CN116341932B - Tidal flat change monitoring method and system based on double remote sensing time sequence indexes - Google Patents
Tidal flat change monitoring method and system based on double remote sensing time sequence indexes Download PDFInfo
- Publication number
- CN116341932B CN116341932B CN202310628404.4A CN202310628404A CN116341932B CN 116341932 B CN116341932 B CN 116341932B CN 202310628404 A CN202310628404 A CN 202310628404A CN 116341932 B CN116341932 B CN 116341932B
- Authority
- CN
- China
- Prior art keywords
- time sequence
- change
- time
- remote sensing
- pixel
- 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
- 230000008859 change Effects 0.000 title claims abstract description 141
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000012544 monitoring process Methods 0.000 title claims abstract description 61
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 51
- 238000001514 detection method Methods 0.000 claims abstract description 13
- 238000012937 correction Methods 0.000 claims abstract description 12
- 230000003595 spectral effect Effects 0.000 claims abstract description 11
- 238000012216 screening Methods 0.000 claims abstract description 5
- 238000001228 spectrum Methods 0.000 claims description 17
- 238000002310 reflectometry Methods 0.000 claims description 14
- 238000011160 research Methods 0.000 claims description 13
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 6
- 238000012886 linear function Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 8
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000012549 training Methods 0.000 abstract description 5
- 238000007689 inspection Methods 0.000 abstract 1
- 238000004458 analytical method Methods 0.000 description 5
- 238000011835 investigation Methods 0.000 description 5
- 239000000047 product Substances 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 240000002044 Rhizophora apiculata Species 0.000 description 1
- 238000005162 X-ray Laue diffraction Methods 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000000985 reflectance spectrum Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- 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
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- 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
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Remote Sensing (AREA)
- Astronomy & Astrophysics (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a tidal flat change monitoring method and system based on double remote sensing time sequence indexes, wherein the method comprises the following steps: after spectral consistency inspection and correction are carried out on the full-time-sequence multispectral remote sensing image, MNDWI and MNDWI remote sensing indexes of each pixel of each image are calculated, variation time detection is carried out in cooperation with the MNDWI and MNDWI double remote sensing time sequence indexes, water coverage frequency of each section of sub-time sequence of each pixel is calculated, land area, tidal flat and water area type division is carried out according to the water coverage frequency, and finally pseudo-variation time is removed, so that remote sensing monitoring on tidal flat variation is completed. Aiming at the high-time-frequency monitoring requirement of tidal flat change under the influence of high-intensity human activities, the invention constructs the tidal flat change high-time-frequency monitoring method based on double remote sensing time sequence indexes, does not need training samples, does not need manual image screening, and has high degree of automation. The method can be used for evaluating the effect of the coast ecological restoration engineering, has great use value, greatly improves the application efficiency of the long-time-sequence multispectral remote sensing image, and is an innovation in the application of the remote sensing information technology.
Description
Technical Field
The invention belongs to the field of remote sensing technology application and coastal wetland monitoring, relates to a tidal flat remote sensing monitoring method based on time sequence multispectral remote sensing images, and particularly relates to a tidal flat change monitoring method and system based on double remote sensing time sequence indexes.
Background
The tidal flat is a tidal zone between the high tide level and the low tide level of the coastal large tide, and has important resource value and strategic status. As a front zone of sea-land interaction, the tidal flat provides important resources such as mangrove forest, salt marsh and the like, is an important component of blue carbon sink, has important ecological service functions of resisting storm surge disasters, purifying environment and the like, and has various social and economic values such as land making, cultivation, water storage, travel and the like. However, in the context of global climate change and high intensity human activity, the problems of tidal flat wetland area loss and ecological system degradation in coastal areas of China are increasingly prominent. With the development of coastal economies, population concentration and the acceleration of the urban process, tidal flat ecosystems are further at risk of loss of area or functional degradation. Therefore, development of a remote sensing monitoring method suitable for tidal flat change under the influence of high-intensity human activities is needed, and scientific basis is provided for ecological restoration of tidal flat wetland resources.
The tidal flat tidal furrows are densely distributed, the sea condition is complex, the accessibility is poor, the exposure time is short, the traditional ground investigation method is high in cost and low in efficiency, and the large-scale popularization and periodic update are difficult. The remote sensing technology has the advantages of wide observation range, strong timeliness, long time sequence and the like, and is an effective means for developing tidal flat monitoring. The traditional remote sensing method for extracting tidal flat is to select cloud-free clear remote sensing images shot at high tide and low tide moments, so as to extract the transition zone between instantaneous water edges at the high tide and the low tide moments. However, the number of clear remote sensing images without cloud coverage, which accords with the tide level conditions, is small due to the comprehensive influence of the relative fixed transit imaging time of the remote sensing satellites, the dynamic nature of tides and the rainy weather of coastal areas, so that the monitoring results of the change of the tide beach range are discontinuous in time and space and are not strong in consistency; meanwhile, the traditional method greatly depends on manual visual interpretation and discrimination, and is time-consuming, labor-consuming and low in automation degree. In summary, the traditional tidal flat remote sensing monitoring method cannot fully utilize the spectrum time sequence index information of the long time sequence remote sensing image, cannot timely and effectively master the change dynamics of the tidal flat, and restricts the effective development of works such as ecological restoration of the tidal flat. Therefore, the invention aims at the remote sensing monitoring requirement of tidal flat change under the influence of high-intensity human activities, builds a double remote sensing time sequence index based on the time sequence multispectral image, and provides a tidal flat change monitoring method based on the double remote sensing time sequence index through a variable point detection method and a time sequence remote sensing water coverage frequency analysis method.
