CN111832430B - Mountain torrent disaster hyperspectral remote sensing image identification method and identification system thereof - Google Patents
Mountain torrent disaster hyperspectral remote sensing image identification method and identification system thereof Download PDFInfo
- Publication number
- CN111832430B CN111832430B CN202010580725.8A CN202010580725A CN111832430B CN 111832430 B CN111832430 B CN 111832430B CN 202010580725 A CN202010580725 A CN 202010580725A CN 111832430 B CN111832430 B CN 111832430B
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- aerial vehicle
- unmanned aerial
- hyperspectral
- sensing image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000012544 monitoring process Methods 0.000 claims abstract description 55
- 238000001228 spectrum Methods 0.000 claims abstract description 38
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 238000012360 testing method Methods 0.000 claims abstract description 5
- 230000003595 spectral effect Effects 0.000 claims description 32
- 239000002689 soil Substances 0.000 claims description 26
- 238000000701 chemical imaging Methods 0.000 claims description 25
- 238000009825 accumulation Methods 0.000 claims description 18
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 14
- 230000005540 biological transmission Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 7
- 239000011435 rock Substances 0.000 claims description 7
- 238000005259 measurement Methods 0.000 claims description 6
- 238000011161 development Methods 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 238000013500 data storage Methods 0.000 claims description 3
- 238000009795 derivation Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 14
- 230000002265 prevention Effects 0.000 description 9
- 238000002310 reflectometry Methods 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 5
- 238000006424 Flood reaction Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 208000027066 STING-associated vasculopathy with onset in infancy Diseases 0.000 description 2
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 2
- 230000003750 conditioning effect Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000012846 near infrared spectroscopy analysis technique Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012502 risk assessment Methods 0.000 description 2
- 229910052710 silicon Inorganic materials 0.000 description 2
- 239000010703 silicon Substances 0.000 description 2
- 241000272814 Anser sp. Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000005375 photometry Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000010792 warming Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Abstract
The invention discloses a mountain torrent disaster hyperspectral remote sensing image identification method. The method comprises the following steps: constructing an unmanned aerial vehicle remote sensing system: step two: preparing a manual and an unmanned aerial vehicle operation manual before flying; step three: testing the working state of an unmanned aerial vehicle-mounted miniature hyperspectral imager; step four: determining image recognition performance and parameters: step five: the unmanned aerial vehicle remote sensing image system acquires data on site: step six: the image recognition system carries out feature spectrum feature judgment and mountain torrent feature factor evolution analysis recognition: step seven: the subsystem of the identification method calculates and judges: step eight: and identifying the torrent characteristic factors, and outputting a torrent characteristic factor spatial distribution map of the torrent ditch in the monitoring area. The invention has the advantages of realizing the accuracy, the reliability and the timeliness of the basic data information of landslide and debris flow areas in mountain torrent disasters. The invention also discloses a mountain torrent disaster hyperspectral remote sensing image recognition system.
Description
Technical Field
The invention relates to the technical field of mountain torrent disaster monitoring and early warning, in particular to a mountain torrent disaster hyperspectral remote sensing image identification method. The invention also relates to a recognition system adopted by the mountain torrent disaster hyperspectral remote sensing image recognition method.
Background
Under the background of global warming and extreme weather frequency, the risks of natural disasters facing China in recent years are continuously increased, and the natural disasters are entering a natural disaster frequency period. Among the numerous disasters, mountain floods and secondary disasters thereof are the most serious flood disasters facing each year, and particularly, towns or residents in hilly areas are located at the openings of debris flow channels, along the coasts of river valleys and even on landslide bodies. The hilly area of China is about two thirds of the area of the country, and many areas lack debris flow and landslide disaster monitoring facilities induced by mountain floods, and the debris flow and landslide monitoring mainly depends on group detection and group prevention, so that dangerous points in areas where mountain floods are likely to occur are not monitored enough. The existing various field mountain torrent disaster monitoring, forecasting and early warning methods and equipment have the defects of low technological content, poor monitoring precision, untimely achievement and insufficient reliability, and become a prominent difficulty in the current flood control and disaster reduction work. Since the implementation of national mountain torrent disaster prevention and control planning in 2006, strengthening mountain torrent disaster prevention and control has become important content for natural disaster prevention and control, wherein monitoring and evaluating mountain torrent disaster-prone areas is one of key technologies for realizing effective mountain torrent disaster prevention and control, and a set of rapid monitoring, statistics, analysis and evaluation methods for mountain torrent disaster conditions are urgently needed for disaster prevention and reduction work in current mountain torrent disaster-prone areas.
In recent years, china has made certain progress in mountain torrent disaster monitoring and evaluation, and many advanced monitoring devices and technologies including satellite remote sensing technology, radar monitoring technology, unmanned aerial vehicle remote sensing monitoring technology and the like are applied, so that the mountain torrent disaster monitoring technology, particularly in a regional large-scale range, has improved early-stage forecasting and early-warning capability to a certain extent. However, satellites are restricted by the running orbit and the running period, radar detection has strict requirements on the monitoring environment, and the like, so that the application range of the existing technologies is limited, and the application fields of the monitoring and forecasting of the mountain torrent disaster prevention and control in advance, maneuverability, timeliness and the like cannot be met. The unmanned aerial vehicle monitoring technology with the advantages of convenience and rapidness in remote sensing becomes the first choice for developing real-time, dynamic and complex environment monitoring of large-scale mountain torrent disasters.
