CN116994072B - Wetland extraction method, device, equipment and medium based on decision tree classification model - Google Patents
Wetland extraction method, device, equipment and medium based on decision tree classification model Download PDFInfo
- Publication number
- CN116994072B CN116994072B CN202311245510.0A CN202311245510A CN116994072B CN 116994072 B CN116994072 B CN 116994072B CN 202311245510 A CN202311245510 A CN 202311245510A CN 116994072 B CN116994072 B CN 116994072B
- Authority
- CN
- China
- Prior art keywords
- decision tree
- classification model
- band reflectivity
- component
- tree classification
- 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
- 238000003066 decision tree Methods 0.000 title claims abstract description 86
- 238000013145 classification model Methods 0.000 title claims abstract description 76
- 238000000605 extraction Methods 0.000 title claims abstract description 54
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 70
- 230000009466 transformation Effects 0.000 claims abstract description 43
- 238000000034 method Methods 0.000 claims abstract description 18
- 238000012544 monitoring process Methods 0.000 claims abstract description 7
- 238000002310 reflectometry Methods 0.000 claims description 107
- 238000012545 processing Methods 0.000 claims description 24
- 238000006243 chemical reaction Methods 0.000 claims description 17
- 238000000179 transient infrared spectroscopy Methods 0.000 claims description 17
- 238000007781 pre-processing Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 12
- 230000005855 radiation Effects 0.000 claims description 7
- 230000015572 biosynthetic process Effects 0.000 claims description 5
- 238000003786 synthesis reaction Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 3
- 241001414989 Thysanoptera Species 0.000 description 2
- 230000006837 decompression Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 241000422653 Thysanus Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013049 sediment Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
-
- 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
Abstract
The application provides a wetland extraction method, a device, equipment and a medium based on a decision tree classification model, and relates to the technical field of remote sensing images; the remote sensing image to be detected comprises Landsat8/9 remote sensing images; calculating a normalized vegetation index, an improved normalized difference water index, a humidity component and a greenness component of the spike cap transformation based on pre-selected sample data; constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation; and monitoring and analyzing the preprocessed remote sensing image to be detected through a decision tree classification model to obtain a wetland extraction result. The method solves the problems that the prior art mainly uses the extraction of water bodies such as lakes, river wetlands and the like as main materials, and the types of partial wetlands such as marshes and vegetation are easy to confuse, the extraction precision is not ideal enough, and the precision of wetland extraction is improved.
Description
Technical Field
The application relates to the technical field of remote sensing images, in particular to a wetland extraction method, device, equipment and medium based on a decision tree classification model.
Background
At present, when wetland extraction is carried out, conventional remote sensing data processing and analysis software such as ENVI, arcGIS and the like are adopted from data preprocessing to final result judgment. However, in the conventional method, when the regional remote sensing wetland is extracted, mainly the extraction of water bodies such as lakes, rivers and the like is mainly performed, and part of wetland types such as marshes and vegetation are easily confused, so that the precision of the wetland extraction is not ideal.
Disclosure of Invention
The purpose of the application is to provide a wetland extraction method, device, equipment and medium based on a decision tree classification model, which solves the problems that the prior art mainly uses the extraction of water bodies such as lakes, river wetlands and the like, and part of wetland types such as marshes and vegetation are easy to be confused, the extraction precision is not ideal enough, and the precision of wetland extraction is improved.
In a first aspect, the present invention provides a method for extracting wetland based on a classification model of a decision tree, comprising:
acquiring a remote sensing image to be detected, and preprocessing the remote sensing image to be detected; the remote sensing image to be detected comprises Landsat8/9 remote sensing images;
calculating a normalized vegetation index, an improved normalized difference water index, a humidity component and a greenness component of the spike cap transformation based on pre-selected sample data;
constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation;
and monitoring and analyzing the preprocessed remote sensing image to be detected through the decision tree classification model to obtain a wetland extraction result.
In an alternative embodiment, preprocessing the remote sensing image to be detected includes:
decompressing the remote sensing image to be detected to obtain decompressed OLI data and TIRS data;
performing radiometric calibration processing on the OLI data and the TIRS data, and performing atmospheric correction processing on the radiometric calibrated data;
and performing projection conversion, band synthesis and data clipping on the data of each band after the atmospheric correction to obtain the remote sensing image to be detected after the pretreatment is completed.