Disclosure of Invention
The invention aims to provide a novel tidal flat change monitoring method based on double remote sensing time sequence indexes aiming at the defects of the prior art.
The invention is realized by the following technical scheme:
a tidal flat change monitoring method based on double remote sensing time sequence indexes comprises the following steps:
(1) And carrying out consistency check and correction on the spectrum values of different sensors. Screening out the earth surface reflectivity data of the time sequence multispectral remote sensing image according to the research time and the region, and correcting the spectral values of blue light wave bands, green light wave bands, red light wave bands, near infrared wave bands and short wave infrared wave bands of different sensors based on a linear function fitting and frequency distribution histogram matching method;
(2) The normalized difference water index NDWI (Normalized Difference Water Index) and the improved normalized difference water index MNDWI (Modified Normalized Difference Water Index) for each clear pixel of each image are calculated, and a time series of remote sensing indexes MNDWI and NDWI is constructed for each pixel of the study area.
(3) Carrying out change time detection in cooperation with the double remote sensing indexes, respectively acquiring the change time of the cluster mean value of the MNDWI index time sequences of each pixel, and dividing the time sequence into a plurality of sub-time sequences with unequal lengths by taking the change time as a dividing point;
(4) Based on the sub-time sequences obtained after the variable point detection, calculating the frequency of each sub-time sequence MNDWI >0 and NDWI >0 of each pixel as a water coverage frequency value, and developing the earth surface coverage classification: if more than 95% of NDWI values in the time sequence are more than 0, judging the NDWI values as water; if the MNCWI value is more than 0 and the frequency is less than 5%, judging that the land is the land; the rest is tidal flat.
(5) And comparing the earth surface coverage types of adjacent sub-time sequences of each pixel, and eliminating the pseudo-change time. Finally, the change time and the change type of each pixel are obtained, and the remote sensing monitoring of the tidal flat change is completed.
In the above technical solution, the method for performing spectrum consistency test and correction of different sensors in step (1) includes: first, roy, D P; kovalsky, V; zhang, H K; vermote, E F; yan, L; kumar, S S; the linear equation of Egorov, A,2016. Characterization of Landsat-7 to Landsat-8 reflective wavelength and normalized difference vegetation index continuity adjusts the spectral values of the various bands of Landsat OLI. Further, the consistency check and correction work of the spectrum values of different sensors are carried out by adopting a frequency distribution histogram matching method, so that the correlation of the remote sensing spectrum frequency distribution histograms of different sensors is more than 0.95. Thus, an image data set with good spectrum consistency of each wave band in a long time sequence is constructed.
Further, on the basis of (1), a normalized difference water index NDWI (Normalized Difference Water Index) and a modified normalized difference water index MNDWI (Modified Normalized Difference Water Index) for each clear pixel of each image are calculated. The specific calculation mode of the index is as follows:
wherein Green, NIR, SWIR is the earth surface reflectivity of the remote sensing image in the green band, near infrared band and short wave infrared band respectively. NDWI and MNDWI are both between-1 and 1. Further, a time sequence scatter diagram of a remote sensing index MNDWI and an NDWI is respectively constructed for each pixel of the research area and used for representing time sequence water index information of each pixel in the research area.
Further, on the basis of the step (2), the change time detection is carried out in cooperation with the double remote sensing indexes. The specific method comprises the following steps: firstly, respectively acquiring the change time of the cluster mean value of MNDWI and MNDWI for each pixel, and referring to a judgment method of the cluster mean value (Sen, A. And M.S. Srivastava, on tests for detecting change in mean, 1975: p.98-108.), wherein according to the change condition of the cluster mean value, the water index time sequence has multiple conditions of no change point, single change point, multiple change points and the like. Next, the time series is divided into a plurality of sub-time series of unequal lengths by using the detected change points as dividing points. The shortest length of all sub-time sequences is set to be not less than 365 days, and if a timing of less than 365 days exists after division, the timing is incorporated into adjacent timing in the vicinity. Further, when dividing the time sequence, detecting the change point of the MNDWI time sequence as a dividing point to divide the time sequence into a plurality of sub-time sequences with different lengths, detecting the change point of the NDWI time sequence, comparing the distance between the change point and the nearest MNDWI change point, and if the distance is more than 365 days, reserving the change point corresponding to the NDWI time sequence, otherwise discarding the change point.