Unmanned aerial vehicle remote sensing is the third generation remote sensing platform behind traditional aviation remote sensing platform. Compared with the traditional aviation and aerospace remote sensing platform, the remote sensing platform has the advantages of low remote sensing data acquisition cost, strong safety operation guarantee capability, high remote sensing data precision, rapid emergency response capability, capability of acquiring data under cloud, and capability of realizing remote sensing monitoring in large area, long endurance, fixed point and fixed area. The hyperspectral imager combines an imaging technology and a spectrum technology, detects two-dimensional geometric space and one-dimensional spectrum information of a target, and acquires continuous and narrow-band image data (spectrum cube) with high spectral resolution. The miniature hyperspectral imager can focus clearly in the whole spectrum range, has high accuracy of photometric measurement, and is particularly suitable for collecting imaging spectrum data of plants, soil, minerals and the like. The hyperspectral remote sensing image system carried by the unmanned aerial vehicle platform is applied to the field of mountain torrent disaster monitoring, and can effectively improve the accuracy, reliability and timeliness of mountain torrent disaster basic data. However, the technology is still in a starting and exploring stage in mountain torrent disaster monitoring application in China.
Therefore, the development of the mountain torrent disaster hyperspectral remote sensing image identification method has a great application prospect.
Disclosure of Invention
The invention provides a mountain torrent disaster hyperspectral remote sensing image identification method, which is based on unmanned aerial vehicle hyperspectral remote sensing image identification technology, can be used for rapidly performing remote sensing monitoring in a mountain torrent disaster prone area, realizing characteristic spectrum identification and mountain torrent disaster factor calculation and discrimination of remote sensing images of landslide bodies, debris flow piles and debris flow ditch beds, and generating a characteristic factor space-time distribution map of the mountain torrent disaster prone area.
The invention aims to provide a mountain torrent disaster hyperspectral remote sensing image recognition system which can realize rapid remote sensing monitoring in a mountain torrent disaster prone area and realize remote sensing image recognition of landslide bodies, debris flow accumulation bodies and debris flow gully beds.
In order to achieve the first object of the present invention, the present invention has the following technical solutions: the identification method of the mountain torrent disaster hyperspectral remote sensing image identification system is characterized by comprising the following steps of: comprises the following steps of the method,
step one: constructing an unmanned aerial vehicle remote sensing system:
based on characteristics of an underlying area in a mountain torrent disaster prone area and characteristics of the mountain torrent disasters, an unmanned aerial vehicle remote sensing platform and a miniature hyperspectral imaging system are optimally combined to form an unmanned aerial vehicle remote sensing image system, and the unmanned aerial vehicle remote sensing image system meets the requirements of remote sensing image recognition performance and parameter selection of the miniature hyperspectral imager under flight conditions;
step two: preparing a working manual and an unmanned aerial vehicle operation manual before the unmanned aerial vehicle flies, and testing and standardizing a measurement and control system, a data transmission and dynamic monitoring support system of a remote sensing platform of the unmanned aerial vehicle;
step three: an operation manual of the unmanned aerial vehicle-mounted micro hyperspectral imager is compiled by combining with an unmanned aerial vehicle remote sensing platform, the working state of the unmanned aerial vehicle-mounted micro hyperspectral imager is tested, and the stability and the flexibility of an unmanned aerial vehicle remote sensing image system are determined;
step four: determining image recognition performance and parameters:
calibrating ground object spectral features by using a remote sensing image of an unmanned aerial vehicle-mounted miniature hyperspectral imager, analyzing reflection spectral features of different factors in a region in different environments, and establishing a regional factor spectral recognition method under dynamic conditions, thereby providing preconditions for identifying landslide, debris flow accumulation bodies and debris flow gully bed remote sensing images;
carrying out spectral characteristic monitoring of a typical mountain torrent ditch, and establishing a mountain torrent disaster characteristic factor database in a specific area;
step five: the unmanned aerial vehicle remote sensing image system acquires data on site:
acquiring and processing mountain torrent disaster site image information by an unmanned airborne micro hyperspectral imager, and transmitting the information to a mountain torrent disaster characteristic factor database in a wireless and airborne memory transmission recording mode;
step six: the image recognition system carries out feature spectrum feature judgment and mountain torrent feature factor evolution analysis recognition:
the image recognition system subsystem automatically classifies the spectral features of the torrent disaster factors according to the regional spectral feature database;
step seven: the subsystem of the identification method calculates and judges:
according to characteristic spectrums of different torrential flood disaster factors, identifying index sets of loose deposit, vegetation, soil and water in a subsystem of the identification method, calculating, distinguishing and determining all the factors;
step eight: and identifying the torrent characteristic factors by adopting normalized vegetation index and geographic information system software, identifying remote sensing images of landslide bodies, debris flow accumulation bodies and debris flow ditch beds, and outputting a torrent characteristic factor spatial distribution map of the torrent ditch in the monitoring area.
In the technical proposal, the method also comprises a step nine,
the step nine is specifically as follows: based on the identification of different factors, landslide body, debris flow accumulation body and debris flow ditch bed are monitored, and the monitoring results are analyzed, summarized and generalized by adopting a spectral evolution characteristic analysis method to study the occurrence and development rules of disasters such as landslide, debris flow and the like.
