CN110569734A - Automatic identification method for newly-built reservoir based on high-time-frequency remote sensing image sequence - Google Patents
Automatic identification method for newly-built reservoir based on high-time-frequency remote sensing image sequence Download PDFInfo
- Publication number
- CN110569734A CN110569734A CN201910735329.5A CN201910735329A CN110569734A CN 110569734 A CN110569734 A CN 110569734A CN 201910735329 A CN201910735329 A CN 201910735329A CN 110569734 A CN110569734 A CN 110569734A
- Authority
- CN
- China
- Prior art keywords
- newly
- breakpoint
- time
- reservoir
- trend
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
the invention provides a newly-built reservoir automatic identification method based on a high time-frequency remote sensing image sequence. The computer automation method for locating reservoir building/water storage activities in time and space by utilizing a time sequence analysis algorithm to decompose and detect a normalized water body index (NDWI) time sequence inverted by remote sensing images in short revisiting periods and according to trend characteristics before and after breakpoints disclosed by the time sequence analysis result effectively realizes the spatial position extraction, water area spatial range delineation and reservoir building/water storage time identification of a newly-built reservoir/artificial lake. The method can adopt the freely obtained high revisiting period remote sensing images and the images with medium and high spatial resolution, expands the application range of the method, and provides important technical method support for establishing efficient large-scale newly-repaired reservoir catalogues and basin reservoir storage and water discharge information parameterization.
Description
Technical Field
the invention relates to the technical field of remote sensing, in particular to a newly built reservoir automatic identification method based on a high time-frequency image sequence.
Background
the reservoir is an artificial lake formed by building a barrage at a narrow opening of a mountain. After the reservoir is built, the functions of flood control, irrigation, fish culture, water supply, power and the like can be achieved, certain influences can be caused on natural environments including climate, water quality, biology and the like, even some geological disasters such as reservoir leakage, bank landslide, reservoir induced earthquake and the like (Xuehua and Liusu Mei, 2007) can be caused, and the lagging phenomenon still exists in the disclosure of the time-space information of the newly-built reservoir by related departments and units, so that the time-space information of the newly-built reservoir is often difficult to obtain in time, and the accuracy of hydrological assessment and the scientificity of basin regulation and control are greatly influenced. Therefore, the method for automatically monitoring the space-time information of the newly built reservoir is designed, and has important significance for scientific research and government decision.
The extraction of the newly-built reservoir mainly refers to the extraction of the space position, the space range and the construction time of the newly-built reservoir in a larger space range, and is mainly used for research in the fields of hydrology, environment and the like and decision-making of departments such as government water conservancy, environment and the like. The high-precision space-time information of the newly-built reservoir can assist relevant researchers to know hydrological changes of a research area more deeply and grasp hydrological characteristics of the research area more deeply, so that necessary support is provided for scientific research. The space-time information of the newly-built reservoir is efficiently and accurately monitored, and artificial intervention and regulation and control decision of the hydrology of the drainage basin can be assisted for all levels of water conservancy departments of the government, so that reasonable utilization and reasonable distribution of water resources are realized; for environmental protection departments at all levels of the government, the development of environmental supervision and environmental protection work can be assisted; other relevant government departments can also obtain relevant necessary auxiliary information, thereby providing scientific support for decision making. The prior art mainly focuses on the aspects of reservoir construction engineering technology, estimation of reservoir areas, water levels, reservoir capacity and the like after the reservoir is constructed, and for the condition of new reservoir construction, except for inquiring relevant data from relevant departments and units, no method for realizing large-scale identification of the new reservoir is available at present; the disclosure time of related departments or units on the data of the reservoir building activities is often relatively lagged, and no similar method can realize the real-time monitoring of the reservoir building activities with low cost, high precision and large range at present, so that the research in the hydrology and environment fields is difficult to obtain the reservoir building information, the related research is difficult to develop smoothly in time, and unnecessary obstacles are formed.