Further, on the basis of the step (3), the water coverage frequency value of each sub-time sequence of each pixel is calculated, and the earth surface coverage classification is developed according to the water coverage frequency value. The specific classification method comprises the following steps: if the frequency of NDWI value greater than 0 in the time sequence is greater than 95%, judging the time sequence as water; if the MNCWI value is more than 0 and the frequency is less than 5%, judging that the land is the land; the rest are classified as tidal beaches. The earth coverage type of each sub-timing of each pixel is acquired so far.
Further, on the basis of the step (4), the earth surface coverage types of adjacent sub-time sequences of the pixels are sequentially compared, and if the earth surface coverage types of the adjacent sub-time sequences are the same, the change time between the adjacent sub-time sequences is corresponding to pseudo change, and the pseudo change needs to be removed. After all the pseudo-change time is removed, the earth surface coverage change time of each pixel and the corresponding change type of the earth surface coverage change time can be obtained, so that the annual tidal flat range is obtained.
Through the working flow, remote sensing monitoring of tidal flat change can be completed, and the annual tidal flat range can be obtained. The final data type is a multi-layer raster data stack, each layer corresponds to the beach range of each year, and the raster attribute values are land area, tidal beach and water area.
In addition, the invention also provides a tidal flat change monitoring system based on the double remote sensing time sequence indexes, which comprises the following steps:
the consistency check and correction module is used for carrying out linear adjustment and histogram matching on the surface reflectivity spectrum values of different sensors so that the correlation coefficient of the frequency distribution histograms of the different sensors is more than 0.95;
the double remote sensing index module is used for calculating MNDWI and NDWI remote sensing indexes of each pixel and constructing double remote sensing time sequence indexes of each pixel;
the variable point segmentation module is used for segmenting the time sequence according to the variable points of the double remote sensing index time sequence clustering mean value to obtain the sub-time sequence of each pixel;
the surface coverage classification module is used for calculating the water coverage frequency value of each sub-time sequence of each pixel and classifying the surface coverage;
and the pseudo-change removing module is used for removing the pseudo-change time according to the consistency of the surface coverage type of the adjacent sub-time sequences of each pixel.
Aiming at the high-time-frequency monitoring requirement of tidal flat change under the influence of high-intensity human activities, the invention constructs the tidal flat change high-time-frequency monitoring method based on double remote sensing time sequence indexes, and has the advantages of no need of training samples, no need of manually screening images, high degree of automation and the like. The method can be used for evaluating the effect of the coast ecological restoration project, provides scientific basis for the marine ecological restoration and the blue bay restoration, has great use value, greatly improves the application efficiency of the long-time-sequence multispectral remote sensing image, is an innovation in the application of the remote sensing information technology, and is a beneficial supplement to a coastal zone remote sensing information extraction method system.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention.
FIG. 2 is a graph showing spectral value consistency correction in the near infrared bands of Landsat 7 and Landsat 8.
FIG. 3 is a schematic diagram of a working area in an embodiment of the present invention.
FIG. 4 is a schematic diagram illustrating remote sensing timing index change point detection according to an embodiment of the present invention.
Figure 5 is a graph showing the annual change in tidal flat area of Yue, hong Kong and Australia in an investigation region in accordance with an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further described in detail below with reference to the accompanying drawings and specific examples.
According to the tidal flat change monitoring method based on double remote sensing time sequence indexes, a technical route diagram of the method is shown in fig. 1, and according to one specific embodiment of the invention, the method comprises the following steps:
and acquiring the surface reflectivity data of the multispectral image in the research area and the research time period. In the embodiment, landsat multispectral data is used as a data source, and the acquisition and processing work of multispectral remote sensing time sequence image data is carried out. All Landsat surface reflectivity level-1 data covering the study area is first obtained from a public database of Google Earth Engine (GEE). Further, the pixel containing cloud image is masked (specific methods can be referred to as Steve Foga, pat L. Scanamuzza, song Guo, zhe Zhu, ronald D. Dilley, tim Beckmann, gail, schmidt, john L. Dwyer, M. Joseph Hughes, brady Laue, cloud detection algorithm comparison and validation for operational Landsat data products, remote Sensing of Environment, volume 194, 2017, pages 379-390, ISSN 0034-4257), and all Landsat time-series surface reflectivity clear pixels are screened out.