In the above technical solution, in step five, functions of the remote sensing system of the unmanned aerial vehicle include data storage, extraction, spectrum identification, analysis and derivation;
in step four, the factors include vegetation, rock, soil, and moisture content;
in step six, the torrential flood disaster factors include loose-packed bodies, vegetation, soil, and water.
In the above technical scheme, in the fifth step, the unmanned aerial vehicle remote sensing image system is composed of an unmanned aerial vehicle remote sensing platform and a micro airborne hyperspectral imaging system; the miniature airborne hyperspectral imaging system comprises an unmanned airborne miniature hyperspectral imager and airborne platform data acquisition software;
the unmanned aerial vehicle-mounted micro hyperspectral imager works on site, and the unmanned aerial vehicle remote sensing platform can select according to actual conditions in two modes of automatic route design and manual control aiming at different area flight operation conditions;
the remote sensing data image is automatically transmitted to an acquisition processing system of an unmanned aerial vehicle-mounted miniature hyperspectral imager on line and is written into a characteristic factor database at the same time.
In the technical scheme, the unmanned aerial vehicle-mounted miniature hyperspectral imager adopts a full-field hyperspectral imaging system.
In order to achieve the second object of the present invention, the present invention has the following technical scheme: mountain torrent calamity hyperspectral remote sensing image identification system, its characterized in that: the system comprises an unmanned aerial vehicle remote sensing image system, a characteristic factor database and a remote sensing image recognition system;
the unmanned aerial vehicle remote sensing system consists of an unmanned aerial vehicle remote sensing platform and a miniature airborne hyperspectral imaging system; the unmanned aerial vehicle remote sensing platform provides a real-time measurement and control system and a monitoring data transmission link for the micro airborne hyperspectral imaging system;
the miniature airborne hyperspectral imaging system comprises an unmanned airborne miniature hyperspectral imager and airborne platform data acquisition software;
the unmanned aerial vehicle remote sensing image system acquires and processes mountain torrent disaster site image information and transmits the mountain torrent disaster site image information to the characteristic factor database;
the image recognition system automatically classifies the characteristic factor database, and spectrum acquisition software interprets relevant parameters of remote sensing image data of the torrent disaster factors and calculates and determines corresponding indexes of the remote sensing image data.
In the technical scheme, the characteristic factor database comprises a factor spectrum identification subsystem, a landslide body debris flow accumulation body and a debris flow ditch bed remote sensing image identification subsystem under dynamic conditions.
In the technical scheme, the remote sensing image recognition system comprises a miniature hyperspectral imager monitoring subsystem, a reflection spectrum feature recognition subsystem and a spectrum evolution analysis feature recognition subsystem.
The invention has the following advantages:
(1) The micro hyperspectral imager and the matched equipment are matched with the unmanned aerial vehicle carrier for the first time, and are applied to monitoring and evaluation of regional mountain torrent disaster prone areas, so that the key technology of unmanned aerial vehicle-mounted hyperspectral remote sensing image identification is developed; the method has the advantages of extracting the spectral characteristics of the ground object with high precision and high efficiency, rapidly identifying landslide and debris flow areas in mountain torrent disasters and monitoring and evaluating the landslide and debris flow areas; the method solves the problems of narrow monitoring surface and low efficiency in the existing large-scale disaster monitoring, and avoids the technical problem of casualties caused by mountain torrent disaster monitoring in dangerous complex environments;
(2) The holographic reflection grating hyperspectral imaging technology is introduced, a large number of optical devices formed by parallel narrow slits with equal width and equal interval are arranged on the grating, so that the optical devices are uniform in light splitting and can be used in the whole spectrum band range; the unmanned aerial vehicle-mounted miniature hyperspectral imager adopts a grating light-splitting method, the photosensitive wavelength range of a silicon CCD is 300-1000nm, a C-T (Czerny-Turner) coaxial reflection optical design based on a holographic grating is adopted, the miniature hyperspectral imager is completely achromatic, and the technical advantages of no image distortion and spectral distortion, a convex reflection holographic grating and the like are realized;
(3) The invention is based on constructing a mountain torrent characteristic factor database, identifying and classifying each factor spectral characteristic, and establishing a regional mountain torrent disaster factor spectral characteristic database, so that the synchronous monitoring of mountain torrent characteristic factors and spectral characteristics under various weather conditions can be ensured to be rapidly and accurately judged, the identification precision and classification of various ground objects in a mountain torrent disaster monitoring region are greatly improved, and the invention has practical values in analysis and judgment of mountain torrent disaster prevention and disaster prevention risk assessment and the like;
(4) The unmanned aerial vehicle-mounted hyperspectral imager can realize automatic classification of bare soil and bare rock; the outstanding problems which cannot be solved by the traditional remote sensing image, the common camera and other equipment are solved (for example, the band range of the traditional remote sensing image and the common camera is small, and bare soil and bare rock are difficult to distinguish); the unmanned aerial vehicle-mounted hyperspectral imager reflects the vegetation spectral characteristics in detail and is used for distinguishing vegetation types and growth states; the unmanned aerial vehicle-mounted hyperspectral imager can distinguish dry soil and wet soil, can invert the water content of soil after calibration, and is used for monitoring the occurrence and development of mountain torrent disasters;
(5) According to characteristic spectrums of different mountain torrents disaster factors, a set of factor identification method and an index calculation formula are provided, and accurate calculation indexes, accurate analysis and judgment are carried out on loose deposit, vegetation, soil, water body and the like; the characteristic factors of the mountain torrent disaster area are rapidly identified by adopting ENVI and ArcGIS software, and a typical mountain torrent characteristic factor spatial distribution map of the mountain torrent is output; the technical problem of regional mountain torrent disaster monitoring and evaluating cores is solved, namely the accuracy, the reliability and the timeliness of basic data information of landslide and debris flow areas in mountain torrent disasters are realized.