The remote sensing technology has the characteristics of large-area synchronous observation, strong timeliness, data comprehensiveness and comparability, multiple means for acquiring information, large information amount and high economic and social benefits, and can serve various aspects of production research. Remote sensing image data of a long time span has been accumulated at present, and related ground object extraction and analysis methods are also very perfect; the time series analysis method is used for extracting key information from mass data, and can extract key information of surface feature changes from remote sensing images accumulated for many years; the BFAST algorithm, as a time series analysis method without limitation to sensors and remote sensing ground physical indexes, can resolve trend information and seasonal information (i.e., trend components and seasonal components) from a large amount of remote sensing data accumulated over a plurality of years, and simultaneously detect possible mutation points (i.e., breakpoints) in the trend components and the seasonal components (verbessel et al, 2010), respectively. Currently, the BFAST algorithm has been widely applied to vegetation change monitoring (Rogier et al, 2012; Liu Bao post et al, 2016). Based on the general characteristic of a BFAST time sequence analysis algorithm, time sequence decomposition and breakpoint detection can be carried out on the remote sensing image water body index time sequence by considering the application of the BFAST algorithm, and reservoir building information is extracted based on structural components and breakpoint information. With the continuous enhancement of data acquisition capacity and computer analysis capacity, reservoir construction and storage regulation activities on regional scale and global scale have increasingly increased influence research, and related research results are closely related to global changes and ecosystems.
Reference to the literature
[1] Xu courts, Liu vegetarian plum, reservoir and its environmental impact [ J ] engineering mechanics, 2007(S2):33-44+74.
[2]S.K.McFEETERS.The use of the Normalized Difference Water Index(NDWI)in the delineation of open water features[J].International Journal of Remote Sensing,1996,17(7):1425-1432.
[3]Jan Verbesselt,Rob Hyndman,Glenn Newnham,Darius Culvenor.Detecting trend and seasonal changes in satellite image time series[J].Remote Sensing of Environment,2010,114(1):106-115.
[4]Jan Verbesselt,Rob Hyndman,AchimZeileis,Darius Culvenor.Phenological change detection while accounting for abrupt andgradual trends in satellite image time series[J],Remote Sensing of Environment,2010,114(12):2970-2980.
[5]Rogier Jong,Jan Verbesselt,Michael E.Schaepman,Sytze Bruin.Trend changes in global greening and browning:contribution of short-term trends tolonger-term change[J].Global Change Biology,2012,18(2).
[6] Liu Bao column, Fangxiuqin, He Qisheng and Rong Qiyuan vegetation change monitoring based on MODIS data and BFAST method [ J ]. remote sensing of national and local resources 2016,28(03):146-153.
Disclosure of Invention
The invention aims to provide a newly-built reservoir automatic identification method based on a high time-frequency remote sensing image sequence, which comprises the steps of space position extraction, space range delineation and construction time measurement.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
The method for automatically identifying the newly-built reservoir based on the high-time-frequency remote sensing image sequence comprises the following steps:
Step 1, collecting remote sensing image data covering a research area;
Step 2, inverting each pixel normalized water index (NDWI) from the remote sensing image obtained in the step 1, and constructing a time sequence of the pixel normalized water index (NDWI);
step 3, decomposing trend, season and noise components of the NDWI time sequence of each pixel by using a BFAST algorithm, and detecting possible breakpoints in the trend components;
Step 4, determining pixels in the inundated area of the newly-built reservoir according to the existence of the break points and trend information before and after the break points;
And 5, assigning values to pixels in the inundated area of the newly-built reservoir in the result grid diagram, wherein the pixel values are the time of breakpoints of the newly-built reservoir.
in the method of the present invention, the step 1 further includes downloading data which is high-time-resolution and medium-space-resolution remote sensing images, providing green bands and near-infrared bands for NDWI calculation, wherein the time span is as long as possible, such as MODIS MOD09a1 earth surface reflection data, the revisit period is 8 days, the space resolution is 500m, and data which has been accumulated for 18 years so far. Along with the increase of the spatial resolution of the newly-put-into-operation remote sensing sensor, the extraction precision of the pixels in the newly-built reservoir inundation area is increased. With the increase of the time span, the time range of the method for automatically identifying the newly built reservoir is widened.