And carrying out consistency check and correction work of the spectrum values of different sensors. First, the surface reflectance spectrum values (Roy, D P; kovalsky, V; zhang, H K; vermote, E F; yan, L; kumar, S S; egorov, A,2016. Characterization of Landsat-7 to Landsat-8 reflective wavelength and normalized difference vegetation index continuity) of the red, green, blue, near infrared and short infrared bands of Landsat8 OLI were adjusted using a linear function. And further carrying out consistency test and correction work on spectral values of each wave band of Landsat8 by adopting a frequency distribution histogram matching method. As shown in fig. 2, the near infrared band spectrum value of Landsat8 OLI is adjusted by using a frequency distribution histogram matching method, and the histogram correlation coefficient of the adjusted near infrared band spectrum values of Landsat8 and Landsat 7 is greater than 0.95. Through the technical process, the spectral consistency checking and correcting work of the surface reflectivity spectral values of the Landsat8 sensor in the red light wave band, the green light wave band, the blue light wave band, the near infrared wave band and the short wave infrared wave band is completed, so that a remote sensing image data set with good long-time sequence spectral consistency is constructed.
And carrying out time sequence remote sensing index calculation. A normalized difference water index NDWI (Normalized Difference Water Index) and a modified normalized difference water index MNDWI (Modified Normalized Difference Water Index) for each clear pixel of each image are calculated. The specific calculation mode of the index is as follows:
NDWI=(Green-NIR)/(Green+NIR)
MNDWI=(Green-SWIR)/(Green+SWIR)
wherein Green, NIR, SWIR is the earth surface reflectivity of the remote sensing image in the green band, near infrared band and short wave infrared band respectively. NDWI and MNDWI are both between-1 and 1. Further, a time sequence scatter diagram of a remote sensing index MNDWI and an NDWI is respectively constructed for each pixel of the research area and used for representing time sequence water index information of each pixel in the research area.
The technical difficulty of the remote sensing monitoring of tidal flat change is that 1, how to break through the limitation of insufficient quantity of low-tide clear images caused by the dynamic inundation factors of the tide in rainy days? 2. How does the remote sensing index time sequence change corresponding to different factors such as dynamic inundation of tide, artificial activity and the like clear? 3. How does the accuracy of extraction of the upper and lower edges of the tidal flat range be compromised? 4. How do the tidal flat change automated monitoring technical process without training samples be implemented? These are not fully considered by the existing remote sensing monitoring technology. Aiming at the technical difficulties, when the remote sensing image set is acquired, not only the clear remote sensing images are acquired, but also all the remote sensing images are fully utilized, and all the clear pixels which are not covered by cloud images are reserved for monitoring tidal flat change, so that the limitation of insufficient quantity of low-tide clear images caused by rainy days and dynamic tidal inundation is broken through. As shown in fig. 3, the tidal flat is a tidal zone between the high and low levels of coastal large tides, and periodic inundation of tides causes difficulty in remote sensing monitoring of the tidal flat range. According to the invention, the observation analysis shows that the change of the earth surface coverage type of the tidal flat can cause the severe fluctuation of the remote sensing time sequence index, and the tidal dynamic inundation corresponds to the small fluctuation of the remote sensing time sequence index, so that the method is used for removing the pseudo change of the remote sensing time sequence index caused by the dynamic influence of the tide by judging the cluster mean value change of the remote sensing time sequence index, thereby obtaining the time of the change of the earth surface coverage type of each pixel. Compared with other remote sensing monitoring methods which directly calculate the water coverage frequency without considering time sequence change detection, the method can effectively avoid the remote sensing monitoring error of the tidal flat range caused by dynamic tide flooding.
According to the invention, after the profile distribution characteristics of the tidal flat are analyzed through investigation and analysis of various remote sensing indexes, the MNCWI and the MNCWI have relatively excellent monitoring capability on the upper edge and the lower edge of the tidal flat range respectively. Through investigation and analysis, the MNDWI is also found to be stable in the different time-space scales at the water-land segmentation threshold value of the MNDWI at the upper edge of the tidal flat and the water-land segmentation threshold value of the NDWI at the lower edge of the tidal flat, so that the tidal flat range can be obtained through a threshold segmentation method without training samples. Therefore, compared with the prior art, the tidal flat change monitoring method for the coordinated MNCWI and NDWI remote sensing indexes has the advantages of no need of training samples, high automation degree and good beneficial effects in extracting the tidal flat range and the change thereof. According to a specific embodiment of the invention, 300 independent sample points are adopted to carry out accuracy verification in a change area, and compared with the accuracy, the overall accuracy of tidal flat change type monitoring by adopting MNDWI and NDWI double remote sensing time sequence indexes is 88%, the overall accuracy of change time monitoring is 86%, the accuracy of change type monitoring by adopting MNDWI and NDWI double remote sensing time sequence indexes is 43% and 23% higher than that of tidal flat change type monitoring by adopting only MNDWI remote sensing indexes and the accuracy of change time monitoring by adopting only the change type monitoring by adopting the MNDWI remote sensing indexes and the change time monitoring by adopting the change type monitoring by adopting the change time indexes are 36% and 14% higher than that of the change type monitoring by adopting the change time indexes.
