KR102526638B1 - Time-series shoreline detection device using multi-time satellite image and artificial intelligence and driving method of the device - Google Patents

Abstract

The present invention relates to a time-series shoreline detection device using a multi-time satellite images and artificial intelligence and a method of driving the device. A time-series shoreline detection device using multi-time satellite images and artificial intelligence according to an embodiment of the present invention may include: a communication interface part that receives multi-time satellite images according to changes in time (t) for any same shoreline area; and a control part that compares and analyzes the received satellite images of multiple periods according to time changes by applying artificial intelligence (AI) and recognizes changes in coastal topography that occurred in the same shoreline area based on the analysis results.

Description

Time-series coastline detection device using multi-period satellite imagery and artificial intelligence and method of driving the device

The present invention relates to a time-series coastline detection device using multi-period satellite imagery and artificial intelligence and a method for operating the device, and more particularly, to be effectively used for monitoring topographical changes and measuring coastal erosion in coastal areas It relates to a time-series coastline detection device using multi-period satellite images and artificial intelligence using N satellite images and artificial intelligence taken at multiple times and a method for operating the device.

The coastline is the boundary line dividing the sea and the land, and is an important marine spatial information data that defines the shape of the land. Coastlines are constantly changing due to natural or artificial causes such as coastal development and climate change, so coastline change information obtained using time-series data is needed to manage vast coastal areas.

Conventionally, a technology for outputting only one coastline from satellite images taken at a single time and a technology for detecting coastlines in small coastal areas have been developed using CCTV images. Acquisition is limited.

Korea Patent Registration No. 10-1690704 (2016.12.22) Korean Registered Patent Publication No. 10-2278419 (2021.07.12) Korean Registered Patent Publication No. 10-1877173 (2018.07.04) Korean Patent Publication No. 10-2022-0000898 (2022.01.04)

An embodiment of the present invention uses multi-period satellite images and artificial intelligence to utilize N satellite images and artificial intelligence captured at multiple times in order to be effectively used for monitoring topographical changes and measuring coastal erosion in coastal areas, for example. The purpose is to provide a time-series coastline detection device using multi-period satellite images and artificial intelligence and a method of driving the device.

An apparatus for detecting a time-series coastline using multi-period satellite images and artificial intelligence according to an embodiment of the present invention includes a communication interface unit for receiving multi-period satellite images according to a change in time (t) for the same coastline area, and and a control unit that compares and analyzes the received multi-time satellite images according to the time change by applying artificial intelligence (AI) and recognizes a change in the coastal topography occurring in the same coastline area based on the analysis result.

The control unit sets a learning area for applying the artificial intelligence to land cover including water, soil or vegetation in the received satellite image, and pixel brightness value information about the location of the set land cover learning area. The above comparative analysis operation may be performed by setting as a training sample for each land cover.

The control unit may obtain the pixel brightness value information of each band from individual multispectral bands of the satellite image.

The control unit generates a coastal land cover image corresponding to the entire satellite image of a plurality of coastline areas, and generates a binary image divided into water and land by merging land cover of soil and vegetation in the generated coastal land cover image. It is possible to remove errors located around the boundary between water and land in the vivid binary image.

The control unit extracts boundary lines from the error-removed binary image in a GIS (Geographic Information System) shape file format, and sets a line object having the longest length among the boundary lines of the extracted GIS shape file format as a coastline. available for detection.

When comparing and analyzing the multi-time satellite images by applying artificial intelligence (AI), the control unit determines whether or not the satellite images are captured by the same sensor or camera, and performs a comparative analysis operation based on the determination result. there is.

In addition, in the driving method of the time-series coastline detection device using multi-period satellite images and artificial intelligence according to an embodiment of the present invention, the communication interface detects multi-period satellite images according to the change in time (t) for the same coastline area. Receiving, and a controller comparing and analyzing the multi-time satellite images according to the received time change by applying artificial intelligence (AI), and recognizing changes in the coastal topography occurring in the same coastline area based on the analysis result. include

The recognizing may include setting a learning area for applying the artificial intelligence to land cover including water, soil or vegetation in the received satellite image, and determining the location of the set land cover learning area. and performing the comparative analysis operation by setting pixel brightness value information as a training sample for each land cover.

In the recognizing, the pixel brightness value information of each band may be obtained from individual multispectral bands of the satellite image.