In the step 2, the NDWI value is calculated by using an NDWI calculation formula based on a GREEN band and a near infrared band proposed by mcfeetts, that is, NDWI is (GREEN-NIR)/(GREEN + NIR). And reading the NDWI continuous observation value of each pixel by using the R language, calling a ts () function, setting start-stop time and observation interval, and finishing time sequence construction.
The step 2 also comprises the processing of a missing value and an abnormal value, and for the missing value and the abnormal value, the average value of two observations of 1 observation before the observation and 1 observation after the observation is taken to replace the original value; if a missing or outlier occurs in the first or last observation in the time series, the mean is filled with 4 observations after the observation or 4 observations before the observation.
in the step 3, in an R language environment, a BFAST algorithm is applied to perform time sequence decomposition and breakpoint detection, a confidence interval is set to be 95%, iteration times are set to be 3 times, the maximum time interval between breakpoints is set to be 10% of the time span of the whole time sequence, and a harmonic model is selected as a seasonal model.
In the step 4, comparing the average value of trend components of all the pixels with a threshold value, wherein the average value is smaller than the threshold value, and the average value is larger than the threshold value, the pixels are located in the inundated area of the newly-built reservoir, otherwise, the pixels are located outside the inundated area of the newly-built reservoir.
According to the quantity characteristics of trend components before and after the breakpoint, three types of pixels can be mainly classified: class I, land-water conversion (in newly built reservoir inundated areas), with breakpoints where both the original NDWI time series and trend components suddenly rise; class II, water type, may detect the breakpoint, but the values of the trend components before and after the breakpoint are higher relative to the land-water conversion type; in class III, the land type, breakpoints may also be detected, but the values of the trend components before and after the breakpoints are lower compared to the land-water conversion type. Therefore, the pixels can be classified according to the following criteria: firstly, eliminating partial II and III breakpoints according to the existence of the breakpoints; then, by comparing trend information before and after the breakpoint, namely, whether the average value of the trend component before the breakpoint is lower than a certain threshold value or not and whether the average value of the trend component after the breakpoint is higher than the threshold value or not are taken as judgment basis, if the two conditions of a and b are met, the land-water body conversion type is realized; if only the condition a is met, namely the average value of the trend component before the breakpoint and the average value of the trend component after the breakpoint are both smaller than the threshold, the trend component is in a land type; and if only the condition b is met, namely the average value of the trend component before the breakpoint and the average value of the trend component after the breakpoint are both greater than the threshold, determining that the trend component is in the water body type.
The threshold is obtained by an iterative solution method, and a threshold value with higher robustness can be obtained. The specific method is that the visual overlay analysis is carried out on the extraction result of the pixels of the reservoir inundation area under different threshold values and the Landsat TM/ETM/OLI true color synthetic image of which the imaging time is closest to the time range end point to which the method is applied, so as to evaluate the filtering effect of the different threshold values. The recommended scheme is that a small sample area is selected in a research area, the prior knowledge of newly-built reservoir information in the sample area is obtained by comprehensively utilizing various means and ways (such as information retrieval, maps, Google Earth images and the like), then threshold iteration solution is carried out in the sample area, and finally the method is popularized to the whole application area of the method. Through application tests (including the embodiments in the specification) of the method in different test areas, the recommended threshold value of the method is-0.20, the value is good in robustness, and the method can be directly used in most cases.
And 5, taking an original image as a base map, assigning the pixels in the inundated area of the newly-built reservoir by using the breakpoint time of the pixels, assigning a 0 value to other pixels by using a law to obtain a result grid, and using the grid as a reservoir space position and space range sketching and building time extraction result. Furthermore, the obtained grid graph of the extraction result of the newly-built reservoir inundation area can be superposed to other image/vector layers to assist relevant research.