And carrying out change time detection in cooperation with the double remote sensing time sequence indexes. The specific method comprises the following steps: firstly, the change time of the cluster mean of MNDWI and MNDWI is obtained for each pixel, and the judgment method of the cluster mean is referred to (Sen, A. And M.S. Srivastava, on tests for detecting change in mean. 1975: p.98-108.). As shown in FIG. 4, according to the cluster mean change condition, the remote sensing time sequence index time sequence can have multiple conditions such as no change point, single change point, multiple change points and the like. Detecting a change point of the MNDWI time sequence as a dividing point to divide the time sequence into a plurality of sub-time sequences with different lengths, detecting the change point of the NDWI time sequence, comparing the distance between the change point and the nearest MNDWI change point, and if the distance is more than 365 days, reserving the change point corresponding to the NDWI time sequence, otherwise, discarding the change point. The shortest length of all sub-time sequences is set to be not less than 365 days at the same time, and if a time sequence less than 365 days exists after division, the time sequence is integrated into adjacent time sequences nearby.
And calculating the water coverage frequency value of each sub-time sequence of each pixel, and developing the earth surface coverage classification according to the water coverage frequency value. According to the invention, through investigation and analysis, the MNDWI and the NDWI are respectively 0 in the water-land segmentation threshold value when extracted in the tidal flat range, and the threshold values are relatively stable in different time-space scales, so that the specific classification method comprises the following steps: if the frequency of NDWI value greater than 0 in the time sequence is greater than 95%, judging the time sequence as water; if the MNCWI value is more than 0 and the frequency is less than 5%, judging that the land is the land; the rest are classified as tidal beaches. The earth coverage type of each sub-timing of each pixel is acquired so far.
And removing the pseudo-change to complete the remote sensing monitoring of the tidal flat change. And sequentially comparing the earth surface coverage types of adjacent sub-time sequences of each pixel, and if the earth surface coverage types of the adjacent sub-time sequences are the same, removing the pseudo-change corresponding to the change time between the adjacent sub-time sequences. After all the pseudo changes are removed, the earth surface coverage change time of each pixel and the corresponding change type of each pixel can be obtained, so that the annual tidal flat range is obtained. The final data type is a multi-layer raster data stack, each layer corresponds to the beach range of each year, and the raster attribute values are land area, tidal beach and water area. As shown in fig. 5, the annual tidal flat, land area and water area change can be counted according to the monitoring result, and a decision basis is provided for coastal zone resource management and ecological restoration.
According to an embodiment of the present invention, there is provided a tidal flat change monitoring system based on dual remote sensing timing indexes, including:
the consistency check and correction module is used for carrying out linear adjustment and histogram matching on the surface reflectivity spectrum values of different sensors so that the correlation coefficient of the frequency distribution histograms of the different sensors is more than 0.95;
the double remote sensing index module is used for calculating MNDWI and NDWI remote sensing indexes of each pixel and constructing double remote sensing time sequence indexes of each pixel;
the variable point segmentation module is used for segmenting the time sequence according to the variable points of the double remote sensing index time sequence clustering mean value to obtain the sub-time sequence of each pixel;
the surface coverage classification module is used for calculating the water coverage frequency value of each sub-time sequence of each pixel and classifying the surface coverage;
and the pseudo-change removing module is used for removing the pseudo-change time according to the consistency of the surface coverage type of the adjacent sub-time sequences of each pixel.
The system operates to realize the monitoring method of the invention, and can realize high-precision monitoring of tidal flat change based on double remote sensing time sequence indexes.
In the experiment, guangdong, hong Kong and Australia (Guangdong province, hong Kong and Australia) are taken as examples, and a tidal flat change monitoring method based on double remote sensing time sequence indexes is applied.
The first step is to acquire all available Landsat series image surface reflectance data covering the research area of cantonese, during 1990 to 2021, based on the Google Earth Engine (GEE) platform, with the area of 1 km inside and outside the coastline of cantonese, as the research area. In this example, 13295 Landsat images were acquired.
The second step is to carry out consistency check and correction work of the surface reflectivity spectrum values of different sensors. And carrying out linear adjustment on the surface reflectivity spectral values of the Landsat8 OLI sensor in the red light wave band, the green light wave band, the blue light wave band, the near infrared wave band and the short wave infrared wave band. Further, the spectral values of the above-mentioned bands of Landsat8 OLI are checked and corrected using a histogram matching method so that the correlation coefficient of the two frequency distribution histograms is greater than 0.95.
And thirdly, calculating MNDWI and NDWI remote sensing indexes for each pixel of the research area, and constructing a double remote sensing time sequence index of each pixel.
And fourthly, carrying out change time detection in cooperation with the double remote sensing indexes. And respectively acquiring the change time of the cluster mean value of the MNDWI index time series of each pixel, and dividing the time series into a plurality of sub-time series with unequal lengths by taking the change time as a dividing point.
The fifth step is to develop sub-timing earth coverage type partitioning. Based on the sub-time sequences obtained after the variable point detection, calculating the frequency of each sub-time sequence MNDWI >0 and NDWI >0 of each pixel as a water coverage frequency value, and developing the earth surface coverage classification: if more than 95% of NDWI values in the time sequence are more than 0, judging the NDWI values as water; if the MNCWI value is more than 0 and the frequency is less than 5%, judging that the land is the land; the rest is tidal flat.