Drawings
FIG. 1 is a schematic flow chart of the method of the invention.
FIG. 2 is a graph of the photosensitive wavelength range of a silicon CCD of the miniature hyperspectral imager of the present invention.
Fig. 3 is a partial torrent characterization factor database in an embodiment of the invention.
Fig. 4 is a graph of spectral characteristics of a typical mountain torrent trench land feature in an embodiment of the invention.
Fig. 5 is a partial torrent characterization factor spectral characterization database in an embodiment of the invention.
Figure 6 is a spatial distribution diagram of typical mountain torrents characteristic factors of a mountain torrent canal in an embodiment of the present invention.
In fig. 2, A1 represents a CCD response curve; a2 represents a blue channel; a3 represents a heat reflecting mirror; a4 represents the green channel; a5 represents a red channel.
In fig. 5, B1 represents vegetation; b2 represents rock; b3 represents dry soil; b4 represents wet soil.
In fig. 6, N represents north.
Detailed Description
The following detailed description of the invention is, therefore, not to be taken in a limiting sense, but is made merely by way of example. While making the advantages of the present invention clearer and more readily understood by way of illustration.
As can be seen with reference to the accompanying drawings: the identifying method of the mountain torrent disaster hyperspectral remote sensing image identifying system comprises the following steps,
step one: constructing an unmanned aerial vehicle remote sensing system: based on characteristics of an underlying area in a mountain torrent disaster prone area and characteristics of the mountain torrent disasters, an unmanned aerial vehicle remote sensing platform and a miniature hyperspectral imaging system are optimally combined to form an unmanned aerial vehicle remote sensing image system, and the unmanned aerial vehicle remote sensing image system meets the requirements of remote sensing image recognition performance and parameter selection of the miniature hyperspectral imager under flight conditions;
step two: preparing a working manual and an unmanned aerial vehicle operation manual before the unmanned aerial vehicle flies, and testing and standardizing a measurement and control system, a data transmission system, a dynamic monitoring support system and the like of a remote sensing platform of the unmanned aerial vehicle;
step three: an operation manual of the unmanned aerial vehicle-mounted micro hyperspectral imager is compiled by combining with an unmanned aerial vehicle remote sensing platform, the working state of the unmanned aerial vehicle-mounted micro hyperspectral imager is tested, and the stability and the flexibility of an unmanned aerial vehicle remote sensing image system are determined;
step four: determining image recognition performance and parameters: calibrating ground object spectral features by using a remote sensing image of an unmanned aerial vehicle-mounted miniature hyperspectral imager, analyzing reflection spectral features of different factors in a region in different environments, and establishing a regional factor spectral recognition method under dynamic conditions, thereby providing preconditions for identifying landslide, debris flow accumulation bodies and debris flow gully bed remote sensing images;
carrying out spectral characteristic monitoring of a typical mountain torrent ditch, and establishing a mountain torrent disaster characteristic factor database in a specific area;
step five: the unmanned aerial vehicle remote sensing image system acquires data on site: acquiring and processing mountain torrent disaster site image information by an unmanned airborne micro hyperspectral imager, and transmitting the information to a mountain torrent disaster characteristic factor database in a wireless and airborne memory transmission recording mode;
step six: the image recognition system carries out feature spectrum feature judgment and mountain torrent feature factor evolution analysis recognition: the image recognition system subsystem automatically classifies the spectral features of the torrent disaster factors according to the regional spectral feature database;
step seven: the subsystem of the identification method calculates and judges:
according to characteristic spectrums of different torrential flood disaster factors, identifying index sets of loose deposit, vegetation, soil and water in a subsystem of the identification method, calculating, distinguishing and determining all the factors;
step eight: and identifying the mountain torrent characteristic factors by adopting normalized vegetation index (ENVI) and geographic information system (ArcGIS) software, identifying remote sensing images of landslide bodies, debris flow accumulation bodies and debris flow ditch beds, and outputting a mountain torrent characteristic factor spatial distribution map of the mountain torrent ditch in the monitoring area.
Further, the method also comprises a step nine,
the step nine is specifically as follows: based on the identification of different factors, landslide body, debris flow accumulation body and debris flow ditch bed are monitored, and the monitoring results are analyzed, summarized and generalized by adopting a spectral evolution characteristic analysis method to study the occurrence and development rules of disasters such as landslide, debris flow and the like (shown in figure 1).
Further, in the fifth step, functions of the remote sensing system of the unmanned aerial vehicle include data storage, extraction, spectrum identification, analysis, derivation and the like;
in step four, the factors include vegetation, rock, soil, moisture content, and the like;
in step six, the torrential flood disaster factors include loose-packed bodies, vegetation, soil, water, and the like.
