method when running.

[0063] In summary, based on the above solution, by respectively performing matter dimension analysis and specific local dimension analysis on the target matter in the target debris flow channel acquisition data, data of potential debris flow hazard points corresponding to the matter can be obtained, and data of local potential debris flow hazard points corresponding to the first matter local analysis result of the target matter can be obtained. Furthermore, based on the data of potential debris flow hazard points corresponding to the matter, the data of local potential debris flow hazard points, and the sample dimension, dimension analysis can be performed on the target matter in the target debris flow channel acquisition data, the missing local important hazard feature description content of the target matter in the target debris flow channel acquisition data can be repaired, the integrity and rationality of the final obtained debris flow channel data acquisition result corresponding to the target matter can be ensured, and thus the accuracy and reliability of the debris flow channel data acquisition result can be improved.

[0064] It should be understood that the systems and their modules shown above can be implemented in various ways. For example, in some embodiments, the systems and their modules can be implemented by hardware, software, or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art can understand that the above methods and systems can be implemented using computer-executable instructions and / or included in processor control code, such as providing such code on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and their modules of the present application can be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips and transistors, or programmable hardware devices such as field programmable gate arrays and programmable logic devices, but also by software executed by various types of processors, or by a combination of the above hardware circuits and software (for example, firmware).

[0065] It should be noted that the beneficial effects that may be produced by different embodiments are different. In different embodiments, the beneficial effects that may be produced can be any one or several combinations of the above, or any other beneficial effects that may be obtained.

Claims

1. A method for collecting debris flow channel data based on an unmanned aerial vehicle, characterized in that, Including: Obtaining the data of potential debris flow hazard points for matters corresponding to the collected data of the target debris flow gully, and the data of local potential debris flow hazard points; The data of potential debris flow hazard points for matters is matched with the target matters in the collected data of the target debris flow gully, and the data of local potential debris flow hazard points is matched with the first local analysis result of the target matters; Obtaining the sample dimensions matched with the target matters, counting the number of the first hazard important feature description contents corresponding to the sample dimensions, and the number of the second hazard important feature description contents corresponding to the data of potential debris flow hazard points for matters; When the number of the first hazard important feature description contents exceeds the number of the second hazard important feature description contents, analyzing and processing the data of potential debris flow hazard points for matters according to the sample dimensions to obtain the first pending matter dimension analysis result; Analyzing and processing the first pending matter dimension analysis result according to the data of local potential debris flow hazard points to obtain the result of collecting data of the debris flow gully corresponding to the target matters.

2. The method according to claim 1, wherein The obtaining the data of potential debris flow hazard points for matters corresponding to the collected data of the target debris flow gully, and the data of local potential debris flow hazard points, includes: Obtaining the collected data of the target debris flow gully from the gully information dataset, loading the collected data of the target debris flow gully into the matter intelligent analysis thread, and outputting the data of potential debris flow hazard points for matters corresponding to the collected data of the target debris flow gully through the matter intelligent analysis thread; Loading the collected data of the target debris flow gully into the local intelligent analysis thread, and outputting the data of local potential debris flow hazard points corresponding to the collected data of the target debris flow gully through the local intelligent analysis thread.

3. The method according to claim 2, wherein The loading the collected data of the target debris flow gully into the matter intelligent analysis thread, and outputting the data of potential debris flow hazard points for matters corresponding to the collected data of the target debris flow gully through the matter intelligent analysis thread, includes: Loading the collected data of the target debris flow gully into the matter intelligent analysis thread, obtaining the matter dimension features corresponding to the target matters in the collected data of the target debris flow gully through the matter intelligent analysis thread, and reading the first classification result corresponding to the matter dimension features; the first classification result is used to represent the types of local analysis results of the hazard important feature description contents of the target matters; Generating a first data intersection description set according to the first classification result and the matter key features of the collected data of the target debris flow gully; Obtaining the mean value of the description attributes corresponding to the first data intersection description set, and determining the position description content of the hazard important feature description contents in the target matters in the collected data of the target debris flow gully according to the mean value of the description attributes; Determining the data of potential debris flow hazard points for matters corresponding to the collected data of the target debris flow gully according to the types of local analysis results of the matters and the position description content.

4. The method according to claim 3, wherein The obtaining the matter dimension features corresponding to the target matters in the collected data of the target debris flow gully through the matter intelligent analysis thread, includes: Obtain the matter description features corresponding to the target matter in the collected data of the target debris flow gully in the intelligent analysis thread of the matter, and output the second classification result corresponding to the matter description features according to the classification unit in the intelligent analysis thread of the matter; Obtain the key matter features of the collected data of the target debris flow gully output by the target feature extraction unit in the intelligent analysis thread of the matter, and perform functional processing on the second classification result and the key matter features to obtain the second data intersection description set corresponding to the collected data of the target debris flow gully; Perform bucketing processing on the collected data of the target debris flow gully according to the second data intersection description set to obtain Z geological state description contents of the scope of the matter local analysis result, and obtain the local description features corresponding to the Z geological state description contents of the scope of the matter local analysis result respectively according to the intelligent analysis thread of the matter; Z is a positive integer; Integrate the matter description features and the local description features corresponding to the Z geological state description contents of the scope of the matter local analysis result into the matter dimension features.