The recognizing may include generating a coastal land cover image corresponding to entire satellite images of a plurality of coastline areas, and merging land cover of soil and vegetation in the generated coastal land cover image to obtain a binary image divided into water and land. It may include generating an image, and removing an error located around a boundary between water and land in the vivid binary image.

In the recognizing, the boundary lines are extracted in the GIS shape file format from the binary image from which the error is removed, and a line object having the longest length among the boundary lines of the extracted GIS shape file format is set as the coast line to be used when detecting the coast line. can

In the step of recognizing, when comparing and analyzing the multi-time satellite image by applying artificial intelligence (AI), it is determined whether the satellite image is captured by the same sensor or camera, and the comparison and analysis operation is performed based on the determination result can do.

According to an embodiment of the present invention, it is possible to calculate related statistics without visiting the site by measuring the moving distance of the time-series coastline calculated from multi-temporal satellite images of topographical changes and erosion phenomena occurring in vast coastal areas. It can be used as an important reference for marine vessel survey work performed by the Korea Hydrographic and Oceanographic Agency.

In addition, according to an embodiment of the present invention, when an insurance company launches a disaster insurance product related to coastal erosion and calculates insurance premiums, differential insurance premiums are set by applying regional weights according to the possibility of coastal erosion based on the calculation results and statistics according to the embodiment of the present invention. You will be able to.

1 is a diagram showing a time-series coastline detection system according to an embodiment of the present invention;
2 to 8 are diagrams for explaining a time-series shoreline detection process of the shoreline detection device of FIG. 1;
9 is a block diagram illustrating a detailed structure of the shoreline detection device of FIG. 1, and
10 is a flowchart illustrating a driving process of the shoreline detecting device of FIG. 1

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

1 is a diagram showing a time-series coastline detection system according to an embodiment of the present invention, and FIGS. 2 to 8 are diagrams for explaining a time-series coastline detection process of the shoreline detection device of FIG. 1 .

As shown in FIG. 1, the time-series coastline detection system 90 according to an embodiment of the present invention includes a satellite (device) 100, a communication network 110, a coastline detection device 120, and a third party device 130. include some or all

Here, “including some or all” means that the time-series coastline detection system 90 is configured by omitting some components such as the third party device 130, or some or all of the components such as the coastline detection device 120 All of them mean things that can be configured by being integrated into a network device (eg, wireless switching device, etc.) constituting the communication network 110, and will be described as including all of them to help a sufficient understanding of the present invention.

The artificial satellite 100 is equipped with a satellite camera for observing the earth in outer space. A satellite camera, referred to as the 'eye of the satellite', is divided into an optical camera, an infrared camera, and a radar. The artificial satellite 100 in FIG. 1 may include various types of satellite devices for obtaining landsat imagery according to an embodiment of the present invention. For example, in an embodiment of the present invention, in order to collect satellite images related to coastline detection, images may be collected by accessing a server such as the National Meteorological Satellite Center. For example, it is possible to receive satellite images related to coastline detection in real time by linking API (Application Programming Interface) to the server of the National Weather Service. Of course, the satellite 100 periodically transmits captured images through communication with the National Weather Service Center.

The artificial satellite 100 may include an optical camera for observing the earth or various types of sensors (e.g., R, G, B sensors constituting a CCD camera, etc.). Characteristics may be somewhat different. Therefore, in the embodiment of the present invention, it is possible to determine whether the same product is used in a sensor or the like for accurate detection of the coastline. It can be done as much as possible by pre-saving and checking it. Above all, the satellite 100 according to an embodiment of the present invention can provide optical and multispectral images.

The communication network 110 performs satellite communication with the satellite 100 . Since satellite communication uses microwaves, high-speed and large-capacity communication is possible, and a wide area (eg, the whole of a specific country) can be covered as a communication area. In addition, even communication is possible regardless of terrain, and communication is not restricted even in the event of a disaster. However, there are disadvantages in that the round-trip time (about 0.24 seconds) of the radio wave is delayed during voice communication, and there is no security of information. In addition, since solar cells are used as a power source, a momentary loss of communication may occur when the satellite is in the shadow of the earth or when heavy rain falls.