In the fifth step, the unmanned aerial vehicle remote sensing image system is composed of an unmanned aerial vehicle remote sensing platform and a micro airborne hyperspectral imaging system; the miniature airborne hyperspectral imaging system comprises an unmanned airborne miniature hyperspectral imager and airborne platform data acquisition software;
the unmanned aerial vehicle-mounted micro hyperspectral imager works on site to collect data images, and the unmanned aerial vehicle remote sensing platform can select according to actual conditions in two modes of automatic route design and manual control aiming at flight operation conditions of different areas;
the remote sensing data image acquired by the unmanned aerial vehicle-mounted micro hyperspectral imager is automatically transmitted to an acquisition processing system of an airborne platform data acquisition software on line, and is simultaneously written into a characteristic factor database;
the unmanned aerial vehicle-mounted miniature hyperspectral imager selects narrow slits (16 um-100 um) with various specifications and changeable widths according to actual application requirements, so that application requirements of different climates are met, and the spatial resolution and the spectral resolution of remote sensing images are improved.
Furthermore, the unmanned aerial vehicle-mounted miniature hyperspectral imager adopts a full-field hyperspectral imaging system;
the full-view hyperspectral imaging system selected by the unmanned aerial vehicle-mounted miniature hyperspectral imager generates two kinds of image deformation (minimum wedge deformation (Smile) and trapezoidal deformation (Keystone)) which are controlled in 1 pixel of the CCD (almost no deformation), and the full-view hyperspectral imaging system does not need to be corrected again when in use; the full-field hyperspectral imaging system actually uses the measured image deformation:
smile is less than or equal to 0.12pixels@1550nm, keystone is less than or equal to 0.04pixels, application requirements of different climates are met, and the spatial resolution and the spectral resolution of the remote sensing image are improved;
wherein CCD is a charge coupled device, which is a detection element for representing signal size by charge amount and transmitting signal by coupling mode, and has self-scanning and sensing functionsSpectrum of waveThe method has the advantages of wide range, small distortion, small volume, light weight, low system noise, small power consumption, long service life, high reliability and the like, and can be made into an assembly with very high integration level.
As can be seen with reference to the accompanying drawings: the mountain torrent disaster hyperspectral remote sensing image recognition system comprises an unmanned aerial vehicle remote sensing image system, a characteristic factor database and a remote sensing image recognition system;
the unmanned aerial vehicle remote sensing system consists of an unmanned aerial vehicle remote sensing platform and a miniature airborne hyperspectral imaging system; the unmanned aerial vehicle remote sensing platform provides a real-time measurement and control system and a monitoring data transmission link for the micro airborne hyperspectral imaging system;
the miniature airborne hyperspectral imaging system comprises an unmanned airborne miniature hyperspectral imager and airborne platform data acquisition software;
the unmanned aerial vehicle remote sensing image system acquires and processes mountain torrent disaster site image information and transmits the mountain torrent disaster site image information to the characteristic factor database;
the image recognition system automatically classifies the characteristic factor database, acquires relevant parameters of remote sensing image data of the software-interpreted mountain torrent disaster factor, and calculates and determines corresponding indexes of the remote sensing image data; identifying the torrent characteristic factors by adopting normalized vegetation index (ENVI) software and geographic information system (ArcGIS) software, and outputting a regional torrent ditch torrent characteristic factor space distribution map;
wherein, ENVI and ArcGIS are both the prior art.
Further, the characteristic factor database comprises a factor spectrum identification subsystem, a landslide body debris flow accumulation body and a debris flow ditch bed remote sensing image identification subsystem under a dynamic condition;
the monitoring area factor identification mainly comprises reflection spectrum characteristics, factor spectrum identification systems and the like of vegetation, rock, soil, water content and the like in a mountain torrent disaster prone area under different environments.
Further, the remote sensing image recognition system comprises a miniature hyperspectral imager monitoring subsystem, a reflection spectrum feature recognition subsystem and an analysis spectrum evolution feature analysis subsystem (shown in fig. 1).
Furthermore, the unmanned aerial vehicle-mounted miniature hyperspectral imager is an unmanned aerial vehicle-mounted miniature hyperspectral imager with high spatial resolution and ultrahigh imaging performance;
the resolution of the unmanned aerial vehicle-mounted miniature hyperspectral imager is 3.5nm;
the unmanned aerial vehicle-mounted micro hyperspectral imager comprises more than 200 wave bands (shown in figure 2) in NIR (900-1700 nm) and VNIR (400-1000 nm);
wherein, NIR and VNIR are both the prior art; NIR is a modern near infrared spectroscopic analysis technique; VNIR is a visible near infrared spectroscopic analysis technique.
The airborne platform data acquisition software is HDPU spectrum acquisition software; the device can acquire images at a faster frame rate so as to meet the shooting requirement of high-speed flying movement.
Examples
The present invention will now be described in detail with reference to the accompanying drawings by way of example only, and not by way of limitation. While making the advantages of the present invention clearer and more readily understood by way of illustration. The embodiment has guiding significance for rapid remote sensing monitoring of mountain torrent disasters in other regions, and realizing remote sensing image identification, monitoring and evaluation of landslide bodies, debris flow accumulation bodies and debris flow ditch beds.