In an alternative embodiment, radiometric scaling of said OLI data and said TIRS data comprises:
performing radiometric calibration processing on the OLI data, wherein the calculation mode is as follows:
wherein,is the reflectivity of the top layer of the atmosphere; />Is the uncorrected atmospheric top layer planetary reflectivity,representing the offset; />Is the zenith angle of the sun; />Is the solar altitude;
performing radiometric calibration processing on the TIRS data, wherein the calculation mode is as follows:
wherein T is the brightness temperature of the sensor; lλ atmospheric top layer reflectivity, K1 and K2 are conversion constants.
In an alternative embodiment, calculating the normalized vegetation index, the improved normalized difference water index, the humidity component and the greenness component of the ear cap transform based on the pre-selected sample data comprises:
calculating a normalized vegetation index based on the red band reflectivity and the near infrared band reflectivity of the pre-selected sample data;
calculating an improved normalized difference water index based on the green light band reflectivity and the first short wave infrared band reflectivity of the pre-selected sample data;
the humidity component and the greenness component of the spike cap transformation are calculated based on the blue band reflectivity, the green band reflectivity, the red band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the pre-selected sample data.
In an alternative embodiment, calculating the humidity component and the greenness component of the ear cap transform based on the blue band reflectivity, the green band reflectivity, the red band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity, and the second short wave infrared band reflectivity of the pre-selected sample data comprises:
the humidity component of the ear cap transformation is calculated as follows:
the green component of the ear cap transformation is calculated as follows:
wherein,moisture component for ear cap conversion, +.>For the green component of the spike cap transformation, blue, green, red, nir, swir, swir2 are respectively the blue band reflectivity, the green band reflectivity, the red band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the sample data.
In an alternative embodiment, constructing a decision tree classification model from the normalized vegetation index, the improved normalized difference water index, the humidity component and the greenness component of the ear cap transform, comprises:
carrying out water body judgment according to the improved normalized difference water body index, and extracting water body areas of lakes and rivers;
dividing paddy fields and marsh wetlands based on the transformed humidity components of the leaf caps and the normalized vegetation indexes, and constructing a first-level branch of a decision tree classification model;
distinguishing vegetation-free coverage and vegetation coverage according to the improved normalized difference water body index and the transformed humidity component of the thysancap, and constructing a second-level branch of the decision tree classification model;
and dividing forests, grasslands and swamp wetlands according to the improved normalized difference water body index and the greenness component of the thysanoptera transformation in the second-level branches, and constructing second-level sub-branches of the decision tree classification model.
In an alternative embodiment, the method further comprises:
and packaging the decision tree classification model and integrating the decision tree classification model into a SMART client.
In a second aspect, the present invention provides a wetland extraction device based on a decision tree classification model, comprising:
the image preprocessing module is used for acquiring a remote sensing image to be detected and preprocessing the remote sensing image to be detected; the remote sensing image to be detected comprises Landsat8/9 remote sensing images;
the characteristic value calculation module is used for calculating a normalized vegetation index, an improved normalized difference water body index, a humidity component and a greenness component of spike cap transformation based on pre-selected sample data;
the decision tree model construction module is used for constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation;
and the wetland extraction module is used for monitoring and analyzing the preprocessed remote sensing images to be detected through the decision tree classification model to obtain a wetland extraction result.
In a third aspect, the present invention provides an electronic device comprising a processor and a memory storing computer executable instructions executable by the processor to implement the method for wetland extraction based on a decision tree classification model according to any one of the preceding embodiments.
In a fourth aspect, the present invention provides a computer readable storage medium storing computer executable instructions that, when invoked and executed by a processor, cause the processor to implement the method for wetland extraction based on a decision tree classification model according to any one of the preceding embodiments.
According to the wetland extraction method, device, equipment and medium based on the decision tree classification model, the constructed decision tree classification model is used for classifying and extracting the remote sensing images to be detected, so that the finally extracted wetland can be distinguished from vegetation information, the problems that the prior art mainly uses the extraction of water bodies such as lakes, river wetlands and the like, part of wetland types such as marshes and vegetation are easily confused, the extraction precision is not ideal enough are solved, and the precision of wetland extraction is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a wetland extraction method based on a decision tree classification model according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a classification model of a decision tree according to an embodiment of the present application;
FIG. 3 is a flowchart of another wetland extraction method based on a decision tree classification model according to an embodiment of the present application;
fig. 4 is a block diagram of a wetland extraction device based on a decision tree classification model according to an embodiment of the present application;
fig. 5 is a block diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
The embodiment of the application provides a wetland extraction method based on a decision tree classification model, which is shown in fig. 1 and mainly comprises the following steps:
step S110, acquiring a remote sensing image to be detected, and preprocessing the remote sensing image to be detected; the remote sensing image to be detected comprises Landsat8/9 remote sensing images.
