NL2030915B1 - Time-space dynamic monitoring method of county soil erosion based on remote sensing and gis - Google Patents

Time-space dynamic monitoring method of county soil erosion based on remote sensing and gis Download PDF

Info

Publication number
NL2030915B1
NL2030915B1 NL2030915A NL2030915A NL2030915B1 NL 2030915 B1 NL2030915 B1 NL 2030915B1 NL 2030915 A NL2030915 A NL 2030915A NL 2030915 A NL2030915 A NL 2030915A NL 2030915 B1 NL2030915 B1 NL 2030915B1
Authority
NL
Netherlands
Prior art keywords
map
soil erosion
remote sensing
interpretation
erosion
Prior art date
Application number
NL2030915A
Other languages
Dutch (nl)
Inventor
Liao Zhiqiang
Hao Yonghe
Zhang Yu
Original Assignee
Univ Jiaying
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Univ Jiaying filed Critical Univ Jiaying
Priority to NL2030915A priority Critical patent/NL2030915B1/en
Application granted granted Critical
Publication of NL2030915B1 publication Critical patent/NL2030915B1/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • 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
    • G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/24—Earth materials

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Multimedia (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Astronomy & Astrophysics (AREA)
  • Remote Sensing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Linguistics (AREA)
  • Pit Excavations, Shoring, Fill Or Stabilisation Of Slopes (AREA)

Abstract

A time—space dynamic monitoring method. of county soil erosion based on remote sensing and GIS. The present disclosure collects a remote sensing image and soil erosion history data in past few years; determines a map spot in the soil erosion map and corresponding soil erosion type and erosion intensity; establishes a soil erosion vector; then extracts the remote sensing image within. a to—be—monitored. scope, and. generates an interpretation map; for the interpretation map, extracts a feature of a mapping domain as a discriminant feature, determines a final division plan of a pre—divided block of the interpretation map; uses the soil erosion vector as an output to establish a support vector machine discriminative model; obtains a updated remote sensing image of the to—be—monitored scope, and synthesizes a updated interpretation. map, performs discriminant monitoring for a soil l5 erosion. situation. of the to—be—monitored. scope is discriminated through the discriminative model.