The mountain torrent disaster hyperspectral remote sensing image identification method comprises the following steps:
step one: regional data collection
Based on the risk assessment requirement of mountain torrents in earthquake areas in the county of the Tibetan nationality of the Abamera of certain province, collecting the characteristics of the areas of the earthquake areas: geographic position, topography characteristic, geological characteristic, meteorological characteristic, mountain torrent disaster characteristic, carrying out mountain torrent disaster investigation and evaluation and reason analysis, and establishing a geographic information basic database;
step two: mountain torrent disaster characteristic factor database
Classifying and naming the aerial images according to features of land features, and distinguishing and describing the aerial images according to color tone, shape, texture and other features to establish a complete typical mountain torrent canal feature factor library (shown in figure 3) in a monitoring area;
step three: implementation steps of remote sensing image recognition method
S7.1: performing identification comparison on the spectrum characteristics of the ground object for carrying out typical mountain torrent canal monitoring investigation, and selecting and determining the representative spectrum of each factor (shown in figure 4);
s7.2: based on the remote sensing image spectrum cube, classifying to determine the names of the factors and extracting corresponding spectrums, and firstly establishing a spectrum characteristic database (shown in figure 5) of the factors;
s7.3: according to characteristic spectrums of different torrential flood disaster factors, determining an identification method of each factor, wherein the identification method of each factor is specifically as follows:
1) Loose stack identification
Soil conditioning vegetation index (SAVI) is used for identifying the loose accumulation bodies of the debris flow gully bed and the slope, L is 0.5 in the test (L can also take other values according to actual conditions), and the loose accumulation body identification method is as follows formula (1):
in the above formula (1): ρ NIR Is the reflectivity of the infrared band; ρ RED Is the red band reflectivity;
SAVI is a soil conditioning vegetation index;
l is a parameter which changes along with the density of vegetation, and the value range is 0-1;
2) Vegetation identification
Using ENVI software (ENVI (The Environment for Visualizing Images) is a complete remote sensing image processing platform, which is the prior art), a method for identifying vegetation using normalized vegetation index NDVI ((NDVI, normalized Difference Vegetation Index), formula (2) below:
in the above formula (2): ρ NIR Is the reflectivity of the infrared band, ρ RED Is the red band reflectivity;
NDVI V NDVI, which is the vegetation cover;
NDVI 0 NDVI, which is a non-vegetation cover portion;
NDVI is normalized vegetation index;
f v is vegetation coverage;
NDVI V and NDVI 0 The determination of two parameters is critical; in actual operation, due to lack of large-area ground surface measured data as reference, confidence interval is generally given according to histogram, and minimum and maximum values in the interval are used as NDVI V And NDVI 0 Values, or NDVI values at 5% and 95% frequencies, are taken as NDVI V And NDVI 0 A value;
this example uses the latter approach (i.e., taking NDVI values at 5% and 95% frequencies as NDVI) V And NDVI 0 A value);
in ENVI, the specific implementation steps are as follows:
(1) selecting a main menu, transforming, and calculating a normalized vegetation index NDVI;
(2) selecting a main menu- & gt Basic Tools- & gt Statistics- & gt Compute Statistics, and performing statistical analysis to obtain NDVI V And NDVI 0 A value;
(3) selecting a main menu, basic Tools, and Band Math, and calculating vegetation coverage f v ；
3) Soil identification
Soil index (SBI) for identifying soil, the method of identifying soil is as follows formula (3):
in the above formula (3): ρ NIR Is the reflectivity of the infrared band, ρ RED Is the red band reflectivity;
SBI is soil index;
4) Water body identification
Normalized water index (NDWI) for identifying a water body, the method of identifying a water body is as follows formula (4):
in the above formula (3): ρ RED Is the red band reflectivity;
NDWI is normalized water index;
step four: and identifying mountain torrent characteristic factors (the identified mountain torrent characteristic factors are shown in figure 6) based on seven ditches and wild goose ditches in the earthquake area of the county of the Tibetan nationality of the Equipped in the province by adopting ENVI and ArcGIS software, identifying landslide bodies, debris flow accumulation bodies and debris flow ditches in the earthquake area of the county of the Tibetan nationality of the Equipped in the province, and outputting a mountain torrent characteristic factor space distribution map of the mountain torrent ditches in the monitoring area.
Conclusion: the embodiment can realize rapid remote sensing monitoring and identification of remote sensing images of landslide bodies, debris flow accumulation bodies and debris flow ditch beds.
Other parts not described are known in the art.