In one embodiment, landsat8/9 remote sensing images may be downloaded from the USGS for image acquisition. The remote sensing image to be detected is preprocessed, and decompression, radiometric calibration, atmospheric correction, projection conversion, band synthesis and data clipping of Landast8/9 data can be realized by using python engineering.
Step S120, calculating a normalized vegetation index, an improved normalized difference water index, a humidity component and a greenness component of the ear cap transformation based on the pre-selected sample data.
In one embodiment, the sample point selection can be performed based on GEE (Google Earth Engine) platform, based on contemporaneous Landsat8/9 remote sensing image and near real-time 10m resolution Sentinel-2 land utilization classification data and Global 30 land utilization classification data in 2020, and combining relevant documents, and the land coverage type of the research area is divided into 7 types of water, woodland, grassland, wetland, cultivated land, construction land and bare land so as to select the sample data.
After the sample data is selected, a calculation of eigenvalues, i.e., normalized vegetation index (Normalized Difference Vegetation Index, NDVI), improved normalized difference water index (Modified Normalized Difference Water Index, MNDWI), moisture component and greenness component of the spike cap transform (Tasseled Cap Transformation, TCT), may be performed based on the pre-selected sample data.
And step S130, constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation.
And step S140, monitoring and analyzing the preprocessed remote sensing image to be detected through a decision tree classification model to obtain a wetland extraction result.
In one embodiment, the threshold may be dynamically adjusted to determine an optimal target model by the decision tree classification model, and wetland extraction may be performed by the thresholded decision tree classification model.
The following describes the means of implementation in detail for ease of understanding.
In an alternative embodiment, the preprocessing of the remote sensing image to be detected may include the following steps 1.1) to 1.3):
step 1.1), decompressing the remote sensing image to be detected to obtain decompressed OLI data and TIRS data. In particular implementations, landsat8/9-L1 level compressed package data may be downloaded from a public resource (such as the USGS), with decompression of the data being accomplished using a python third party library.
Step 1.2), carrying out radiometric calibration processing on the OLI data and the TIRS data, and carrying out atmospheric correction processing on the radiometric calibrated data. Specifically, the radiation calibration may be based on decompressed data, and the python code may be written to implement the radiation calibration of OLI data and TIRS data.
In one embodiment, performing radiation scaling processing on OLI data and TIRS data includes:
(1) Carrying out radiometric calibration processing on the OLI data, and converting DN values of the OLI data into the atmospheric top layer reflectivity by using reflectivity calibration coefficients of corresponding wave bands of a sensor of a metadata file by the radiometric calibration of the OLI data, wherein the calculation mode is as follows:
wherein,is the reflectivity of the top layer of the atmosphere; />Is the uncorrected atmospheric top layer planetary reflectivity,representing the offset; />Is the zenith angle of the sun; />Is the solar altitude. The parameters may be obtained from metadata.
(2) Because the brightness temperature of the TIRS sensor is converted from spectrum radiation through a thermal constant, the calculation mode of carrying out radiometric calibration processing on the TIRS data is as follows:
wherein T is the brightness temperature of the sensor; l (L) λ Reflectivity of atmosphere top layer, K 1 And K 2 Is a conversion constant.
And 1.3) performing projection conversion, band synthesis and data clipping processing on the atmospheric corrected data in each band to obtain a remote sensing image to be detected after pretreatment is completed. In practical application, the atmospheric correction can utilize python to call a Py6S packet of a 6S radiation transmission model to perform 6S atmospheric correction on the radiation-calibrated data; the projective transformation can utilize the third-party GDAL library of python to perform projective transformation-band synthesis-data clipping functions on the data of each band after the atmospheric correction.
In an alternative embodiment, calculating the normalized vegetation index, the improved normalized difference water index, the humidity component and the greenness component of the ear cap transform based on the pre-selected sample data comprises:
1) And calculating a normalized vegetation index based on the red band reflectivity and the near infrared band reflectivity of the pre-selected sample data. The normalized vegetation index (NDVI) is calculated as follows:
wherein, NDVI is normalized vegetation index, nir is near infrared band reflectivity, red is Red band reflectivity.
2) Calculating an improved normalized difference water index based on the green light band reflectivity and the first short wave infrared band reflectivity of the pre-selected sample data;
wherein MNCWI is an improved normalized difference water index, green is a Green wave band reflectivity, swir1 is a first short wave infrared band reflectivity.