Description

P1113/NLpd
TIME-SPACE DYNAMIC MONITORING METHOD OF COUNTY SOIL EROSION BASED
ON REMOTE SENSING AND GIS
TECHNICAL FIELD
The present disclosure relates to a monitoring method of soil erosion.
BACKGROUND ART
The existing soil erosion monitoring methods first obtain professional maps such as topographic maps, geological maps, geo- morphic maps, vegetation type maps, soil maps, desertification maps, soil erosion maps, land use maps, watershed boundary maps, and China soil erosion zoning maps and TM images; at the same time, hydrological and meteorological observation data of a site should be obtained: including hydrological and sediment data of a hydrological site, soil erosion observation data of an experi- mental station, sedimentation data of a sediment storage dam; and other relevant research reports.
After checking topographic map data and TM data provided by a data group, the topographic map data is marked with national boundaries, provincial boundaries, and county boundaries, and then administrative boundaries are corrected and modified based on a topographic map in new version. After passing the quality inspec- tion, control point selection and digital entry of administrative boundary and control points are carried out according to the TM images, and an image processing analysis system is used to perform geometric correction and mosaic processing on the TM images, and finally an integration of the provincial boundaries, county bound- aries and images is realized, processing results are recorded on a tape in units of provinces, and submitted to each job group for image interpretation and analysis.
Through the analysis of soil erosion intensity reference in- dex elements, and based on a soil erosion type and intensity clas- sification, a human-computer interaction technology is used to outline to obtain a soil erosion type and intensity distribution map and soil erosion type and intensity classification data.
Although this method has a high accuracy rate, this accuracy rate is based on the accurate interpretation of images by experts with rich experience. Therefore, in order to obtain an accurate soil erosion type and intensity distribution map, a large number of professional and technical personnel must be equipped to achieve through human-computer interaction.
SUMMARY
A time-space dynamic monitoring method of county soil erosion based on remote sensing and GIS, includes following steps: step 1, determining a to-be-monitored scope for counties di- vided by administrative regions across the country, collecting a remote sensing image and soil erosion history data within the to- be-monitored scope in past few years; step 2, extracting a soil erosion map in the soil erosion history data, determining a map spot in the soil erosion map and corresponding soil erosion type and erosion intensity and extract- ing a soil erosion situation in the soil erosion map; establishing a soil erosion vector (a, b, c¢) according to the soil erosion sit- uation, wherein a represents the erosion type, b represents an erosion type sub-category, and c represents the erosion intensity; representing by 0 if the erosion type has no erosion type sub- category, wherein the erosion type does not need to be divided in- to erosion intensity, that is, representing by 0 if there is no erosion intensity. step 3, extracting the remote sensing image within the to-be- monitored scope, and using GIS to synthesize the remote sensing image and corresponding watershed boundary maps, topographic maps and geological maps to generate an interpretation map; step 4, for the interpretation map within the to-be-monitored scope, according to a location and acreage of the map spot in the soil erosion map, marking an area with a same position and acreage as the map spot in the interpretation map as a mapping domain; ex- tracting color, shape, location, acreage, shadow, and texture in- formation of the mapping domain, and storing the color, shape, lo- cation, acreage, shadow, and texture information in a database as a discriminant feature; step 5, using an image recognition technology to divide the interpretation map to obtain a pre-divided block of the interpre- tation map, comparing the pre-divided block with the mapping do- main described in step 4, and setting a preset value q, adjusting a division plan through adjusting the determining feature, and us- ing the division plan as a final division plan of the pre-divided block of the interpretation map when a similarity between the map- ping domain and the pre-divided block is greater than the preset value qd; step 6, using the discriminant feature as an input and the soil erosion vector (a, b, c) as an output to establish a support vector machine discriminative model, and train the discriminative model, regarding the discriminative model as a qualified discrimi- native model when a discriminant accuracy rate of the discrimina- tive model is greater than w; otherwise , returning to step 4; step 7, obtaining a updated remote sensing image of the to- be-monitored scope, and synthesizing a updated interpretation map, using the final division plan of step 5 to divide the updated in- terpretation map, wherein the discriminative feature of each block after the updated interpretation map is divided is extracted and substituted into the qualified discriminative model, and determin- ing a soil erosion result corresponding to each block of the up- dated interpretation map and performing dynamic monitoring.
In order to realize dynamic and obvious monitoring, the pre- sent disclosure also includes: step 8, obtaining a soil erosion vector (a’, b’, c'’) of the updated interpretation map according to the soil erosion result of the updated interpretation map; for the to-be-monitored scope, pre-comparing a change between the soil erosion situation corresponding to an update period with the soil erosion situation at a certain time in the past, and ex- tract the soil erosion vector (a, b, c) corresponding to a certain time in the past; calculating a distance d between the soil erosion vector (a’, b’, Cc’) of the updated interpretation map and the soil erosion vector (a, b, c) corresponding to the certain time in the past;
according to the soil erosion vector (a', b', c') of the up- dated interpretation map and the distance d, performing the soil erosion situation to realize a time-space dynamic monitoring of soil erosion.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a flow schematic diagram of the present disclosure.
DETAILED DESCRIPTION OF THE EMBODIMENTS
Specific embodiment 1: the embodiment is described with ref- erence to FIG. 1.