5. The method according to claim 2, wherein Loading the collected data of the target debris flow gully into the local intelligent analysis thread, and outputting the local debris flow hidden danger point data corresponding to the collected data of the target debris flow gully through the local intelligent analysis thread, including: Load the collected data of the target debris flow gully into the local intelligent analysis thread, and analyze the first matter local analysis result of the target matter in the collected data of the target debris flow gully in the local intelligent analysis thread; If the first matter local analysis result is analyzed in the collected data of the target debris flow gully, obtain the geological state description content of the scope covering the first matter local analysis result from the collected data of the target debris flow gully, obtain the position of the local hidden danger important feature description content corresponding to the first matter local analysis result according to the geological state description content of the scope, and determine the local debris flow hidden danger point data corresponding to the collected data of the target debris flow gully based on the position of the local hidden danger important feature description content; If the first matter local analysis result is not analyzed in the collected data of the target debris flow gully, determine that the local debris flow hidden danger point data corresponding to the collected data of the target debris flow gully is a null value.

6. The method according to claim 5, wherein The above-mentioned if the first matter local analysis result is analyzed in the collected data of the target debris flow gully, obtain the geological state description content of the scope covering the first matter local analysis result from the collected data of the target debris flow gully, obtain the position of the local hidden danger important feature description content corresponding to the first matter local analysis result according to the geological state description content of the scope, and determine the local debris flow hidden danger point data corresponding to the collected data of the target debris flow gully based on the position of the local hidden danger important feature description content, including: If the first matter local analysis result is analyzed in the collected data of the target debris flow gully, filter the collected data of the target debris flow gully to obtain the geological state description content of the scope covering the first matter local analysis result; Obtain the local edge features corresponding to the geological state description content of the said range, and regressively analyze the position of the local hidden danger important feature description content corresponding to the local analysis result of the first matter based on the local edge features; Based on the position of the local hidden danger important feature description content, correlate the hidden danger important feature description content of the local analysis result of the first matter to obtain the local debris flow hidden danger point data corresponding to the target debris flow channel collection data.

7. The method according to claim 1, characterized in that Analyze and process the first undetermined matter dimension analysis result according to the local debris flow hidden danger point data to obtain the debris flow channel data collection result corresponding to the target matter, including: If the matter debris flow hidden danger point data covers the dimension of the local analysis result of the second matter and does not cover the dimension of the local analysis result of the third matter, then determine the first local dimension corresponding to the local analysis result of the third matter according to the position of the hidden danger important feature description content of the local analysis result of the first matter covered in the local debris flow hidden danger point data; the local analysis result of the second matter and the local analysis result of the third matter are in a mirror image relationship; Obtain the first local data volume of the local analysis result of the second matter in the first undetermined matter dimension analysis result, and determine the position of the hidden danger important feature description content of the local analysis result of the third matter according to the first local data volume and the first local dimension; Load the position of the hidden danger important feature description content of the local analysis result of the third matter into the first undetermined matter dimension analysis result to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel collection data.

8. The method according to claim 7, wherein The step of loading the position of the hidden danger important feature description content of the local analysis result of the third matter into the first undetermined matter dimension analysis result to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel collection data includes: Load the position of the hidden danger important feature description content of the local analysis result of the third matter into the first undetermined matter dimension analysis result to obtain the second undetermined matter dimension analysis result corresponding to the target matter in the target debris flow channel collection data, and obtain the dimension difference value between the sample dimension and the second undetermined matter dimension analysis result; If the dimension difference value exceeds the target difference value, then perform debugging on the hidden danger important feature description content of the second undetermined matter dimension analysis result based on the sample dimension to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel collection data.

9. The method according to claim 1, characterized in that, The target debris flow channel collection data is the a-th debris flow channel collection data in the channel information dataset, where a is a positive integer; the step of analyzing and processing the first undetermined matter dimension analysis result according to the local debris flow hidden danger point data to obtain the debris flow channel data collection result corresponding to the target matter includes: If the debris flow hazard point data of the said matters does not include the dimensions of the local analysis results of the second matter and the third matter, then according to the position of the hazard important feature description content of the local analysis result of the first matter covered in the local debris flow hazard point data, determine the second local dimension corresponding to the local analysis result of the second matter, and the third local dimension corresponding to the local analysis result of the third matter; the local analysis results of the second matter and the third matter are in a mirror image relationship; In the data collection of the (a - 1)th debris flow channel, obtain the second local data volume corresponding to the local analysis result of the second matter, and the third local data volume corresponding to the local analysis result of the third matter, and determine the position of the hazard important feature description content of the local analysis result of the second matter according to the second local data volume and the second local dimension; Determine the position of the hazard important feature description content of the local analysis result of the third matter according to the third local data volume and the third local dimension, and load the position of the hazard important feature description content of the local analysis result of the second matter and the position of the hazard important feature description content of the local analysis result of the third matter into the analysis result of the first undetermined matter dimension to obtain the debris flow channel data collection result corresponding to the target matter in the target debris flow channel data collection.

10. A debris flow channel data acquisition system based on an unmanned aerial vehicle, characterized in that, It includes a processor and a memory that communicate with each other, and the processor is configured to read and execute a computer program from the memory to implement the method according to any one of claims 1 - 9.

Images