Compared with the normalized difference water index NDWI, the improved normalized difference water index MNDWI can reveal water body micro-characteristics such as distribution of suspended sediment and change of water quality, so that the subsequently extracted wetland is more accurate, and confusion with vegetation data is avoided. In addition, the MNDWI can easily distinguish shadows from water bodies, and solves the problem that shadows are difficult to eliminate in water body extraction.
3) The humidity component and the greenness component of the spike cap transformation are calculated based on the blue band reflectivity, the green band reflectivity, the red band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the pre-selected sample data. In one example:
the humidity component of the ear cap transformation is calculated as follows:
the green component of the ear cap transformation is calculated as follows:
wherein,moisture component for ear cap conversion, +.>For the green component of the spike cap transformation, blue, green, red, nir, swir and Swir2 are respectively the blue light wave band reflectivity, the green light wave band reflectivity, the red light wave band reflectivity, the near infrared wave band reflectivity, the first short wave infrared wave band (1.57-1.65 mu m short wave infrared) reflectivity and the second short wave infrared wave band (2.1-2.35 mu m short wave infrared) reflectivity of the sample data.
The humidity component (KT_Wetness) obtained by remote sensing leaf-cap transformation is calculated, so that the humidity conditions of surface vegetation, water and soil can be better reflected, and the humidity is closely related to ecological environment changes such as soil degradation; the green degree component can reflect the vegetation coverage rate, the larger the green degree index is, the higher the vegetation coverage rate is, and the vegetation information is more abundant, so that the vegetation information can be extracted more accurately when a decision tree classification model is built later, the difference between the wetland and the vegetation is improved, and the extraction precision of the wetland information is improved.
For easy understanding, referring to the schematic structural diagram of the classification model of the decision tree shown in fig. 2, a mode of setting the model threshold is shown in the schematic diagram, and in practical application, the threshold may be adjusted according to practical situations, which is only an example and not limited in particular.
Further, the constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the ear cap transformation may include the following steps 2.1) to 2.4):
and 2.1) carrying out water body judgment according to the improved normalized difference water body index, and extracting water body areas of lakes and rivers. In one embodiment, the improved normalized difference water index is greater than or equal to 0 and is determined to be a water body and less than 0 is determined to be a non-water body by setting a threshold for partitioning.
And 2.2) dividing the paddy field and the marsh wetland based on the transformed humidity component of the thysanus cap and the normalized vegetation index, and constructing a first-level branch of the decision tree classification model. In one example, referring to FIG. 2, the decision tree branches to the right, MNDWI is used to distinguish between high water content land use types, and the combination of the tassel-cap transformed moisture component and NDVI is used to distinguish between paddy fields and wetlands.
And 2.3) distinguishing vegetation-free coverage and vegetation coverage according to the improved normalized difference water body index and the transformation humidity component of the tassel cap, and constructing a second-level branch of the decision tree classification model. In one example, MNTVI is used to combine with the tassel cap transformed humidity (kt_wet) component to distinguish between vegetation coverage and vegetation coverage of buildings, bare land and the like, and to construct the left branch of the decision tree model.
And 2.4), dividing forests, grasslands and swamp wetlands in the second-level branches according to the improved normalized difference water body index and the greenness component of the thysanoptera transformation, and constructing the second-level sub-branches of the decision tree classification model. In one example, on the basis of step 2.3), the MNDVI is reused in combination with a thysancap transform greenness (kt_greenness) component to distinguish between forests, grasslands and swamp wetlands with high vegetation coverage.
In one embodiment, the land is removed by combining the GlobLand30 land use coverage type with the result extracted in the above steps, and finally the extraction result of the natural wetland is obtained.
Further, the decision tree classification model may also be encapsulated and integrated into the SMART client. For example, python can be utilized to package the decision tree model into an algorithm plug-in and integrate the algorithm plug-in into a SMART client, so that the dynamic adjustment of each parameter in the decision tree model in a man-machine interaction mode is realized, and the result is directly output and displayed.
Further, as shown in fig. 3, a Landsat8/9 multispectral image is first obtained, and the image is preprocessed, including radiation calibration, atmospheric correction, projection conversion and region of investigation clipping; and further calculating characteristic values, including normalized vegetation indexes, improved normalized difference water indexes, humidity components and greenness components of spike cap transformation. And constructing a decision tree classification model, and performing parameter tuning to extract natural wetland from the Landsat8/9 multispectral image of the pretreated research area. During extraction, a threshold value is set by combining manual interaction, and the cultivated land is removed by combining the GlobLand30 land use coverage type, so that an extraction result of the natural wetland is finally obtained.