A time-space dynamic monitoring method of county soil erosion based on remote sensing and GIS, includes following steps: step 1, determining a to-be-monitored scope for counties di- vided by administrative regions across the country, collecting a remote sensing image and soil erosion history data within the to- be-monitored scope in past few years; step 2, extracting a soil erosion map in the soil erosion history data, determining a map spot in the soil erosion map and corresponding soil erosion type and erosion intensity and extract- ing a soil erosion situation in the soil erosion map; establishing a soil erosion vector (a, b, c) according to the soil erosion sit- uation, wherein a represents the erosion type, b represents an erosion type sub-category, and c represents the erosion intensity; representing by 0 if the erosion type has no erosion type sub- category, wherein the erosion type does not need to be divided in- to erosion intensity, that is, representing by 0 if there is no erosion intensity. step 3, extracting the remote sensing image within the to-be- monitored scope, and using GIS to synthesize the remote sensing image and corresponding watershed boundary maps, topographic maps and geological maps to generate an interpretation map; step 4, for the interpretation map within the to-be-monitored scope, according to a location and acreage of the map spot in the soil erosion map, marking an area with a same position and acreage as the map spot in the interpretation map as a mapping domain; ex- tracting color, shape, location, acreage, shadow, and texture in-
formation of the mapping domain, and storing the color, shape, lo- cation, acreage, shadow, and texture information in a database as a discriminant feature; step 5, using an image recognition technology to divide the 5 interpretation map to obtain a pre-divided block of the interpre- tation map, comparing the pre-divided block with the mapping do- main described in step 4, and setting a preset value q, adjusting a division plan through adjusting the determining feature, and us- ing the division plan as a final division plan of the pre-divided block of the interpretation map when a similarity between the map- ping domain and the pre-divided block is greater than the preset value di step 6, using the discriminant feature as an input and the soil erosion vector (a, b, ¢) as an output to establish a support vector machine discriminative model, and train the discriminative model, regarding the discriminative model as a qualified discrimi- native model when a discriminant accuracy rate of the discrimina- tive model is greater than w; otherwise, returning to step 4; step 7, obtaining a updated remote sensing image of the to- be-monitored scope, and synthesizing a updated interpretation map, using the final division plan of step 5 to divide the updated in- terpretation map, wherein the discriminative feature of each block after the updated interpretation map is divided is extracted and substituted into the qualified discriminative model, and determin- ing a soil erosion result corresponding to each block of the up- dated interpretation map and performing dynamic monitoring.
Specific embodiment 2:
The time-space dynamic monitoring method of county soil ero- sion based on remote sensing and GIS further includes: step 8, obtaining a soil erosion vector (a’, b’, cf) of the updated interpretation map according to the soil erosion result of the updated interpretation map; for the to-be-monitored scope, pre-comparing a change between the soil erosion situation corresponding to an update period with the soil erosion situation at a certain time in the past, and ex- tract the soil erosion vector (a, b, c¢) corresponding to a certain time in the past;
calculating a distance d between the soil erosion vector (a’, b’, cc’) of the updated interpretation map and the soil erosion vector (a, b, c¢) corresponding to the certain time in the past; according to the soil erosion vector (a', b', c') of the up- dated interpretation map and the distance d, performing the soil erosion situation to realize a time-space dynamic monitoring of soil erosion.
Other steps and parameters are the same as the specific em- bodiment 1.
Specific embodiment 3:
The specific process of step 5 in the embodiment is as fol- lows: step 5.1, using the image recognition technology to divide the interpretation map to obtain the pre-divided block of the in- terpretation map; step 5.2, comparing number of pre-divided block and the map- ping domain; if the number is not the same, adjusting the color, shape, location, acreage, shadow, and texture information of each mapping domain, and returning to step 5.1; otherwise, executing step 5.3; step 5.3, comparing whether the location of each pre-divided block and each mapping domain is the same, if the location is not the same, returning to step 5.1, if t the location is the same, executing step 5.4; step 5.4, comparing the shape and acreage of the pre-divided block and the mapping domain at the same position, when a shape similarity and an acreage similarity of the pre-divided block and the mapping domain are both greater than the preset value g, using the division plan as the final division plan of the pre-divided block of the interpretation map; otherwise, returning to step 5.1.
Other steps and parameters are the same as the specific em- bodiment 1 or embodiment 2.
Specific embodiment 4:
The specific process of step 5.3 in the embodiment is as fol- lows: calculating a geographic coordinate of a grid of the pre- divided block in the GIS, calculating another geographic coordi-
nate of the grid of the mapping domain in the GIS, and calculating the number of grids where the geographic coordinate of the grid of the pre-divided block is the same as the geographic coordinate of the grid of the mapping domain; if the number of grids occupies more than 97% of the number of grids in the pre-divided block and more than 97% of the number of grids in the mapping domain, con- sidering that the locations of the pre-divided block and the map- ping domain are the same, and returning step 5.4; otherwise, con- sidering different, and returning to step 5.1.
Other steps and parameters are the same as the specific em- bodiment 3.
Specific embodiment 5:
The specific process of step 5.4 in the embodiment is as fol- lows: selecting the pre-divided block and the mapping domain corre- sponding to its location, calculating an acreage size of the two, and determining an acreage similarity of the pre-divided block and the mapping domain corresponding to its location according to the acreage size of the two; calculating a boundary line similarity of the pre-divided block and the mapping domain at the same time, and using the boundary line similarity as a shape similarity; and make the acreage similarity and the shape similarity greater than the set preset value g; at this time, use image recognition technology to divide the interpretation map The division plan is the final division plan; otherwise, return to step 5.1; at this point, re- garding the division plan that uses the image recognition technol- ogy to divide the interpretation map as the final division plan; otherwise, returning to step 5.1.
Other steps and parameters are the same as the specific em- bodiment 4.
Specific embodiment 6:
The remote sensing image described in the embodiment is a TM remote sensing image after correction.
Other steps and parameters are the same as the specific em- bodiment 5.