According to the invention, by utilizing freely-available short revisit period MODIS satellite remote sensing data and high-resolution Landsat series satellite TM/ETM +/OLI data, the space position and the construction time of a newly-built reservoir can be accurately extracted, the blank of related methods in the field is filled, the space position and the space range of the newly-built reservoir are accurately obtained, the time information of reservoir construction activities is accurately grasped, the observation precision is improved, and important scientific and technological support is provided for the fields of hydrologic data updating, hydrologic analysis and research, river water and sand transportation research, river runoff and change research, regional climate change, government decision, environment monitoring and the like.
drawings
The drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. Embodiments of various aspects of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:
FIG. 1 is a sample area of the study in the examples.
FIG. 2 is a flow chart of the method of the present invention.
fig. 3 is three cases of classifying the image elements according to the trend component information.
And FIGS. 4a to f are the filtering results of the pixels of the non-newly built reservoir under different breakpoint-trend threshold values in the embodiment.
FIG. 5 shows the results of extraction of new reservoirs in the research sample area in the examples.
In the above-mentioned diagrams 1 to 7, the coordinates, marks or other representations expressed in english are all known in the art and are not described in detail in this embodiment.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are provided to illustrate the present invention, but are not intended to limit the scope of the present invention.
the embodiment of the invention takes a certain sub-basin in the Poyang lake basin as an example to further explain the method of the invention.
as shown in FIG. 1, the research area is located in the Poyang lake region, where the Jiangxiang branch is flowing through the grain, and the total area is about 3086 square kilometers. The image data used MODIS MOD09A1 surface reflectance data from 2000 to 2018 and Lansat-8OLI data at 11 months of 2018.
As shown in fig. 2, which is a flow chart of the present invention, the present embodiment includes the following steps:
Step 1, acquiring data of all available MODIS sensors MOD09A1 from 2000 to 2018 in a research area, and a Landsat-8OLI RGB true-color synthetic image imaged in 11 months of 2018.
and2, carrying out NDWI inversion pixel by pixel in an ENVI + IDL environment based on an NDWI algorithm proposed by McFeeters. For MODISMOD09A1 image, the NDWI calculation equation is
NDWI＝(Band4-Band2)/(Band4+Band2)
In the R language environment, the ts () function is called for the continuous observation of the NDWI of each pel to perform time series construction, and since MOD09a1 data is observed once every 8 days and 46 observations are made in one year, the frequency parameter is set to 46, and since the data accuracy of the first observation (corresponding to the 7 th observation in 2000) of MOD09a1 is low, the NDWI time series is constructed from the 8 th observation in 2000, and the start parameter is set to c (2000, 8).
Traversing each observation value of each pixel NDWI time sequence, and for all the missing values, taking the average value of two observations before the missing value and two observations after the missing value to fill the missing value; for abnormal values, the replacement method is the same. If the missing value/abnormal value appears in the first observation or the last observation, the mean value of 4 observations after/before the missing value/abnormal value is taken for replacing the missing value/abnormal value.
And 3, installing and loading a 'BFAST' program package in an R language environment, and calling a BFAST () function aiming at the time sequence of each pixel, wherein the h parameter is set to be 0.1, namely the minimum interval between breakpoints detected by a BFAST time sequence analysis algorithm is about 1.8 years, so that a large number of misjudgment points are prevented from occurring due to too narrow breakpoint intervals, and the correct breakpoints are prevented from being covered by the time intervals due to too loose breakpoint intervals. The seasonal model represented by the Season parameters was set to the harmonic model because related studies have demonstrated that it is more robust to breakpoint detection than the dummy model (verbessel et al, 2010). The parameter of the maximum iteration number (max. iter) is set to be 3, because subsequent observation shows that the iteration number of most of the pixels is less than or equal to 3, the time overhead can be effectively reduced on the premise of ensuring the precision. The parameter of the maximum number of breakpoints detected (breaks) is set to 1 because if it is assumed that the reservoir is not dismantled after being built, the ideal land-water body conversion pixel may only have a single lifting with the largest amplitude in the trend component. The confidence interval is set to 95%, i.e. the level parameter is set to 0.05, ensuring a sufficient confidence level.