In conclusion, the embodiment of the application realizes Landsat8/9 remote sensing data preprocessing based on python engineering, and improves the data processing efficiency; by combining satellite monitoring analysis and a remote sensing application System (SMART), the method can set each parameter threshold in the decision tree model through man-machine interaction, extract the wetland range of images in different remote sensing, and meet the requirements of real scenes.
Based on the above method embodiment, the embodiment of the present application further provides a wetland extraction device based on a decision tree classification model, as shown in fig. 4, where the device mainly includes the following parts:
the image preprocessing module 410 is configured to obtain a remote sensing image to be detected, and perform preprocessing on the remote sensing image to be detected; the remote sensing image to be detected comprises Landsat8/9 remote sensing images;
a eigenvalue calculation module 420 for calculating a normalized vegetation index, an improved normalized difference water index, a humidity component and a greenness component of the spike cap transform based on pre-selected sample data;
the decision tree model construction module 430 is configured to construct a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation;
the wetland extraction module 440 is configured to monitor and analyze the preprocessed remote sensing image to be detected through the decision tree classification model, so as to obtain a wetland extraction result.
In a possible embodiment, the image preprocessing module 410 is further configured to:
decompressing the remote sensing image to be detected to obtain decompressed OLI data and TIRS data;
performing radiometric calibration processing on the OLI data and the TIRS data, and performing atmospheric correction processing on the radiometric calibrated data;
and performing projection conversion, band synthesis and data clipping on the data of each band after the atmospheric correction to obtain the remote sensing image to be detected after the pretreatment is completed.
In a possible embodiment, the image preprocessing module 410 is further configured to:
carrying out radiometric calibration processing on the OLI data, wherein the calculation mode is as follows:
wherein,is the reflectivity of the top layer of the atmosphere; />Is the uncorrected atmospheric top layer planetary reflectivity,representing the offset; />Is the zenith angle of the sun; />Is the solar altitude;
performing radiometric calibration processing on the TIRS data, wherein the calculation mode is as follows:
wherein T is the brightness temperature of the sensor; l (L) λ Reflectivity of atmosphere top layer, K 1 And K 2 Is a conversion constant.
In a possible embodiment, the feature value calculating module 420 is further configured to:
calculating a normalized vegetation index based on the red band reflectivity and the near infrared band reflectivity of the pre-selected sample data;
calculating an improved normalized difference water index based on the green light band reflectivity and the first short wave infrared band reflectivity of the pre-selected sample data;
the humidity component and the greenness component of the spike cap transformation are calculated based on the blue band reflectivity, the green band reflectivity, the red band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the pre-selected sample data.
In a possible embodiment, the feature value calculating module 420 is further configured to:
the humidity component of the ear cap transformation is calculated as follows:
the green component of the ear cap transformation is calculated as follows:
wherein,moisture component for ear cap conversion, +.>For the green component of the spike cap transformation, blue, green, red, nir, swir, swir2 are respectively the blue band reflectivity, the green band reflectivity, the red band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the sample data.
In a possible embodiment, the decision tree model building module 430 is further configured to:
carrying out water body judgment according to the improved normalized difference water body index, and extracting water body areas of lakes and rivers;
dividing paddy fields and marsh wetlands based on the transformed humidity components of the leaf caps and the normalized vegetation indexes, and constructing a first-level branch of a decision tree classification model;
distinguishing vegetation coverage and vegetation coverage according to the improved normalized difference water body index and the transformed humidity component of the thysancap, and constructing a second-level branch of the decision tree classification model;
and dividing forests, grasslands and swamp wetlands in the second-level branches according to the improved normalized difference water body index and the greenness component of the thysancap transformation again, and constructing the second-level sub-branches of the decision tree classification model.
In a possible embodiment, the above apparatus further includes a package integration module for:
and packaging the decision tree classification model and integrating the decision tree classification model into the SMART client.
The implementation principle and the generated technical effects of the wetland extraction device based on the decision tree classification model provided by the embodiment of the application are the same as those of the embodiment of the method, and for the sake of brief description, reference may be made to corresponding contents in the embodiment of the wetland extraction method based on the decision tree classification model, where the embodiment of the wetland extraction device based on the decision tree classification model is not mentioned.