The communication network 110 includes both wired and wireless communication networks. For example, as the communication network 110, a wired or wireless Internet network may be used or interlocked. Here, the wired network includes an Internet network such as a cable network or a public switched telephone network (PSTN), and the wireless communication network includes CDMA, WCDMA, GSM, EPC (Evolved Packet Core), LTE (Long Term Evolution), and Wibro networks. meaning to include Of course, the communication network according to the embodiment of the present invention is not limited thereto, and can be used as an access network of a next-generation mobile communication system to be implemented in the future, for example, a cloud computing network under a cloud computing environment, a 5G network, and the like. For example, if the communication network is a wired communication network, the access point in the communication network 110 can access an exchange center of a telephone company, but in the case of a wireless communication network, access to an SGSN or GGSN (Gateway GPRS Support Node) operated by a communication company to process data, Data can be processed by connecting to various repeaters such as BTS (Base Transceiver Station), NodeB, and e-NodeB.

The communication network 110 may include an access point. Access points include small base stations such as femto or pico base stations that are often installed in buildings. Here, the femto or pico base stations are classified according to the maximum number of shoreline detection devices 120 or third party devices 130 that can be connected to the small base stations. Of course, the access point includes a short-range communication module for performing short-range communication such as ZigBee and Wi-Fi with the shoreline detection device 120 or the third party device 130. The access point may use TCP/IP or Real-Time Streaming Protocol (RTSP) for wireless communication. Here, short-range communication may be performed in various standards such as Bluetooth, ZigBee, infrared (IrDA), RF (Radio Frequency) such as UHF (Ultra High Frequency) and VHF (Very High Frequency), and ultra-wideband communication (UWB) in addition to Wi-Fi. can Accordingly, the access point can extract the location of the data packet, designate the best communication path for the extracted location, and forward the data packet to the next device, for example, the coastline detecting device 120, along the designated communication path. An access point may share several lines in a general network environment, and includes, for example, a router, a repeater, and a repeater.

The coastline detection device 120 detects N time-series coastlines by using artificial intelligence techniques on N multi-temporal satellite images acquired using the same sensor (eg, optical camera, etc.), and detects them as a GIS (Geographic Information System) shape file. (shapefile) format. Alternatively, the shoreline detection device 120 may provide managers with a screen for coastline detection by loading and executing a program for outputting in a GIS shape file format. Here, the shape file refers to a vector-type file structure used to store location and attribute information of points, lines, and planes. In this process, the coastline detecting device 120 collects satellite images of a plurality of coastline areas and combines them, so that it can be effectively used for monitoring topographic changes and measuring coastal erosion in coastal areas.

The coastline detection device 120 detects N coastlines by applying an artificial intelligence technique to N satellite images acquired in a vast coastal area at multiple times. Of course, the coastline detecting device 120 can directly receive the satellite image provided by the artificial satellite 100, but the satellite image cannot receive related images from a third party device 130 such as the National Meteorological Satellite Center. may be

For the shoreline detection operation, as shown in FIG. 2, the shoreline detection device 120 sets the main land cover (eg, water, soil, vegetation, etc.) area in the satellite image and the brightness of pixels located in the main land cover. Value information is obtained from individual multispectral bands and defined (or set) as training samples for each land cover. In this process, the shoreline detection device 120 may classify and store brightness value information of pixels in each band with respect to the location and area of a training sample for water as land cover, for example, as shown in FIG. 2 .

Multispectral imaging is an image for acquiring image data belonging to a specific wavelength range of the electromagnetic spectrum. Wavelengths can be separated using filters or instruments sensitive to specific wavelengths, including light from frequencies outside the range of visible light, such as microwave, infrared, ultraviolet, and X-rays. That is, the multispectral image is a collection of several monochromatic images of the same scene, which are taken with different sensors. Here, each monochromatic image is called a band. A well-known multispectral (or multiband image) is an RGB color image composed of red, green, and blue images, which are captured by sensors sensitive to different wavelengths. Such spectral imaging may allow extraction of additional information that the human eye cannot capture with the receptors for red, green and blue colors. Multispectral imaging typically measures light in 3 to 15 spectral bands (or bands).

In addition, as shown in FIG. 3, the coastline detection device 120 learns training samples for each land cover with a Deep Neural Network (DNN) algorithm, which is an artificial intelligence technique, and applies the learned model to the entire satellite image to apply the coastal land cover. Produce (or create) a video. 3, (a) shows a satellite image and (b) shows a coastal land cover image, respectively.

Then, as shown in FIG. 4, the shoreline detection device 120 merges land cover (eg, soil, vegetation, etc.) from the coastal land cover image to finally produce a binary image divided into water and land. . Figure 4 (a) shows a coastal land cover image, and (b) shows a binary image.