Claims (8)
1. The mountain torrent disaster hyperspectral remote sensing image identification method is characterized by comprising the following steps of: comprises the following steps of the method,
step one: constructing an unmanned aerial vehicle remote sensing system:
based on characteristics of an underlying area in a mountain torrent disaster prone area and characteristics of the mountain torrent disasters, an unmanned aerial vehicle remote sensing platform and a miniature hyperspectral imaging system are optimally combined to form an unmanned aerial vehicle remote sensing image system, and the unmanned aerial vehicle remote sensing image system meets the requirements of remote sensing image recognition performance and parameter selection of the miniature hyperspectral imager under flight conditions;
step two: preparing a working manual and an unmanned aerial vehicle operation manual before the unmanned aerial vehicle flies, and testing and standardizing a measurement and control system, a data transmission and dynamic monitoring support system of a remote sensing platform of the unmanned aerial vehicle;
step three: an operation manual of the unmanned aerial vehicle-mounted micro hyperspectral imager is compiled by combining with an unmanned aerial vehicle remote sensing platform, the working state of the unmanned aerial vehicle-mounted micro hyperspectral imager is tested, and the stability and the flexibility of an unmanned aerial vehicle remote sensing image system are determined;
step four: determining image recognition performance and parameters:
calibrating ground object spectral features by using a remote sensing image of an unmanned aerial vehicle-mounted miniature hyperspectral imager, analyzing reflection spectral features of different factors in a region in different environments, and establishing a regional factor spectral recognition method under dynamic conditions, thereby providing preconditions for identifying landslide, debris flow accumulation bodies and debris flow gully bed remote sensing images;
carrying out spectral characteristic monitoring of a typical mountain torrent ditch, and establishing a mountain torrent disaster characteristic factor database in a specific area;
step five: the unmanned aerial vehicle remote sensing image system acquires data on site:
acquiring and processing mountain torrent disaster site image information by an unmanned airborne micro hyperspectral imager, and transmitting the information to a mountain torrent disaster characteristic factor database in a wireless and airborne memory transmission recording mode;
step six: the image recognition system carries out feature spectrum feature judgment and mountain torrent feature factor evolution analysis recognition:
the image recognition system subsystem automatically classifies the spectral features of the torrent disaster factors according to the regional spectral feature database;
step seven: the subsystem of the identification method calculates and judges:
according to characteristic spectrums of different torrential flood disaster factors, identifying index sets of loose deposit, vegetation, soil and water in a subsystem of the identification method, calculating, distinguishing and determining all the factors;
step eight: and identifying the torrent characteristic factors by adopting normalized vegetation index and geographic information system software, identifying remote sensing images of landslide bodies, debris flow accumulation bodies and debris flow ditch beds, and outputting a torrent characteristic factor spatial distribution map of the torrent ditch in the monitoring area.
2. The mountain torrent disaster hyperspectral remote sensing image identification method according to claim 1, wherein the method is characterized by comprising the following steps of: also comprises a step nine of the method,
the step nine is specifically as follows: based on the identification of different factors, landslide body, debris flow accumulation body and debris flow ditch bed are monitored, and the monitoring results are analyzed, summarized and generalized by adopting a spectral evolution characteristic analysis method to study the occurrence and development rules of disasters such as landslide, debris flow and the like.
3. The mountain torrent disaster hyperspectral remote sensing image identification method as claimed in claim 2, wherein the method comprises the following steps: in step four, the factors include vegetation, rock, soil, and moisture content;
in the fifth step, the functions of the unmanned aerial vehicle remote sensing system comprise data storage, extraction, spectrum identification, analysis and derivation;
in step six, the torrential flood disaster factors include loose-packed bodies, vegetation, soil, and water.
4. The mountain torrent disaster hyperspectral remote sensing image identification method as claimed in claim 3, wherein the method comprises the following steps: in the fifth step, the unmanned aerial vehicle remote sensing image system consists of an unmanned aerial vehicle remote sensing platform and a miniature airborne hyperspectral imaging system; the miniature airborne hyperspectral imaging system comprises an unmanned airborne miniature hyperspectral imager and airborne platform data acquisition software;
the unmanned aerial vehicle-mounted micro hyperspectral imager works on site, and the unmanned aerial vehicle remote sensing platform can select according to actual conditions in two modes of automatic route design and manual control aiming at different area flight operation conditions;
the remote sensing data image is automatically transmitted to an acquisition processing system of an unmanned aerial vehicle-mounted miniature hyperspectral imager on line and is written into a characteristic factor database at the same time.
5. The mountain torrent disaster hyperspectral remote sensing image identification method as claimed in claim 4, wherein the method comprises the following steps: the unmanned aerial vehicle-mounted miniature hyperspectral imager adopts a full-field hyperspectral imaging system.
6. The recognition system adopted by the mountain torrent disaster hyperspectral remote sensing image recognition method according to any one of claims 1 to 5, wherein the recognition system is characterized in that: the system comprises an unmanned aerial vehicle remote sensing image system, a characteristic factor database and a remote sensing image recognition system;
the unmanned aerial vehicle remote sensing system consists of an unmanned aerial vehicle remote sensing platform and a miniature airborne hyperspectral imaging system; the unmanned aerial vehicle remote sensing platform provides a real-time measurement and control system and a monitoring data transmission link for the micro airborne hyperspectral imaging system;
the miniature airborne hyperspectral imaging system comprises an unmanned airborne miniature hyperspectral imager and airborne platform data acquisition software;
the unmanned aerial vehicle remote sensing image system acquires and processes mountain torrent disaster site image information and transmits the mountain torrent disaster site image information to the characteristic factor database;
the image recognition system automatically classifies the characteristic factor database, acquires relevant parameters of remote sensing image data of the software-interpreted mountain torrent disaster factor, and calculates and determines corresponding indexes of the remote sensing image data.
7. The mountain torrent disaster hyperspectral remote sensing image recognition system of claim 6, wherein: the characteristic factor database comprises a factor spectrum identification subsystem, a landslide body debris flow accumulation body and a debris flow ditch bed remote sensing image identification subsystem under dynamic conditions.