The embodiment of the present application further provides an electronic device, as shown in fig. 5, which is a schematic structural diagram of the electronic device, where the electronic device 100 includes a processor 51 and a memory 50, where the memory 50 stores computer executable instructions that can be executed by the processor 51, and the processor 51 executes the computer executable instructions to implement any one of the above wetland extraction methods based on the decision tree classification model.
In the embodiment shown in fig. 5, the electronic device further comprises a bus 52 and a communication interface 53, wherein the processor 51, the communication interface 53 and the memory 50 are connected by the bus 52.
The memory 50 may include a high-speed random access memory (RAM, random Access Memory), and may further include a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The communication connection between the system network element and at least one other network element is achieved via at least one communication interface 53 (which may be wired or wireless), and the internet, wide area network, local network, metropolitan area network, etc. may be used. Bus 52 may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The bus 52 may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, only one bi-directional arrow is shown in FIG. 5, but not only one bus or type of bus.
The processor 51 may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in the processor 51 or by instructions in the form of software. The processor 51 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also digital signal processors (Digital Signal Processor, DSP for short), application specific integrated circuits (Application Specific Integrated Circuit, ASIC for short), field-programmable gate arrays (Field-Programmable Gate Array, FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present application may be embodied directly in hardware, in a decoded processor, or in a combination of hardware and software modules in a decoded processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in the memory, and the processor 51 reads the information in the memory, and combines the hardware thereof to complete the steps of the wetland extraction method based on the decision tree classification model in the foregoing embodiment.
The embodiment of the application also provides a computer readable storage medium, which stores computer executable instructions that, when being called and executed by a processor, cause the processor to implement the above wetland extraction method based on the decision tree classification model, and the specific implementation can refer to the foregoing method embodiment and will not be described herein.
The method, apparatus, device and medium for extracting wetland based on the decision tree classification model provided in the embodiments of the present application include a computer readable storage medium storing program codes, where the instructions included in the program codes may be used to execute the method described in the foregoing method embodiment, and specific implementation may refer to the method embodiment and will not be described herein.
The relative steps, numerical expressions and numerical values of the components and steps set forth in these embodiments do not limit the scope of the present application unless specifically stated otherwise.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In the description of the present application, it should be noted that the terms "first," "second," and the like are used merely to distinguish between descriptions and should not be construed as indicating or implying relative importance.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.
Claims (7)
1. A wetland extraction method based on a decision tree classification model is characterized by comprising the following steps:
acquiring a remote sensing image to be detected, and preprocessing the remote sensing image to be detected; the remote sensing image to be detected comprises Landsat8/9 remote sensing images;
calculating a normalized vegetation index, an improved normalized difference water index, a humidity component and a greenness component of the spike cap transformation based on pre-selected sample data;
constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation;
monitoring and analyzing the preprocessed remote sensing image to be detected through the decision tree classification model to obtain a wetland extraction result;
constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation, wherein the decision tree classification model comprises the following steps:
carrying out water body judgment according to the improved normalized difference water body index, and extracting water body areas of lakes and rivers;
dividing paddy fields and marsh wetlands based on the transformed humidity components of the leaf caps and the normalized vegetation indexes, and constructing a first-level branch of a decision tree classification model;
distinguishing vegetation-free coverage and vegetation coverage according to the improved normalized difference water body index and the transformed humidity component of the thysancap, and constructing a second-level branch of the decision tree classification model;
dividing forests, grasslands and swamp wetlands in the second-level branches according to the improved normalized difference water body index and the greenness component of the thysancap transformation again, and constructing second-level sub-branches of a decision tree classification model;
calculating a normalized vegetation index, an improved normalized difference water index, a humidity component and a greenness component of the ear cap transform based on pre-selected sample data, comprising:
calculating a normalized vegetation index based on the red band reflectivity and the near infrared band reflectivity of the pre-selected sample data;
calculating an improved normalized difference water index based on the green light band reflectivity and the first short wave infrared band reflectivity of the pre-selected sample data;
calculating humidity components and green components of the spike cap conversion based on the blue light band reflectivity, the green light band reflectivity, the red light band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the pre-selected sample data, specifically:
the humidity component of the ear cap transformation is calculated as follows:
the green component of the ear cap transformation is calculated as follows:
wherein,moisture component for ear cap conversion, +.>For the green component of the ear cap transformation, blue, green, red, nir, swir and Swir2 are respectively the blue light wave band reflectivity, the green light wave band reflectivity, the blue light wave band reflectivity, the green light wave band reflectivity and the blue light wave band reflectivity of the sample data,Red light band reflectivity, near infrared band reflectivity, first short wave infrared band reflectivity and second short wave infrared band reflectivity.