Step 4, the breakpoints detected by the BFAST algorithm are not all generated due to the reservoir construction, and the fluctuation of the NDWI value caused by other reasons can also cause the BFAST algorithm to detect the breakpoints, and if the breakpoints are not filtered, the accuracy of the breakpoints cannot be accepted.
As shown in fig. 3, the trends and breakpoint characteristics of three different types of ground features selected in the three types of regions are significantly different. In this patent, the ground features can be divided into three types: 1) when the reservoir is built and starts to store water, the NDWI value is obviously increased, the trend is also obviously increased at almost the same position (namely a breakpoint), and the trend before and after the increase is stable. At a significant elevation, the BFAST algorithm will detect a breakpoint; 2) a water body region, in which a BFAST algorithm may detect a breakpoint, but before and after the breakpoint, the trend will remain at a relatively high value, which is an important feature that is distinguished from a land-to-water body region; 3) land areas where the BFAST algorithm may detect breakpoints, but before and after breakpoints the trend will remain relatively low, an important feature that distinguishes it from land-to-water areas.
Based on the characteristic of the trend component of the newly-built reservoir, namely that the storage of the reservoir causes the conversion of the ground features of the reservoir area from land to water, so that the NDWI value and the trend component are suddenly increased at the construction time (breakpoint), the trends before and after the breakpoint are respectively kept relatively stable, and in the pixel time sequence of other types of ground features (such as water and land), although the breakpoint can be detected, the characteristic is not existed, so that a method for comparing the average value of the trends before and after the breakpoint with the threshold can be adopted, and the 'the average value of the trends before the breakpoint is smaller than the threshold and the average value of the trends after the breakpoint is larger than the threshold' is adopted as the basis for judging whether the breakpoint of the pixel is caused by reservoir construction, thereby filtering the non-land-water conversion pixel.
For setting the threshold, the invention adopts an iterative solution method, traverses various threshold setting schemes from-0.05 to-0.30 in sequence by the step length of 0.01, judges the effects of different schemes by overlapping the threshold setting schemes in the Landsat-8OLI true color synthetic image for visual interpretation, and obtains the approximate trend of the influence of different threshold values on the pixel extraction effect in the inundation area of the newly-built reservoir. When the threshold value is small, the extraction of the pixels in the inundation range of the newly-built reservoir is complete, but the pixels of which many breakpoint causes are not reservoir building/water storage activities are not properly eliminated, so that the condition can be considered as an under-correction scheme; when the threshold value is large, the effect of delineating the inundation range of the newly-built reservoir is poor, and the pixels positioned in the inundation area of the reservoir are not extracted completely, so that the situation can be regarded as an overcorrection scheme, but the pixels of the newly-built reservoir are eliminated completely along with the increase of the threshold value. The reasonable situation is that a compromise scheme is adopted, the space position and the space range of the reservoir are guaranteed to obtain a precise sketching effect on the whole, and meanwhile, a certain number of misjudgment points which can be smoothly removed through river buffer area analysis and remote sensing images can be reserved. FIGS. 4 a-f show the trend of the influence of different threshold values on the result in step length of 0.05, and when the threshold values are-0.05, -0.10, -0.15, the pixels in the inundated area of the newly-built reservoir are not effectively extracted; when the threshold value is-0.25 or-0.30, the pixels in the inundated area of the newly-built reservoir are basically extracted completely, but have large error and leakage, which is particularly shown in the occurrence of a large number of error judgment points; compared with the threshold value of-0.20, the reservoir has better delineation effect of the spatial position and the spatial range, can completely cover the newly-built reservoir inundation range disclosed by the Landsat-8OLI true color synthetic image, and has no wrong division point.