The sixth step is to reject the spurious changes. And comparing the earth surface coverage types of adjacent sub-time sequences of each pixel, and eliminating the corresponding pseudo-change time if the earth surface coverage types of the adjacent sub-time sequences are consistent. Finally, the change time and the change type of each pixel are obtained, and the remote sensing monitoring of the tidal flat change is completed.
Through the process, the application of the tidal flat change monitoring method based on the double remote sensing time sequence indexes in Guangdong, port and Australia in 1986-2021 is completed, and land, beach and water distribution products year by year are obtained. The data type is a multi-layer grid stack, each layer corresponds to the tidal flat range monitoring result of each year, and the grid attribute values are divided into three types of land areas, tidal flat areas and water areas, and the results are shown in fig. 5. The tidal flat change remote sensing monitoring product can be used as a thematic information product to be provided for related departments such as natural resource management, coastal zone ecological restoration and the like, provides decision basis for coastal zone ecological restoration engineering, and has great practical value.
In conclusion, the invention provides an advanced and feasible method for monitoring the tidal flat long time sequence change with high time frequency, fully utilizes the superiority of double remote sensing time sequence indexes, and well overcomes the defect that the existing tidal flat change monitoring method cannot timely and effectively master the dynamic change of the tidal flat year by year. The method provided by the invention can be used for constructing tidal flat change remote sensing monitoring products, so that the method is provided for related departments such as natural resource management and coastal zone ecological restoration, and the like, decision basis is provided for coastal zone ecological restoration engineering, and the method has great practical value.
Claims (6)
1. The tidal flat change monitoring method based on the double remote sensing time sequence indexes is characterized by comprising the following steps of:
(1) Acquiring a full-time sequence multispectral remote sensing dataset, screening clear pixels, and correcting spectral values of blue light wave bands, green light wave bands, red light wave bands, near infrared wave bands and short wave infrared wave bands of different sensors based on a linear function fitting and frequency distribution histogram matching method;
(2) Calculating a normalized water index NDWI and an improved normalized water index MNDWI of each pixel of each image, so as to construct a MNDWI and NDWI remote sensing index time sequence scatter diagram of each pixel in a research area;
(3) Carrying out change time detection in cooperation with the double remote sensing indexes, respectively acquiring the change time of the cluster mean value of the MNDWI and MNDDWI remote sensing indexes of each pixel, and dividing the time sequence into a plurality of sub-time sequences with unequal lengths by taking the change time as a dividing point; the method comprises the following steps: aiming at each pixel, respectively acquiring the change time of the cluster mean value of MNDWI and NDWI, wherein according to the change condition of the cluster mean value, the water index time sequence can have multiple conditions of no change point, single change point and multiple change points, and secondly, the detected change point is used as a dividing point to divide the time sequence into a plurality of sub-time sequences with unequal lengths, and if no change point exists, the time sequence is not divided into a single time sequence; the shortest length of all sub-time sequences should be not less than 365 days, if a time sequence less than 365 days exists after the segmentation, the time sequence is combined into adjacent time sequences nearby;
(4) Based on the sub-time sequences obtained after the variable point detection, calculating the frequency of each sub-time sequence MNDWI >0 and NDWI >0 of each pixel as a water coverage frequency value, and developing the earth surface coverage classification: if more than 95% of NDWI values in the time sequence are more than 0, judging the NDWI values as water; if more than 95% of MNCWI values in the time sequence are less than 0, judging that the time sequence is land; the rest is tidal beaches;
(5) And comparing the earth surface coverage types of adjacent sub-time sequences of each pixel, removing the pseudo-change time, finally obtaining the change time and the change type of each pixel, completing remote sensing monitoring on the change of the tidal flat, and obtaining the annual tidal flat range.
2. The tidal flat change monitoring method based on double remote sensing time sequence indexes of claim 1, wherein the step (1) is specifically: for the acquired multispectral image earth surface reflectivity data in the research period of the research area, a pixel mask containing cloud images is used for screening out clear pixels of time sequence earth surface reflectivity, firstly, a linear function is adopted for adjusting the spectrum values of all wave bands, and a frequency distribution histogram matching method is adopted for carrying out consistency check and correction work on the spectrum values of different sensors, so that the correlation of remote sensing spectrum frequency distribution histograms corresponding to all wave bands of different sensors is more than 0.95, and an image data set with good spectrum consistency of each wave band of a long time sequence is constructed.
3. The tidal flat change monitoring method based on double remote sensing time sequence indexes according to claim 1, wherein when the time sequence is divided, the change point of the MNDWI time sequence is detected as a dividing point to divide the time sequence into a plurality of sub-time sequences with different lengths, then the change point of the NDWI time sequence is detected, the distance between the change point and the nearest MNDWI change point is compared, if the distance is more than 365 days, the change point corresponding to the NDWI time sequence is reserved, and otherwise, the change point is abandoned.
4. The tidal flat change monitoring method based on double remote sensing time sequence indexes according to claim 1, wherein the step (4) is specifically: calculating the water coverage frequency value of each sub-time sequence of each pixel, and developing the earth surface coverage classification according to the water coverage frequency value; if the frequency of NDWI value greater than 0 in the time sequence is greater than 95%, judging the time sequence as water; if the MNCWI value is more than 0 and the frequency is less than 5%, judging that the land is the land; the rest are classified as tidal beaches; the earth coverage type of each sub-timing of each pixel is acquired so far.
5. The tidal flat change monitoring method based on double remote sensing time sequence indexes according to claim 1, wherein the step (5) is specifically: sequentially comparing the earth surface coverage types of adjacent sub-time sequences of each pixel, if the earth surface coverage types of the adjacent sub-time sequences are the same, the change time between the adjacent sub-time sequences is corresponding to pseudo change, all the pseudo change time is required to be removed, and the earth surface coverage change time of each pixel and the corresponding change type can be obtained after all the pseudo change time is removed, so that the annual tidal flat range is obtained; the final data type is a multi-layer raster data stack, each layer corresponds to the beach range of each year, and the raster attribute values are land area, tidal beach and water area.
6. Tidal flat change monitoring system based on two remote sensing time sequence indexes, characterized by comprising:
the consistency check and correction module is used for carrying out linear adjustment and histogram matching on the surface reflectivity spectrum values of different sensors so that the correlation coefficient of the frequency distribution histograms of the different sensors is more than 0.95;
the double remote sensing index module is used for calculating MNDWI and NDWI remote sensing indexes of each pixel and constructing double remote sensing time sequence indexes of each pixel;
the variable point segmentation module is used for segmenting the time sequence according to the variable points of the double remote sensing index time sequence clustering mean value to obtain the sub-time sequence of each pixel; the method comprises the following steps: aiming at each pixel, respectively acquiring the change time of the cluster mean value of MNDWI and NDWI, wherein according to the change condition of the cluster mean value, the water index time sequence can have multiple conditions of no change point, single change point and multiple change points, and secondly, the detected change point is used as a dividing point to divide the time sequence into a plurality of sub-time sequences with unequal lengths, and if no change point exists, the time sequence is not divided into a single time sequence; the shortest length of all sub-time sequences should be not less than 365 days, if a time sequence less than 365 days exists after the segmentation, the time sequence is combined into adjacent time sequences nearby;
the surface coverage classification module is used for calculating the water coverage frequency value of each sub-time sequence of each pixel and classifying the surface coverage; the method comprises the following steps: if the frequency of the NDWI value greater than 0 in the calculated time sequence is greater than 95%, judging the frequency as water; if the MNCWI value in the calculated time sequence is more than 0 and less than 5%, judging that the time sequence is land; the rest are classified as tidal beaches; so far, obtaining the earth surface coverage type of each sub-time sequence of each pixel;
and the pseudo-change removing module is used for removing the pseudo-change time according to the consistency of the surface coverage type of the adjacent sub-time sequences of each pixel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310628404.4A CN116341932B (en) | 2023-05-31 | 2023-05-31 | Tidal flat change monitoring method and system based on double remote sensing time sequence indexes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310628404.4A CN116341932B (en) | 2023-05-31 | 2023-05-31 | Tidal flat change monitoring method and system based on double remote sensing time sequence indexes |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116341932A CN116341932A (en) | 2023-06-27 |
| CN116341932B true CN116341932B (en) | 2023-08-22 |
Family
ID=86891562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310628404.4A Active CN116341932B (en) | 2023-05-31 | 2023-05-31 | Tidal flat change monitoring method and system based on double remote sensing time sequence indexes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116341932B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117253155B (en) * | 2023-11-17 | 2024-03-15 | 山东大学 | Human activity detection method and system based on deep learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105957079A (en) * | 2016-04-28 | 2016-09-21 | 淮阴师范学院 | Lake water area information extraction method based on Landsat OLI multispectral image |
| CN111402169A (en) * | 2020-03-23 | 2020-07-10 | 宁波大学 | Method for repairing remote sensing vegetation index time sequence under influence of coastal tide |
| CN111597930A (en) * | 2020-04-30 | 2020-08-28 | 河海大学 | Coastline extraction method based on remote sensing cloud platform |
| CN114581784A (en) * | 2022-05-07 | 2022-06-03 | 自然资源部第二海洋研究所 | Construction method of long-time-sequence yearly mangrove remote sensing monitoring product |
| CN114612793A (en) * | 2022-02-22 | 2022-06-10 | 中国自然资源航空物探遥感中心 | Multi-temporal remote sensing coastline and tidal flat detection method based on high-frequency observation of water sideline |
| CN115690086A (en) * | 2022-11-17 | 2023-02-03 | 电子科技大学长三角研究院(湖州) | Object-based high-resolution remote sensing image change detection method and system |
| CN115841618A (en) * | 2022-12-05 | 2023-03-24 | 南京邮电大学 | Remote sensing image coastline extraction method based on clustering and edge detection |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11676244B2 (en) * | 2018-10-19 | 2023-06-13 | Mineral Earth Sciences Llc | Crop yield prediction at field-level and pixel-level |
-
2023
- 2023-05-31 CN CN202310628404.4A patent/CN116341932B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105957079A (en) * | 2016-04-28 | 2016-09-21 | 淮阴师范学院 | Lake water area information extraction method based on Landsat OLI multispectral image |
| CN111402169A (en) * | 2020-03-23 | 2020-07-10 | 宁波大学 | Method for repairing remote sensing vegetation index time sequence under influence of coastal tide |
| CN111597930A (en) * | 2020-04-30 | 2020-08-28 | 河海大学 | Coastline extraction method based on remote sensing cloud platform |
| CN114612793A (en) * | 2022-02-22 | 2022-06-10 | 中国自然资源航空物探遥感中心 | Multi-temporal remote sensing coastline and tidal flat detection method based on high-frequency observation of water sideline |
| CN114581784A (en) * | 2022-05-07 | 2022-06-03 | 自然资源部第二海洋研究所 | Construction method of long-time-sequence yearly mangrove remote sensing monitoring product |
| CN115690086A (en) * | 2022-11-17 | 2023-02-03 | 电子科技大学长三角研究院(湖州) | Object-based high-resolution remote sensing image change detection method and system |
| CN115841618A (en) * | 2022-12-05 | 2023-03-24 | 南京邮电大学 | Remote sensing image coastline extraction method based on clustering and edge detection |
Non-Patent Citations (1)
| Title |
|---|
| Monitoring long-term annual urban expansion (1986–2017) in the largest archipelago of China;Wenting Cao，et al.;《Science of the Total Environment 》(第2021期);第2页第1栏第1段至12页第2栏第2段 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116341932A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | A 30-year (1984–2013) record of annual urban dynamics of Beijing City derived from Landsat data | |
| Yu et al. | A new approach for land cover classification and change analysis: Integrating backdating and an object-based method | |
| Dronova et al. | Object-based analysis and change detection of major wetland cover types and their classification uncertainty during the low water period at Poyang Lake, China | |
| Rogers et al. | Mapping of mangrove extent and zonation using high and low tide composites of Landsat data | |
| CN111062368B (en) | City update region monitoring method based on Landsat time sequence remote sensing image | |
| Pu et al. | A protocol for improving mapping and assessing of seagrass abundance along the West Central Coast of Florida using Landsat TM and EO-1 ALI/Hyperion images | |
| CN116341932B (en) | Tidal flat change monitoring method and system based on double remote sensing time sequence indexes | |
| CN114581784B (en) | Construction method of long-time-sequence yearly mangrove remote sensing monitoring product | |
| Zhang et al. | Automatically monitoring impervious surfaces using spectral generalization and time series Landsat imagery from 1985 to 2020 in the Yangtze River Delta | |
| CN112418506B (en) | Coastal zone wetland ecological safety pattern optimization method and device based on machine learning | |
| Kwon et al. | Hyperspectral retrievals of suspended sediment using cluster-based machine learning regression in shallow waters | |
| Elmahdy et al. | Monitoring and analysing the Emirate of Dubai’s land use/land cover changes: An integrated, low-cost remote sensing approach | |
| Singh et al. | Application of earth observation data for estimation of changes in land trajectories in Varanasi District, India | |
| Phuong et al. | Remote sensing for monitoring surface water quality in the Vietnamese Mekong delta: the application for estimating chemical oxygen demand in river reaches in Binh Dai, Ben Tre | |
| CN116385867A (en) | Ecological land block monitoring, identifying and analyzing method, system, medium, equipment and terminal | |
| Kara et al. | Impact of rapid urbanisation on land cover in Istanbul Province | |
| CN116543300A (en) | Cloud-aerosol hierarchical classification method based on semantic segmentation | |
| Avci et al. | A comparison of pixel-based and object-based classification methods, a case study: Istanbul, Turkey | |
| CN116310864A (en) | Automatic identification method, system, electronic equipment and medium for crop lodging | |
| Trivero et al. | Automatic oil slick detection from SAR images: Results and improvements in the framework of the PRIMI pilot project | |
| CN115761493A (en) | Water body extraction method based on combined water body index frequency | |
| CN112729562B (en) | Sea ice distribution detection method based on improved U-shaped convolutional neural network | |
| CN115661675A (en) | Multi-cloud area earthquake landslide remote sensing identification method, system, equipment and storage medium | |
| Zheng et al. | Classification of salt marsh vegetation in the Yangtze River Delta of China using the pixel-level time-series and XGBoost algorithm | |
| Parkan | Combined use of airborne laser scanning and hyperspectral imaging for forest inventories |
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 |