2. The method for extracting the wetland based on the classification model of the decision tree according to claim 1, wherein preprocessing the remote sensing image to be detected comprises the following steps:
decompressing the remote sensing image to be detected to obtain decompressed OLI data and TIRS data;
performing radiometric calibration processing on the OLI data and the TIRS data, and performing atmospheric correction processing on the radiometric calibrated data;
and performing projection conversion, band synthesis and data clipping on the data of each band after the atmospheric correction to obtain the remote sensing image to be detected after the pretreatment is completed.
3. The decision tree classification model based wetland extraction method according to claim 2, wherein performing radiometric scaling processing on said OLI data and said TIRS data comprises:
performing radiometric calibration processing on the OLI data, wherein the calculation mode is as follows:
wherein,is the reflectivity of the top layer of the atmosphere; />Is the uncorrected atmospheric top layer planetary reflectivity,，M ρ to scale the coefficient of radiation, Q cal For quantized and calibrated pixel values, A ρ Representing the offset; />Is the zenith angle of the sun; />Is the solar altitude;
performing radiometric calibration processing on the TIRS data, wherein the calculation mode is as follows:
wherein T is the brightness temperature of the sensor; l (L) λ Reflectivity of atmosphere top layer, K 1 And K 2 Is a conversion constant.
4. The decision tree classification model based wetland extraction method according to claim 1, wherein said method further comprises:
and packaging the decision tree classification model and integrating the decision tree classification model into a SMART client.
5. Wetland extraction device based on decision tree classification model, characterized by comprising:
the image preprocessing module is used for acquiring a remote sensing image to be detected and preprocessing the remote sensing image to be detected; the remote sensing image to be detected comprises Landsat8/9 remote sensing images;
the characteristic value calculation module is used for calculating a normalized vegetation index, an improved normalized difference water body index, a humidity component and a greenness component of spike cap transformation based on pre-selected sample data;
the decision tree model construction module is used for constructing a decision tree classification model according to the normalized vegetation index, the improved normalized difference water body index, the humidity component and the greenness component of the spike cap transformation;
the wetland extraction module is used for monitoring and analyzing the preprocessed remote sensing images to be detected through the decision tree classification model to obtain a wetland extraction result;
the decision tree model building module is further configured to:
carrying out water body judgment according to the improved normalized difference water body index, and extracting water body areas of lakes and rivers;
dividing paddy fields and marsh wetlands based on the transformed humidity components of the leaf caps and the normalized vegetation indexes, and constructing a first-level branch of a decision tree classification model;
distinguishing vegetation-free coverage and vegetation coverage according to the improved normalized difference water body index and the transformed humidity component of the thysancap, and constructing a second-level branch of the decision tree classification model;
dividing forests, grasslands and swamp wetlands in the second-level branches according to the improved normalized difference water body index and the greenness component of the thysancap transformation again, and constructing second-level sub-branches of a decision tree classification model;
the characteristic value calculation module is further used for calculating a normalized vegetation index based on the red-light wave band reflectivity and the near-infrared wave band reflectivity of the pre-selected sample data;
calculating an improved normalized difference water index based on the green light band reflectivity and the first short wave infrared band reflectivity of the pre-selected sample data;
calculating humidity components and green components of the spike cap conversion based on the blue light band reflectivity, the green light band reflectivity, the red light band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the pre-selected sample data, specifically:
the humidity component of the ear cap transformation is calculated as follows:
the green component of the ear cap transformation is calculated as follows:
wherein,moisture component for ear cap conversion, +.>For the green component of the spike cap transformation, blue, green, red, nir, swir, swir2 are respectively the blue band reflectivity, the green band reflectivity, the red band reflectivity, the near infrared band reflectivity, the first short wave infrared band reflectivity and the second short wave infrared band reflectivity of the sample data.
6. An electronic device comprising a processor and a memory, the memory storing computer executable instructions executable by the processor to implement the decision tree classification model-based wetland extraction method of any one of claims 1 to 4.
7. A computer readable storage medium storing computer executable instructions which, when invoked and executed by a processor, cause the processor to implement the decision tree classification model-based wetland extraction method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311245510.0A CN116994072B (en) | 2023-09-26 | 2023-09-26 | Wetland extraction method, device, equipment and medium based on decision tree classification model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311245510.0A CN116994072B (en) | 2023-09-26 | 2023-09-26 | Wetland extraction method, device, equipment and medium based on decision tree classification model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116994072A CN116994072A (en) | 2023-11-03 |
| CN116994072B true CN116994072B (en) | 2024-01-30 |
Family
ID=88525104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311245510.0A Active CN116994072B (en) | 2023-09-26 | 2023-09-26 | Wetland extraction method, device, equipment and medium based on decision tree classification model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116994072B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650689A (en) * | 2016-12-30 | 2017-05-10 | 厦门理工学院 | Coastal city time sequence land utilization information extracting method |
| CN111507303A (en) * | 2020-04-28 | 2020-08-07 | 同济大学 | Wetland plant species detection method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111122449A (en) * | 2019-12-31 | 2020-05-08 | 北京师范大学 | Urban impervious surface remote sensing extraction method and system |
-
2023
- 2023-09-26 CN CN202311245510.0A patent/CN116994072B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650689A (en) * | 2016-12-30 | 2017-05-10 | 厦门理工学院 | Coastal city time sequence land utilization information extracting method |
| CN111507303A (en) * | 2020-04-28 | 2020-08-07 | 同济大学 | Wetland plant species detection method |
Non-Patent Citations (3)
| Title |
|---|
| 一种基于植被指数的遥感影像决策树分类方法;潘琛;杜培军;罗艳;袁林山;;计算机应用(第03期);第777-780、797页 * |
| 利用决策树模型的湿地提取与分类;杨冬;李恩宝;张亚男;赵泉华;宋伟东;;辽宁工程技术大学学报(自然科学版)(第05期);全文 * |
| 基于决策树的高寒湿地类型遥感分类方法研究;邹文涛;张怀清;鞠洪波;刘华;;林业科学研究(第04期);第464-469页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116994072A (en) | 2023-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mondejar et al. | Near infrared band of Landsat 8 as water index: a case study around Cordova and Lapu-Lapu City, Cebu, Philippines | |
| Zhang et al. | A simple automated dynamic threshold extraction method for the classification of large water bodies from landsat-8 OLI water index images | |
| Berberoglu et al. | Assessing different remote sensing techniques to detect land use/cover changes in the eastern Mediterranean | |
| Chen et al. | Mapping forest and their spatial–temporal changes from 2007 to 2015 in tropical hainan island by integrating ALOS/ALOS-2 L-Band SAR and landsat optical images | |
| CN107145891B (en) | Water body extraction method and system based on remote sensing image | |
| CN106501186B (en) | A kind of soil moisture content product NO emissions reduction method | |
| CN115372282B (en) | Farmland soil water content monitoring method based on hyperspectral image of unmanned aerial vehicle | |
| Chen et al. | Estimation of vegetation coverage in semi-arid sandy land based on multivariate statistical modeling using remote sensing data | |
| CN111652092A (en) | Method for monitoring forest coverage change based on Sentinel-2A data | |
| Xu et al. | A novel automatic method on pseudo-invariant features extraction for enhancing the relative radiometric normalization of high-resolution images | |
| CN113486975A (en) | Ground object classification method, device, equipment and storage medium for remote sensing image | |
| CN114778483A (en) | Method for correcting terrain shadow of remote sensing image near-infrared wave band for monitoring mountainous region | |
| CN116824384A (en) | Soybean identification method based on standard curve | |
| Shirazinejad et al. | Applying multidate Sentinel-2 data for forest-type classification in complex broadleaf forest stands | |
| Verma et al. | Development of LR-PCA based fusion approach to detect the changes in mango fruit crop by using landsat 8 OLI images | |
| CN111175231B (en) | Inversion method and device of canopy vegetation index and server | |
| CN112257531B (en) | Remote sensing monitoring method for forest land change based on diversity feature combination | |
| CN116994072B (en) | Wetland extraction method, device, equipment and medium based on decision tree classification model | |
| Tan et al. | Estimation of leaf color variances of Cotinus coggygria based on geographic and environmental variables | |
| CN110489505B (en) | Method for identifying low cloud and large fog by dynamic threshold value method | |
| CN115631419B (en) | Rice planting area and spatial distribution extraction method and device based on change detection | |
| CN111693463B (en) | Antarctic peninsula optimized lichen coverage index extraction method | |
| Değermenci | Estimation of land surface temperature from Landsat-8 OLI: comparative analysis of two periods for Duzce in Turkey | |
| CN113392377A (en) | Swamp wetland remote sensing extraction method and system based on soil humidity | |
| Ren et al. | Surface water classification from GF-4 images using a time series water index |
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 |