And 5, taking the acquired MOD09A1 image of the research area obtained at any imaging time as a grid base map, setting the grid value of the pixel of the breakpoint as the time of the breakpoint, and uniformly setting the grid value of the pixel which does not contain the breakpoint as 0 to obtain a result grid map, wherein the grid map stores the spatial position information and the construction time information of the pixel of the newly-built reservoir, and can be superposed on a related map layer (such as Landsat series satellite images and Google Earth Pro images) to assist the development of related research.
Fig. 5 is the space position and space range sketching and construction time extraction results of the newly-built reservoir in the research sample area by adopting the method of the invention, and the base map is a Landsat-8OLI true color synthetic image. And superposing the extracted space position and space range of the reservoir to the Landsat-8 image in the figure, and comparing the extracted building time of the reservoir with the building time disclosed by the Google Earth historical image of the reservoir dam, thereby showing that the method provided by the invention has better extraction results on the space position, the space range and the building time of the newly-built reservoir.
By the method, the space range and the space position of the newly-built reservoir can be accurately extracted, the building time can be accurately measured, the space-time information of the newly-built reservoir can be correctly grasped, the accuracy of hydrological analysis and environmental monitoring can be improved, and an important scientific and technological support can be provided for hydrological and environmental change research.
although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the protection scope of the present invention should be determined by the appended claims.
Claims (10)
1. A newly-built reservoir automatic identification method based on a high time-frequency remote sensing image sequence is characterized by comprising the following steps:
Step 1, collecting remote sensing image data covering a research area;
Step 2, inverting each pixel normalized water index (NDWI) from the remote sensing image obtained in the step 1, and constructing a time sequence of the pixel normalized water index (NDWI);
step 3, decomposing trend, season and noise components of the NDWI time sequence of each pixel by using a BFAST algorithm, and detecting possible breakpoints in the trend components;
Step 4, determining pixels in the inundated area of the newly-built reservoir according to the existence of the break points and trend information before and after the break points;
and 5, assigning values to pixels in the inundated area of the newly-built reservoir in the result grid diagram, wherein the pixel values are the time of breakpoints of the newly-built reservoir.
2. The method of claim 1, wherein in step 1, the source of the remote sensing image data is MODISMOD09A1 surface reflection data.
3. The method of claim 1, wherein in step 1, the remotely sensed image data comprises all available sequential images over as long a time span as possible.
4. The method according to claim 1 or 2, wherein in step 2, the NDWI value is calculated by:
NDWI ═ (GREEN-NIR)/(GREEN + NIR); when applied to modimod 09a1 surface reflection data, the NDWI values are inverted by the formula NDWI ═ (Band4-Band2)/(Band4+ Band 2).
5. The method according to claim 1, wherein the step 2 further comprises processing missing values and abnormal values; filling missing or outliers with the mean of 1 observation before and 1 observation after the observation; if a missing or outlier occurs in the first or last observation in the time series, the mean is filled with 4 observations after the observation or 4 observations before the observation.
6. The method according to claim 1, wherein in step 3, a BFAST algorithm is applied to perform time series decomposition and breakpoint detection, the confidence interval is set to 95%, the number of iterations is set to 3, the maximum interval between breakpoints is set to 10% of the time span of the whole time series, and a harmonic model is selected as a seasonal model.
7. The method according to claim 1, wherein in the step 4, based on the trend information before and after the break point, the method for determining the picture elements in the inundated area of the newly-built reservoir is as follows:
Overlapping and analyzing each pixel and a Landsat/ETM +/OLI true color synthetic image of which the imaging time is closest to the end point of a time range to which the method is applied, observing the trend component and the breakpoint position of the true color synthetic image, and dividing the pixels into three types:
(a) The land-water body conversion pixel, namely the newly-built reservoir inundation area pixel, has the advantages that the original time sequence and trend components suddenly rise at the breakpoint, the trend before and after the breakpoint is kept stable, and the trend component before the breakpoint is obviously lower than the trend component after the breakpoint;
(b) The water body pixel is possible to detect a breakpoint and also possible not to detect the breakpoint, and even if the breakpoint is detected, the values of trend components before and after the breakpoint are in relatively high values;
(c) The land pixel is possible to detect a breakpoint and also may not detect the breakpoint, and even if the breakpoint is detected, the values of trend components before and after the breakpoint are all in relatively low values;
And (4) filtering the pixels (b) and (c) based on a threshold value aiming at the characteristic of the pixels in the inundated area of the newly-built reservoir.
8. the method of claim 7, wherein the threshold value is obtained by an iterative solution.
9. The method according to claim 1, wherein in the step 5, with one scene in the original remote sensing image data obtained in the step 1 as a base map, the pixel value of the newly-built reservoir inundation area is assigned with the time of the break point, and other pixels are assigned with 0 value in a unified manner, so as to obtain the newly-built reservoir inundation area grid.
10. The method as claimed in claim 1, further comprising superimposing the grid map of the extraction results of the newly-built reservoir inundation area obtained in step 5 on other image layers to assist in relevant research.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910735329.5A CN110569734A (en) | 2019-08-09 | 2019-08-09 | Automatic identification method for newly-built reservoir based on high-time-frequency remote sensing image sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910735329.5A CN110569734A (en) | 2019-08-09 | 2019-08-09 | Automatic identification method for newly-built reservoir based on high-time-frequency remote sensing image sequence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110569734A true CN110569734A (en) | 2019-12-13 |
Family
ID=68775026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910735329.5A Withdrawn CN110569734A (en) | 2019-08-09 | 2019-08-09 | Automatic identification method for newly-built reservoir based on high-time-frequency remote sensing image sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110569734A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117708548A (en) * | 2024-02-05 | 2024-03-15 | 北京开运联合信息技术集团股份有限公司 | Remote sensing satellite application efficiency evaluation method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101963664A (en) * | 2010-09-28 | 2011-02-02 | 中国科学院东北地理与农业生态研究所 | Microwave remote sensing pixel element decomposing method based on land and water living beings classifying information |
| CN106443701A (en) * | 2016-08-31 | 2017-02-22 | 华中农业大学 | Flood pre-disaster early warning method based on sequential water scope remote sensing image |
| CN106840409A (en) * | 2017-01-23 | 2017-06-13 | 北京师范大学 | A kind of forest fire point detection method based on MODIS |
| CN107944368A (en) * | 2017-11-16 | 2018-04-20 | 中国科学院遥感与数字地球研究所 | A kind of Clean water withdraw method based on sequential remote sensing images |
-
2019
- 2019-08-09 CN CN201910735329.5A patent/CN110569734A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101963664A (en) * | 2010-09-28 | 2011-02-02 | 中国科学院东北地理与农业生态研究所 | Microwave remote sensing pixel element decomposing method based on land and water living beings classifying information |
| CN106443701A (en) * | 2016-08-31 | 2017-02-22 | 华中农业大学 | Flood pre-disaster early warning method based on sequential water scope remote sensing image |
| CN106840409A (en) * | 2017-01-23 | 2017-06-13 | 北京师范大学 | A kind of forest fire point detection method based on MODIS |