8. The mountain torrent disaster hyperspectral remote sensing image recognition system of claim 7, wherein: the remote sensing image recognition system comprises a miniature hyperspectral imager monitoring subsystem, a reflection spectrum characteristic recognition subsystem and a spectrum evolution characteristic analysis subsystem.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010580725.8A CN111832430B (en) | 2020-06-23 | 2020-06-23 | Mountain torrent disaster hyperspectral remote sensing image identification method and identification system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010580725.8A CN111832430B (en) | 2020-06-23 | 2020-06-23 | Mountain torrent disaster hyperspectral remote sensing image identification method and identification system thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111832430A CN111832430A (en) | 2020-10-27 |
| CN111832430B true CN111832430B (en) | 2024-01-09 |
Family
ID=72898294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010580725.8A Active CN111832430B (en) | 2020-06-23 | 2020-06-23 | Mountain torrent disaster hyperspectral remote sensing image identification method and identification system thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111832430B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112801007B (en) * | 2021-02-07 | 2022-09-20 | 中国自然资源航空物探遥感中心 | Hot melt pond state identification method combining long time sequence data and machine learning algorithm |
| CN115290578B (en) * | 2022-08-15 | 2023-04-28 | 中国科学院空天信息创新研究院 | High-spectrum detection method and system for cohesive force of debris flow trench fine particle accumulation |
| CN115272874B (en) * | 2022-09-28 | 2022-12-09 | 中国科学院、水利部成都山地灾害与环境研究所 | Debris flow disaster identification and frequency calculation method based on remote sensing image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107195164A (en) * | 2017-07-03 | 2017-09-22 | 长江水利委员会长江科学院 | Mountain flood on-line monitoring identification method for early warning and its early warning system |
| CN110570462A (en) * | 2019-09-12 | 2019-12-13 | 中山大学 | flood inundation range automatic extraction method based on polarized radar remote sensing image |
| KR20200055388A (en) * | 2018-11-13 | 2020-05-21 | 한국건설기술연구원 | Information analysis system of rivers |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8775219B2 (en) * | 2012-07-13 | 2014-07-08 | Northrop Grumman Systems Corporation | Spectral image classification of rooftop condition for use in property insurance |
-
2020
- 2020-06-23 CN CN202010580725.8A patent/CN111832430B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107195164A (en) * | 2017-07-03 | 2017-09-22 | 长江水利委员会长江科学院 | Mountain flood on-line monitoring identification method for early warning and its early warning system |
| KR20200055388A (en) * | 2018-11-13 | 2020-05-21 | 한국건설기술연구원 | Information analysis system of rivers |
| CN110570462A (en) * | 2019-09-12 | 2019-12-13 | 中山大学 | flood inundation range automatic extraction method based on polarized radar remote sensing image |
Non-Patent Citations (1)
| Title |
|---|
| 舟曲特大山洪泥石流灾害遥感应急监测评估方法研究;李珊珊;宫辉力;范一大;陈世荣;温奇;胡卓玮;;农业灾害研究(01);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111832430A (en) | 2020-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111832430B (en) | Mountain torrent disaster hyperspectral remote sensing image identification method and identification system thereof | |
| Marti et al. | Mapping snow depth in open alpine terrain from stereo satellite imagery | |
| Polcyn | Monsoon flood boundary delineation and damage assessment using space borne imaging radar and Landsat data | |
| EP1248939B1 (en) | Method for determination of stand attributes and a computer program to perform the method | |
| Kääb et al. | The new remote-sensing-derived Swiss glacier inventory: II. First results | |
| Adams et al. | Multitemporal accuracy and precision assessment of unmanned aerial system photogrammetry for slope-scale snow depth maps in Alpine terrain | |
| CN109946235B (en) | Multilayer cloud inversion method of wind cloud 4A meteorological satellite multichannel scanning imaging radiometer | |
| CN111781146A (en) | Wave parameter inversion method using high-resolution satellite optical image | |
| CN114241331B (en) | Remote sensing modeling method for ground biomass of reed in wetland by taking UAV as ground and Septinel-2 medium | |
| Haack et al. | National land cover mapping by remote sensing | |
| Kruse | Mineral mapping using spectroscopy: from field measurements to airborne and satellite-based imaging spectrometry | |
| Brigante et al. | USE OF MULTISPECTRAL SENSORS WITH HIGH SPATIAL RESOLUTION FOR TERRITORIAL AND ENVIRONMENTAL ANALYSIS. | |
| CN114239379A (en) | Transmission line geological disaster analysis method and system based on deformation detection | |
| Jensen et al. | Remote sensing of impervious surfaces and building infrastructure | |
| Mahboubi et al. | Detection of Wet Riparian Areas using Very High Resolution Multispectral UAS Imagery Based on a Feature-based Machine Learning Algorithm | |
| King et al. | Evaluation of MK-4 multispectral satellite photography in land cover classification of eastern Ontario | |
| Mancini et al. | Unmanned Aerial System Applications to Coastal Environments | |
| Villa et al. | Multispectral transform and spline interpolation for mapping flood damages | |
| Hopkins | Remote Sensing and Spatial Variability of Leaf Area Index of Irrigated Wheat Fields | |
| Yuan et al. | On the Accuracy and Application of UAV-Based Coastal Monitoring | |
| Heimes et al. | Preliminary applications of Landsat images and aerial photography for determining land-use, geologic, and hydrologic characteristics: Yampa River basin, Colorado and Wyoming | |
| Root et al. | Identification of urban watershed units using remote multispectral sensing | |
| Xu et al. | Application of airborne laser radar technology of UAV based on river embankment survey | |
| Goksel et al. | Using satellite data for land use change detection: a case study for Terkos | |
| Borg et al. | Remote Sensing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |