CN116129145B - A method and system for extracting sandy coastlines from high-resolution remote sensing images - Google Patents

Abstract

This disclosure provides a method and system for extracting sandy coastlines from high-resolution remote sensing images, which acquires multiple remote sensing images at different times, converts the remote sensing images into NDWI images and composes the NDWI images into NDWI image sequences, and performs NDWI image sequence Through the processing, the coastline extraction points are obtained, and the coastline is corrected according to the coastline extraction points, which is beneficial to improve the accuracy of the coastline.

Description

A method and system for extracting sandy coastlines from high-resolution remote sensing images

technical field

The disclosure belongs to the field of data processing, and in particular relates to a method and system for extracting sandy coastlines from high-resolution remote sensing images.

Background technique

With the continuous development and application of remote sensing technology, high-resolution remote sensing images have been widely used in coastline extraction. Traditional coastline extraction methods mainly use threshold segmentation or edge detection methods, but due to the complex terrain of coastlines and the influence of factors such as illumination, traditional methods are prone to large errors. Therefore, many new methods for coastline extraction have emerged in recent years. For example, some scholars have proposed to use convolutional neural network (CNN) to classify remote sensing images to improve accuracy; some scholars have proposed a method based on deep learning to use fully convolutional network (FCN) to classify remote sensing images. Split, and extract the coastline accordingly. However, these methods also have some limitations and deficiencies. For example, a coastline remote sensing extraction method described in the patent document CN113221813B, due to the complexity and variability of the coastline topography, as well as factors such as noise and ambiguity of the remote sensing image itself, these methods are still in practice. There are certain errors. In addition, some algorithms require a large amount of manually labeled data, making it difficult to promote and popularize them in practical applications. At the same time, these algorithms require high-performance computing equipment for calculation, which increases operating costs and time costs.

Contents of the invention

The purpose of the present invention is to propose a method and system for extracting sandy coastlines from high-resolution remote sensing images, so as to solve one or more technical problems in the prior art, and at least provide a beneficial option or create conditions.

This disclosure provides a method and system for extracting sandy coastlines from high-resolution remote sensing images, which acquires multiple remote sensing images at different times, converts the remote sensing images into NDWI images and composes the NDWI images into NDWI image sequences, and performs NDWI image sequence Through the processing, the coastline extraction points are obtained, and the coastline is corrected according to the coastline extraction points, which is beneficial to improve the accuracy of the coastline.

In order to achieve the above purpose, according to one aspect of the present disclosure, a method for extracting a sandy coastline from a high-resolution remote sensing image is provided, and the method includes the following steps:

S100, acquiring multiple remote sensing images at different times;

S200, converting the remote sensing image into an NDWI image, and composing the NDWI image into an NDWI image sequence;

S300, performing penetration processing on the NDWI image sequence to obtain coastline extraction points;

S400, obtain the corrected coastline according to the coastline extraction points.

Further, in S100, the time intervals between the times in the multiple different times are the same, and the arrangement of the times in the multiple different times is in chronological order, where each remote sensing image corresponds to each Each moment has one and only one corresponding remote sensing image, and the remote sensing image is a high-resolution remote sensing image of the coastline.

Further, in S200, the remote sensing images at each moment are converted into NDWI images by calculating the normalized difference water index (Normalized Difference Water Index), and the obtained NDWI images are sequentially composed into an NDWI image sequence, so The pixel values on each of the above NDWI images have been grayscaled and normalized and can be mapped to values belonging to [0,1]. Preferably, the closer the pixel value is to 0, the closer to black and the closer the pixel value is to 1 means the closer to white, the closer to black means the less water content in the corresponding position on the NDWI image, and the closer to white means the more water content in the corresponding position on the NDWI image.

Further, in S300, the method of performing penetration processing on the NDWI image sequence to obtain coastline extraction points is as follows:

Note that the NDWI image sequence is a sequence NDseq, the number of NDWI images as its elements in the NDWI image sequence is denoted as t, and the sequence number of the element in the NDWI image sequence is q, and q belongs to the interval from 1 to t, and the number of The sequence number of each moment in different moments is consistent with the sequence number q, and the sequence number is the qth element in the described NDWI image sequence is ND(q);

Each NDWI image in the NDWI image sequence is an image matrix with a size of n rows and m columns, wherein in each NDWI image, i is the serial number of the row and j is the serial number of the column, i belongs to the interval from 1 to n, and j Belonging to the interval from 1 to m, the value at the position of row i and column j in ND(q) is expressed as ND(q)[i,j];

The next of the last element in the NDWI image sequence is pointed to the first element, that is, the next of the last element ND(t) is pointed to the first element ND(1). This operation is because the change of the tide is Periodically back and forth, the tidal advance and retreat is a great interference factor for the blurred image of the coastline, so the end-to-end data sequence can better reflect the periodic data distribution of the coastline, and facilitate the subsequent extraction of the coastline;

The function TimeGAP(,) indicates the time interval between the moment of the serial number in the bracket and the time of the previous serial number, specifically: TimeGAP(1,q) indicates the time interval between the time of serial number q and the time of serial number 1, TimeGAP(1 ,t) represents the time interval between the time of serial number t and the time of serial number 1, and TimeGAP(q-1,q) represents the time interval of the time of serial number q from the time of serial number q to the previous time q-1 of serial number q, where when When the q of TimeGAP(q-1,q) represents the time of serial number 1, the time of the previous time q-1 of the serial number q is the time of serial number t, that is, the time of the time of serial number q minus the time of serial number 1 interval, the time interval is only an absolute value without positive or negative points, for example, the time of serial number q is 23:00 of a day, and the time of serial number 1 is 3:00 of the same day, then the time of serial number q is subtracted from the serial number The time interval of time 1 is 20 hours. If the time of serial number t is 24:00 of the same day, when the q of TimeGAP(q-1,q) represents the time of serial number 1, there will be TimeGAP(q-1,q) Equal to TimeGAP(t,1) is 21 hours, at this time TimeGAP(1,t) is also equal to 21 hours;

Calculate the time series penetration of each NDWI image, specifically:

Denote the timing penetration of ND(q) as perpet(q), and obtain the last element ND(q-1) of the ND(q) in the NDWI image sequence and its pixel values ND(q-1)[i ,j], after dimensionless processing, the calculation formula of perpet(q) can be:

Among them, exp is to calculate the exponential function based on the natural constant e, and exp(| ND(q)[i,j]-ND(q-1)[i,j]| ) is the NDWI of serial number q and its previous serial number The difference between the numerical values of the same row and column position in the image is expressed as an exponential representation after taking the absolute value, and the denominator is calculated by traversing i and j in each row and column position, and calculating the number of the same row and column position in the NDWI image of the serial number q and its previous serial number The difference between the values is taken as an exponential representation after taking the absolute value and accumulated and summed, so that the distribution of the numerical characteristics of the normalized water index corresponding to each position can be better extracted as the time and place change. The denominator part has traversal The accumulated symbol is enclosed in square brackets in the denominator part to indicate that the traversal loops of the sequence numbers j and i are firstly summed up, and then the value of the numerator is divided by the denominator part, TimeGAP(q-1,q)/TimeGAP (1,t) represents the ratio of the time interval between the serial number q and the time of the previous serial number to the total time change interval (TimeGAP(1,t)), which can extract the fluctuation degree of time change, combined with the normalized water index as Time-series penetration is calculated based on the distribution of numerical features due to changes in time and location and the degree of fluctuation in time changes, and the coastline contour features are penetrated through the data features of the image sequence at each moment, which is conducive to accurately capturing the tides caused by time changes. The contour features of the coastline are blurred, which improves the matching accuracy of sandy coastline extraction. The larger the time-series penetration value, the higher the response of the NDWI image to the contour features of the coastline, which is more conducive to improving Extraction accuracy;

The arithmetic mean, median or mode of the time-series penetration of each NDWI image in the NDWI image sequence can be calculated as the time-series penetration threshold of the NDWI image sequence (this can be beneficial to obtain the average level of the normal distribution in it), and the In the NDWI image sequence, the NDWI images whose time-series penetration reaches (the value is greater than or equal to) the threshold of the time-series penetration are screened out as super-threshold images, and the screened super-threshold images still retain their corresponding values in the NDWI image sequence. The time series penetration of ;

For each super-threshold image, multiply the pixel value of each pixel by its corresponding time-series penetration, so as to convert each super-threshold image into a time-series penetration image, and then calculate the time-series penetration images together The matrix obtained after multiplying the Hadamard product is the cumulative image matrix, (from the calculation of the time series penetration to the calculation of each time series penetration image together with the Hadamard product multiplication to obtain the cumulative image matrix, which has been widely used all the time. For the multiplication between arrays, here the Hadamard product is used to perform point-to-point numerical multiplication between the image matrices, and the numerical linearity of the multiplication of multiple variables is also continuously derivable. This numerical linear correlation is very useful for coastline extraction. helpful);

On the cumulative image matrix, the value of each row and column position is converted into a positive or negative value through a binarization algorithm, and then the point of the row and column position converted into a positive value is selected as the coastline extraction point, wherein the The positive or negative value can be the value of 0 or 1, logic and logical not, Boolean value (or the value of True or False), etc., which can represent the value or symbol of the logically opposite relationship, specifically: for each The value of the row and column position is converted into a value of 0 or 1 through a binarization algorithm (Binarization) or a decision tree classification algorithm (for example, a binary classification network based on Keras, a cart_tree decision tree based on sklearn, etc.), and then selected The points in the rows and columns of the values converted to 1 are used as coastline extraction points. Compared with the method of directly using the binarization algorithm or the decision tree classification algorithm to convert and then directly extract the coastline with the original NDWI image, the method of the present invention uses the value of the normalized water index that changes with time and place. Based on the feature distribution law and the fluctuation degree of time changes, the periodic data distribution of the coastline is measured, which is beneficial to avoid the influence of time and further improve the accuracy of the coastline.

Further, in S400, the method of obtaining the corrected coastline according to the coastline extraction points is: using the curve obtained by curve fitting or interpolation on the coastline extraction points as the corrected coastline.

The present disclosure also provides a system for extracting sandy coastlines from high-resolution remote sensing images. The system for extracting sandy coastlines from high-resolution remote sensing images includes: a processor, a memory, and a memory that is stored in the memory and can be stored in the memory. A computer program running on the processor, when the processor executes the computer program, the steps in the sandy coastline method of the high-resolution remote sensing image are realized, and the sandy coastline of the high-resolution remote sensing image is The quality coastline extraction system can run on computing devices such as desktop computers, notebook computers, palmtop computers, and cloud data centers, and the operable systems can include, but are not limited to, processors, memories, and server clusters. The processors execute Said computer program runs in units of the following systems:

An image acquisition unit, configured to acquire a plurality of remote sensing images at different times;

A serialization unit is used to convert the remote sensing image into an NDWI image, and form the NDWI image into an NDWI image sequence;

The extraction unit is used to perform penetration processing on the NDWI image sequence to obtain coastline extraction points;

The correction output unit is used to obtain the correction coastline according to the coastline extraction points.

The beneficial effects of the present disclosure are: the sandy coastline extraction method and system of high-resolution remote sensing images provided by the present disclosure, by converting multiple remote sensing images at different times into NDWI image sequences, and performing through-through processing on the NDWI image sequences, to obtain Coastline extraction points, and further get corrected coastlines. Compared with the existing methods, the method provided in the present disclosure also has the advantages of less calculation, less manual intervention, and high accuracy, and is expected to be more widely promoted and applied in practical applications.

Description of drawings

The above and other features of the present disclosure will be more apparent through a detailed description of the embodiments shown in the drawings. The same reference numerals in the drawings of the present disclosure represent the same or similar elements. Obviously, the appended The drawings are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative work. In the drawings:

Figure 1 shows a flow chart of a method for extracting sandy coastlines from high-resolution remote sensing images;

Figure 2 shows a system structure diagram of a sandy coastline extraction system for high-resolution remote sensing images.

Detailed ways

The concept, specific structure and technical effects of the present disclosure will be clearly and completely described below in conjunction with the embodiments and drawings, so as to fully understand the purpose, scheme and effect of the present disclosure. It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

In the description of the present invention, several means one or more, and multiple means more than two. Greater than, less than, exceeding, etc. are understood as not including the original number, and above, below, within, etc. are understood as including the original number. If the description of the first and second is only for the purpose of distinguishing the technical features, it cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features relation.

As shown in Figure 1, it is a flow chart of a sandy coastline extraction method of a high-resolution remote sensing image according to the present invention, and the sandy quality of a high-resolution remote sensing image according to an embodiment of the present invention will be described below in conjunction with Figure 1 Coastline extraction method and system.

The present invention includes a method for extracting a sandy coastline from a high-resolution remote sensing image, and the method specifically includes the following steps:

S100, acquiring multiple remote sensing images at different times;

S200, converting the remote sensing image into an NDWI image, and composing the NDWI image into an NDWI image sequence;

S300, performing penetration processing on the NDWI image sequence to obtain coastline extraction points;

S400, obtain the corrected coastline according to the coastline extraction points.

Further, in S100, the time intervals between the times in the multiple different times are the same, and the arrangement of the times in the multiple different times is in chronological order, where each remote sensing image corresponds to each Each moment has one and only one corresponding remote sensing image, and the remote sensing image is a high-resolution remote sensing image of the coastline.

Further, in S200, the remote sensing images at each moment are converted into individual NDWI images by calculating the normalized water index, and the obtained NDWI images are sequentially composed into an NDWI image sequence, and each NDWI image The pixel value above has been grayscaled and normalized to a value belonging to 0,1. Among them, the closer the pixel value is to 0, the closer it is to black, and the closer the pixel value is to 1, the closer it is to white, and the closer it is to black, it means NDWI The less water content in the corresponding position on the image, the closer to white, the more water content in the corresponding position on the NDWI image. (Preferably, the time interval between each moment can be 1 hour.)

Further, in S300, the method of performing penetration processing on the NDWI image sequence to obtain coastline extraction points is as follows:

Note that the NDWI image sequence is a sequence NDseq, the number of NDWI images as its elements in the NDWI image sequence is denoted as t, and the sequence number of the element in the NDWI image sequence is q, and q belongs to the interval from 1 to t, preferably , the interval from 1 to t can be the interval from 1 hour to 24 or 25 hours, preferably forming a closed and recyclable time period, the sequence number of each moment in the multiple different moments is consistent with the sequence number q, and the NDWI image The qth element in the sequence is ND(q);

Each NDWI image in the NDWI image sequence is an image matrix with a size of n rows and m columns, wherein in each NDWI image, i is the serial number of the row and j is the serial number of the column, i belongs to the interval from 1 to n, and j Belonging to the interval from 1 to m, the value at the position of row i and column j in ND(q) is expressed as ND(q)[i,j];

The next of the last element in the NDWI image sequence is pointed to the first element, that is, the next of the last element ND(t) is pointed to the first element ND(1). This operation is because the change of the tide is Periodically back and forth, the tidal advance and retreat is a great interference factor for the blurred image of the coastline, so the end-to-end data sequence can better reflect the periodic data distribution of the coastline, and facilitate the subsequent extraction of the coastline;

The function TimeGAP(,) indicates the time interval between the moment of the serial number in the brackets and the time of the previous serial number, specifically: TimeGAP(1,q) indicates the time interval between the time of the serial number q and the time of the serial number 1, that is, the serial number q The time interval between the time of serial number 1 and the time of serial number 1 is subtracted; TimeGAP(1,t) indicates the time interval between the time of serial number t and the time of serial number 1; TimeGAP(q-1,q) indicates the distance between the time of serial number q and the time of serial number q The time interval elapsed at the moment of the last moment q-1; wherein, when the q of TimeGAP(q-1,q) represents the moment of sequence number 1, the moment of the previous moment q minus 1 of the sequence number q is the sequence number At the time of t, it is equivalent to saying that in the serialized arrangement of serial number q, since serial number t and serial number 1 are connected end to end, when q is equal to 1, q minus 1 is equal to t; the time interval only takes the absolute value without positive Negative points; for example, the time of serial number q is 23:00 of a day, and the time of serial number 1 is 3:00 of the same day, then the time interval of the time of serial number q minus the time of serial number 1 is 20 hours, if the serial number t The moment of time is 24:00 of the same day, when the q of TimeGAP(q-1,q) represents the moment of serial number 1, then TimeGAP(q-1,q) is equal to TimeGAP(t,1) which is 21 hours, here TimeGAP(1,t) is also equal to 21 hours;

Calculate the time series penetration of each NDWI image, specifically:

Denote the timing penetration of ND(q) as perpet(q), and obtain the last element ND(q-1) of the ND(q) in the NDWI image sequence and its pixel values ND(q-1)[i ,j], after dimensionless processing, the calculation formula of perpet(q) is:

Among them, exp() is to calculate the exponential function based on the natural constant e, and exp(| ND(q)[i,j]-ND(q-1)[i,j] | ) is the serial number q and its previous serial number The difference between the numerical values of the same row and column position in the NDWI image is expressed as an exponential representation after taking the absolute value. The denominator is calculated by traversing i and j in each row and column position, and calculating the same row and column in the NDWI image with the serial number q and its previous serial number The difference between the numerical values of the positions is taken as the absolute value of the exponential representation and accumulated and summed, so that the distribution of the numerical characteristics of the normalized water index corresponding to each position changes with time and place can be better extracted, TimeGAP( q-1,q)/TimeGAP(1,t) indicates the ratio of the time interval between the serial number q and its previous serial number to the total time change interval (TimeGAP(1,t)), which can extract the fluctuation degree of time change, Combining the distribution of numerical features of the normalized water index with the change of time and place and the degree of fluctuation of time changes, the time series penetration is calculated, and the data features of the coastline contour features on the image sequence are broken through at each time, which is beneficial to Accurately capture the outline features of coastlines that are blurred due to tidal advance and retreat due to time changes, and improve the matching accuracy of sandy coastline extraction. The larger the time-series penetration value, the more responsive the NDWI image is to the outline features of the coastline. The higher the value, the more beneficial it is to improve the accuracy of extraction;

Calculate the arithmetic mean of the time-series penetration of each NDWI image in the NDWI image sequence as the time-series penetration threshold of the NDWI image sequence, and make the corresponding time-series penetration in the NDWI image sequence reach (the value is greater than or equal to) the time-series penetration The thresholded NDWI image is screened out as a super-threshold image, and the screened super-threshold image still retains its corresponding time series penetration in the NDWI image sequence;

For each super-threshold image, multiply the pixel value of each pixel by its corresponding time-series penetration, so as to convert each super-threshold image into a time-series penetration image, and then calculate the time-series penetration images together The matrix obtained after multiplying the Hadamard product is the cumulative image matrix, (from the calculation of the time series penetration to the calculation of each time series penetration image together with the Hadamard product multiplication to obtain the cumulative image matrix, which has been widely used all the time. For the multiplication between arrays, here the Hadamard product is used to perform point-to-point numerical multiplication between the image matrices, and the numerical linearity of the multiplication of multiple variables is continuously derivable, which is interpretable in mathematical analysis. This is different from the non-linear, non-continuous, derivable and inexplicable mathematical models based on machine learning, deep neural network, etc. This kind of interpretable numerical linear correlation is more guaranteed for coastline extraction);

On the cumulative image matrix, the value of each row and column position is converted into a positive or negative value through a binarization algorithm, and then the point of the row and column position converted into a positive value is selected as the coastline extraction point, wherein the The positive or negative value can be the value of 0 or 1, logic is and logical not, Boolean value, etc. can represent the value or symbol of the logically opposite relationship, etc., specifically: the value of each row and column position through binarization Algorithm (Binarization) or decision tree classification algorithm (for example, Keras-based binary classification network, sklearn-based cart_tree decision tree) is mapped to a value of 0 or 1, and then the row and column position of the value converted to 1 is selected Points are used as coastline extraction points. Compared with the method of directly using the binarization algorithm or the decision tree classification algorithm to convert and then directly extract the coastline with the original NDWI image, the method of the present invention uses the value of the normalized water index that changes with time and place. Based on the feature distribution law and the fluctuation degree of time changes, the periodic data distribution of the coastline is measured, which is beneficial to avoid the influence of time and further improve the accuracy of the coastline.

Further, in S400, the method of obtaining the corrected coastline according to the coastline extraction points is: using the curve obtained by curve fitting (Least Squares) or interpolation (Spline Interpolation) on the coastline extraction points as the corrected coastline.

The present invention also includes a sandy coastline extraction system for high-resolution remote sensing images, the sandy coastline extraction system for high-resolution remote sensing images runs on desktop computers, notebook computers, palmtop computers or cloud data In any computing device in the center, the computing device includes: a processor, a memory, and a computer program stored in the memory and running on the processor, and the processor implements the computer program when executing the computer program. In the steps of a method for extracting sandy coastlines from high-resolution remote sensing images, an operable system may include, but not limited to, a processor, a memory, a server cluster, and the like.

An embodiment of the present disclosure provides a system for extracting a sandy coastline from a high-resolution remote sensing image, as shown in FIG. 2 , the system for extracting a sandy coastline from a high-resolution remote sensing image in this embodiment includes: a processor, a memory and a computer program stored in the memory and operable on the processor, when the processor executes the computer program, the steps in the above embodiment of a sandy coastline extraction method for high-resolution remote sensing images are realized , the processor executes the computer program running in the following system units:

An image acquisition unit, configured to acquire a plurality of remote sensing images at different times;

A serialization unit is used to convert the remote sensing image into an NDWI image, and form the NDWI image into an NDWI image sequence;

The extraction unit is used to perform penetration processing on the NDWI image sequence to obtain coastline extraction points;

The correction output unit is used to obtain the correction coastline according to the coastline extraction points.

Among them, preferably, all undefined variables in the present invention, if not clearly defined, can be manually set thresholds.

The sandy coastline extraction system for high-resolution remote sensing images can be run on computing devices such as desktop computers, notebook computers, palmtop computers, and cloud data centers. The sandy coastline extraction system for high-resolution remote sensing images includes, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above example is only an example of a method and system for extracting a sandy coastline from a high-resolution remote sensing image, and does not constitute an introduction to a method and system for extracting a sandy coastline from a high-resolution remote sensing image. limited, may include more or less components, or combine certain components, or different components, for example, the sandy coastline extraction system for high-resolution remote sensing images may also include input and output devices, network interfaces input devices, buses, etc.

The so-called processor can be a central processing unit (Central Processing Unit, CPU), and can also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), on-site Programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete component gate circuits or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., and the processor is the control center of the sandy coastline extraction system of the high-resolution remote sensing image, using various interfaces Each sub-region of a high-resolution remote sensing image sandy coastline extraction system is connected with lines.

The memory can be used to store the computer programs and/or modules, and the processor realizes the one by running or executing the computer programs and/or modules stored in the memory and calling the data stored in the memory. A method for extracting sandy coastline from high-resolution remote sensing images and various functions of the system. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playback function, an image playback function, etc.) and the like; the storage data area may store Data created based on the use of the mobile phone (such as audio data, phonebook, etc.), etc. In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disk, internal memory, plug-in hard disk, smart memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card , a flash memory card (Flash Card), at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

This disclosure provides a method and system for extracting sandy coastlines from high-resolution remote sensing images, which acquires multiple remote sensing images at different times, converts the remote sensing images into NDWI images and composes the NDWI images into NDWI image sequences, and performs NDWI image sequence Through the processing, the coastline extraction points are obtained, and the coastline is corrected according to the coastline extraction points, which is beneficial to improve the accuracy of the coastline.

While the description of the present disclosure has been presented with considerable detail and in particular has described a few described embodiments, it is not intended to be limited to any such details or embodiments or to any particular embodiment, effectively encompassing the intended scope of the present disclosure. Furthermore, the disclosure has been described above in terms of embodiments foreseeable by the inventors for the purpose of providing a useful description, and insubstantial modifications of the disclosure which are not presently foreseeable may still represent equivalent modifications of the disclosure.

Claims (4)

1. The method for extracting the sandy coastline of the high-resolution remote sensing image is characterized by comprising the following steps of:

s100, acquiring a plurality of remote sensing images at different moments;

s200, converting the remote sensing image into an NDWI image, and forming the NDWI image into an NDWI image sequence;

s300, carrying out penetration processing on the NDWI image sequence to obtain coastline extraction points;

s400, obtaining a corrected coastline according to the coastline extraction points;

in S200, remote sensing images at each moment are respectively converted into corresponding NDWI images through calculating normalized water indexes, and the obtained NDWI images sequentially form an NDWI image sequence, wherein pixel values on the NDWI images are values belonging to [0,1] after being subjected to graying and normalization processing;

in S300, the method for obtaining the coastline extraction point by performing the penetration processing on the NDWI image sequence includes:

recording the NDWI image sequence as a sequence NDseq, recording the number of NDWI images serving as elements in the NDWI image sequence as t, enabling the sequence number of the elements in the NDWI image sequence to be q, keeping the sequence number of each moment in a plurality of different moments consistent with the sequence number q, and enabling the element with the q-th sequence number in the sequence NDseq to be ND (q);

each NDWI image in the NDWI image sequence is an image matrix with the size of n rows and m columns, i is the serial number of a row and j is the serial number of a column in each NDWI image, and the numerical value of the j-th column position of the i-th row in ND (q) is represented as ND (q) [ i, j ];

directing the next of the last element in the NDWI image sequence to the first element;

the function TimeGAP (,) represents the time interval that the time of the subsequent sequence number in the bracket is calculated to pass from the time of the preceding sequence number;

calculating the time sequence penetration degree of each NDWI image: recording the time sequence penetration degree of ND (q) as a period (q), and acquiring the last element ND (q-1) of ND (q) in the NDWI image sequence and the numerical value ND (q-1) [ i, j ] of the element ND (q), wherein the calculation formula of period (q) is as follows:

calculating an arithmetic average value of time sequence penetration of each NDWI image in an NDWI image sequence as a time sequence penetration threshold of the NDWI image sequence, and screening out the NDWI images of which the corresponding time sequence penetration in the NDWI image sequence reaches the time sequence penetration threshold as super-threshold images;

multiplying each super-threshold image by the corresponding time sequence penetration degree on the pixel value of each pixel point of each super-threshold image so as to convert each super-threshold image into time sequence penetration degree images, and multiplying each time sequence penetration degree image together by Hadamard product to obtain an accumulated image matrix;

on the cumulative image matrix, the numerical value of each row and column position is converted into a numerical value with the positive or the negative through a binarization algorithm, wherein the numerical value with the positive or the negative represents the numerical value or sign with the logical inverse relation, the numerical value including the numerical value of 0 or 1, the logical AND logic NOT and the Boolean value, and then the point of the row and column position converted into the positive numerical value is selected as a coastline extraction point.

2. The method for extracting a coastline from a sand of a high-resolution remote sensing image according to claim 1, wherein in S100, time intervals among the time points in the plurality of different time points are the same, the order of arrangement of the time points in the plurality of different time points is in time sequence, each remote sensing image corresponds to each time point, wherein each time point corresponds to only one remote sensing image, and the remote sensing image is the high-resolution remote sensing image of the coastline.

3. The method for extracting a coastline from a sand of a high-resolution remote sensing image according to claim 1, wherein in S400, the method for obtaining a corrected coastline according to the coastline extraction point comprises: and taking a curve obtained by curve fitting or interpolation on the coastline extraction points as the corrected coastline.

4. A sandy coastline extraction system of high resolution remote sensing image, characterized in that the sandy coastline extraction system of high resolution remote sensing image is operated in any computing device of a desktop computer, a notebook computer or a cloud data center, the computing device comprising: a processor, a memory and a computer program stored in the memory and running on the processor, the processor implementing the steps in a method for extracting a sandy coastline of a high resolution remote sensing image as claimed in any one of claims 1 to 2 when the computer program is executed.

Images