Claims (1)

CONCLUSIESCONCLUSIONS 1. Tijd-ruimte dynamische monitoringmethode van bodemerosie op ba- sis van teledetectie en GIS, omvattende de volgende stappen: stap 1, het bepalen van een te monitoren bereik, verzamelen van een teledetectiebeeld en historische gegevens van bodemerosie binnen de te monitoren bereik in de afgelopen jaren; stap 2, het extraheren van een bodemerosiekaart in de historische gegevens van bodemerosie, het bepalen van een kaartplek in de bo- demerosiekaart en het bijbehorende bodemerosietype en erosie- intensiteit en het extraheren van een bodemerosiesituatie in de bodemerosiekaart; stap 3, het extraheren van het teledetectiebeeld binnen het te mo- nitoren bereik, en het gebruiken van GIS om het teledetectiebeeld en de bijbehorende stroomgebiedgrenskaarten, topografische kaarten en geologische kaarten te synthetiseren om een interpretatiekaart te genereren; stap 4, volgens een locatie en areaal van de kaartplek in de bo- demerosiekaart, het markeren van een gebied met dezelfde positie en areaal als de kaartplek in de interpretatiekaart als een karte- ringsdomein; een kenmerk van het mappingdomein extraheren als een onderscheidend kenmerk; stap 5, het gebruiken van een beeldherkenningstechnologie om de interpretatiekaart te verdelen om een vooraf verdeeld blok van de interpretatiekaart te verkrijgen, het vooraf verdeelde blok verge- lijken met het mappingdomein beschreven in stap 4, en een vooraf ingestelde waarde q instellen, het aanpassen van een indelingsplan door het bepalende kenmerk aan te passen, en het indelingsplan te gebruiken als een definitief indelingsplan van het vooraf verdeel- de blok van de interpretatiekaart wanneer een overeenkomst tussen het afbeeldingsdomein en het vooraf verdeelde blok groter is dan de vooraf ingestelde waarde gq; stap 6, het gebruikt maken van de discriminantfunctie als invoer en een bodemerosievector (a, b, c) als uitvoer om een discrimine- rend model voor de ondersteuningsvectormachine vast te stellen en het discriminerende model te trainen, waarbij het discriminerende model wordt beschouwd als een gekwalificeerd discriminerend model wanneer een discriminant nauwkeurigheidspercentage van het discri- minerende model groter is dan w; anders terugkeren naar stap 4; stap 7, het verkrijgen van een bijgewerkt remote sensing-beeld van de te bewaken bereik, en het samenstellen van een bijgewerkte in- terpretatiekaart, gebruikmakend van het definitieve indelingsplan van stap 5 om de bijgewerkte interpretatiekaart te verdelen, waar- bij het onderscheidende kenmerk van elk blok na de bijgewerkte in- terpretatiekaart is verdeeld, wordt geëxtraheerd en vervangen in het gekwalificeerde discriminerende model, en het bepalen van een bodemerosieresultaat dat overeenkomt met elk blok van de bijge- werkte interpretatiekaart en het uitvoeren van dynamische monito- ring.1. Time-space dynamic monitoring method of soil erosion based on remote sensing and GIS, comprising the following steps: step 1, determining a range to be monitored, collecting a remote sensing image and historical data of soil erosion within the range to be monitored in the In recent years; step 2, extracting a soil erosion map in the soil erosion historical data, determining a map spot in the soil erosion map and the corresponding soil erosion type and erosion intensity, and extracting a soil erosion situation in the soil erosion map; step 3, extracting the remote sensing image within the range to be monitored, and using GIS to synthesize the remote sensing image and associated river basin boundary maps, topographic maps, and geologic maps to generate an interpretation map; step 4, according to a location and area of the map spot in the soil erosion map, marking an area having the same position and area as the map spot in the interpretation map as a mapping domain; extract a feature from the mapping domain as a distinguishing feature; step 5, using an image recognition technology to divide the interpretation map to obtain a pre-divided block of the interpretation map, comparing the pre-divided block with the mapping domain described in step 4, and setting a preset value q, adjusting a mapping plan by adjusting the determinant, and using the mapping plan as a final mapping plan of the pre-divided block of the interpretation map when a match between the mapping domain and the pre-divided block is greater than the preset value gq; step 6, using the discriminant function as input and a soil erosion vector (a, b, c) as output to establish a discriminant model for the support vector machine and train the discriminant model, considering the discriminant model as a qualified discriminatory model when a discriminant accuracy percentage of the discriminatory model is greater than w; otherwise return to step 4; step 7, obtain an updated remote sensing image of the range to be monitored, and construct an updated interpretation map, using the final layout plan from step 5 to distribute the updated interpretation map, where the distinguishing feature of each block after the updated interpretation map is divided, extracted and replaced in the qualified discriminatory model, and determining a soil erosion result corresponding to each block of the updated interpretation map and performing dynamic monitoring.
NL2030915A 2022-02-11 2022-02-11 Time-space dynamic monitoring method of county soil erosion based on remote sensing and gis NL2030915B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
NL2030915A NL2030915B1 (en) 2022-02-11 2022-02-11 Time-space dynamic monitoring method of county soil erosion based on remote sensing and gis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
NL2030915A NL2030915B1 (en) 2022-02-11 2022-02-11 Time-space dynamic monitoring method of county soil erosion based on remote sensing and gis

Publications (1)

Publication Number Publication Date
NL2030915B1 true NL2030915B1 (en) 2023-08-18

Family

ID=87654405

Family Applications (1)

Application Number Title Priority Date Filing Date
NL2030915A NL2030915B1 (en) 2022-02-11 2022-02-11 Time-space dynamic monitoring method of county soil erosion based on remote sensing and gis

Country Status (1)

Country Link
NL (1) NL2030915B1 (en)

Similar Documents

Publication Publication Date Title
WO2022184088A1 (en) Flood sensitivity risk assessment method based on ensemble learning
CN109738970B (en) Method, device and storage medium for realizing lightning early warning based on lightning data mining
CN110490415B (en) A construction progress assessment method based on multi-UAV visual coordination
Leyk et al. A conceptual framework for uncertainty investigation in map‐based land cover change modelling
CN116337146A (en) Ecological quality evaluation and partitioning method and device based on improved remote sensing ecological index
CN111985389B (en) Basin similarity discrimination method based on basin attribute distance
CN111984701B (en) Country aggregation evolution prediction method, device, equipment and storage medium
CN108960622A (en) A kind of Assessment method of the reclaimed land in mining area quality based on remote sensing image
CN112329559A (en) Method for detecting homestead target based on deep convolutional neural network
CN105467100B (en) County's region soil based on remote sensing and GIS corrodes space-time dynamic monitoring method
CN110569523A (en) Soil Wind Erosion Model Establishment Method and Wind Erosion Rapid Estimation System
Zhang et al. Impervious surface extraction from high-resolution satellite image using pixel-and object-based hybrid analysis
CN113963268B (en) Construction state change monitoring method for large-scale engineering construction
CN107798357A (en) A kind of village land used reclamation planning analogy method based on KNN
CN118036785A (en) Model training method and model training device, storage medium and electronic equipment
Mansor et al. Optimization of land use suitability for agriculture using integrated geospatial model and genetic algorithms
NL2030915B1 (en) Time-space dynamic monitoring method of county soil erosion based on remote sensing and gis
Wang et al. [Retracted] Processing Methods for Digital Image Data Based on the Geographic Information System
CN117633068B (en) A construction land supply monitoring and supervision method and system
CN114842356B (en) A method, system and device for automatic generation of high-resolution surface type samples
CN115808690A (en) Intelligent revision method and system for weather forecast
CN115457386A (en) Village land informatization generation method
Song et al. Urban landscape modeling and algorithms under machine learning and remote sensing data
Nandi et al. Geographical Information System (GIS) in water resources engineering
Wang et al. Identification of Winter Wheat-Growing Areas Based on the XGBoost Algorithm