| CN107944368A (en) * | 2017-11-16 | 2018-04-20 | 中国科学院遥感与数字地球研究所 | A kind of Clean water withdraw method based on sequential remote sensing images |
Non-Patent Citations (5)
| Title |
|---|
| CHUNQIAO SONG等: "Long-term surface water changes and driving cause in Xiong’an, China:from dense Landsat time series images and synthetic analysis", 《SCIENCE BULLETIN 》 * |
| OLOLADE O等: "MAPPING THE STATUS OF MINING WASTE USING LANDSAT DATA IN THE RUSTENBURG REGION", 《7TH AARSE CONFERENCE, ACCRA》 * |
| S.K.MCFEETERS: "The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features", 《INTERNATIONAL JOURNAL OF REMOTE SENSING》 * |
| WENBO LI等: "A Comparison of Land Surface Water Mapping Using the Normalized Difference Water Index from TM, ETM+ and ALI", 《REMOTE SENSING》 * |
| WENSONG ZHANG等: "Identifying Emerging Reservoirs along Regulated Rivers Using Multi-Source Remote Sensing Observations", 《REMOTE SENSING》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117708548A (en) * | 2024-02-05 | 2024-03-15 | 北京开运联合信息技术集团股份有限公司 | Remote sensing satellite application efficiency evaluation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11591762B2 (en) | Method for tracking, monitoring and evaluating ecological impact of channel project based on long-term time series satellite remote sensing data | |
| Deng et al. | Continuous subpixel monitoring of urban impervious surface using Landsat time series | |
| Cao et al. | Mapping changes in coastlines and tidal flats in developing islands using the full time series of Landsat images | |
| Wu | GIS and remote sensing applications in wetland mapping and monitoring | |
| Jia et al. | Phenological response of vegetation to upstream river flow in the Heihe Rive basin by time series analysis of MODIS data | |
| CN111062368B (en) | City update region monitoring method based on Landsat time sequence remote sensing image | |
| Wang et al. | A Quantitative Study of Gully Erosion Based on Object‐Oriented Analysis Techniques: A Case Study in Beiyanzikou Catchment of Qixia, Shandong, China | |
| Noto et al. | Low-cost stage-camera system for continuous water-level monitoring in ephemeral streams | |
| Jing et al. | A novel approach for quantifying high-frequency urban land cover changes at the block level with scarce clear-sky Landsat observations | |
| Liu et al. | Seasonal pattern of tidal-flat topography along the Jiangsu middle coast, China, using HJ-1 optical images | |
| Zhou et al. | Improving time series anomaly detection based on exponentially weighted moving average (EWMA) of season-trend model residuals | |
| Zhou et al. | Retrieving dynamics of the surface water extent in the upper reach of Yellow River | |
| Yang et al. | Bank gully extraction from DEMs utilizing the geomorphologic features of a loess hilly area in China | |
| CN110569733B (en) | Lake long time sequence continuous water area change reconstruction method based on remote sensing big data platform | |
| Gu et al. | Mapping inter-and intra-annual dynamics in water surface area of the Tonle Sap Lake with Landsat time-series and water level data | |
| Li et al. | Combined gully profiles for expressing surface morphology and evolution of gully landforms | |
| El Kafrawy | Shoreline Change Rate Detеction and Futurе Prediction Using Rеmote Sensing and GIS Tеchniques: A Case Study... | |
| Kolli et al. | Assessment of change in the extent of mangrove ecosystems using different spectral indices in Google Earth Engine based on random forest model | |
| Wang et al. | Human activities dominate a staged degradation pattern of coastal tidal wetlands in Jiangsu province, China | |
| CN110569734A (en) | Automatic identification method for newly-built reservoir based on high-time-frequency remote sensing image sequence | |
| Afifi et al. | Urban sprawl of greater Cairo and its impact on the agricultural land using remote sensing and digital soil map | |
| Al Kuwari et al. | Impact of North Gas Field development on landuse/landcover changes at Al Khore, North Qatar, using remote sensing and GIS | |
| Singh et al. | Estimation of revised capacity in Gobind Sagar reservoir using Google earth engine and GIS | |
| Milani et al. | Characterizing flood impact on Swiss floodplains using interannual time series of satellite imagery | |
| Cucchi et al. | Intrinsic vulnerability assessment of Sette Comuni plateau aquifer (Veneto region, Italy) |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191213 |
|
| WW01 | Invention patent application withdrawn after publication |