Furthermore, as shown in FIG. 5, the shoreline detection device 120 applies dilation and erosion operations of geomorphology to binary images to detect errors located around the boundary between water and land in binary images. Remove. (a) of FIG. 5 shows an original binary image, and (b) shows a binary image from which errors have been removed. For example, erosion marks a 0 if there is even a single 1 in the kernel. Erosion is an operation in which, when a mask is placed on each pixel of an input binary image, the result value is 255 only when the input image also has a value of 255 for all pixel positions where the mask has a value of 255. If even one pixel at the target location has a value of 0, the resulting value becomes 0, resulting in a reduction in the area with a value of 255 as a whole. On the other hand, the dilation operation, in contrast to the erosion operation, serves to brighten all the pixels in the effective area of the mask.

Subsequently, as shown in FIG. 6, the coastline detection device 120 extracts the boundary lines in the GIS shape file format from the error-removed binary image, and defines (or designates) the line object with the longest length as the coastline . In FIG. 6, (a) represents a boundary line extracted from a binary image, and (b) represents a coastline selected from among the boundary lines.

Next, the coastline detection device 120 detects N time-series coastlines from N multi-period satellite images obtained by using the same sensor in the same coastal area and calculates them in the form of individual GIS shape files. 7 shows an exemplary view of coastlines individually detected from time-series satellite images.

Finally, the coastline detection device 120 compares multi-temporal satellite images to recognize changes in coastal topography caused by natural and artificial causes such as coastal erosion and port construction. 8 shows an example of recognizing changes in the coastal topography caused by port construction by utilizing coastlines detected from the first and second satellite images.

The third party device 130 may include various types of devices operated by the National Weather Service Center or may include various devices of program producers that develop programs according to embodiments of the present invention. For example, in the case of the national meteorological center, after receiving and storing a satellite image captured through the artificial satellite 100, it can be provided in real time upon request from the coastline detection device 120. In addition, in the case of a program manufacturer, a program for detecting a shoreline according to an embodiment of the present invention may be developed and installed in the shoreline detection device 120.

As a result of the configuration above, by using the technology and system according to the embodiment of the present invention, the movement distance of the time-series coastline calculated from multi-temporal satellite images of topographical changes and erosion phenomena occurring in vast coastal areas is measured without visiting the site. As related statistics can be calculated, it can be used as an important reference for marine vessel survey work performed by the National Oceanographic Survey Agency under the Ministry of Maritime Affairs and Fisheries. In addition, the embodiment of the present invention is based on the results and statistics calculated through this technology when an insurance company launches a coastal erosion-related disaster insurance product and calculates insurance premiums by applying regional weights according to the possibility of coastal erosion to set differential insurance premiums. You will be able to.

9 is a block diagram illustrating a detailed structure of the shoreline detection device of FIG. 1 .

As shown in FIG. 9, the coastline detection device 120 or the third party device 130 according to an embodiment of the present invention includes a communication interface unit 900, a control unit 910, a coastline detection unit 920, and a storage unit. All or part of (930).

Here, "including some or all" means that some components such as the storage unit 930 are omitted to form the coastline detection device 120 or some components such as the coastline detection unit 920 are configured by the control unit 910. It means that it can be integrated with other components such as, etc., and will be described as including all to help a sufficient understanding of the invention.

The communication interface unit 900 communicates with the satellite 100 or the third party device 130 via the communication network 110 of FIG. 1 . Through satellite communication with the artificial satellite 100, satellite images of the coastline 97 can be received in real time or periodically. Of course, it is preferable that the communication interface unit 900 periodically receives satellite images from the satellite 100 having the same optical characteristics as the optical camera of the satellite 100 . For example, the accuracy of detecting N time-series coastlines can be increased by applying artificial intelligence techniques to N multi-time satellite images acquired through the same sensor of an optical camera. The periodically received multi-period satellite images may be systematically classified and stored in the DB 120a of FIG. 1 .

In addition, when the communication interface unit 900 does not directly communicate with the satellite 100 through the communication network 110 of FIG. A satellite image of the artificial satellite 100 can be received. The third party device 130 may periodically receive satellite images related to the entire coastline through API interworking and provide the information to the control unit 910 .

Furthermore, the communication interface unit 900 may perform operations such as modulation/demodulation, muxing/demuxing, encoding/decoding, resolution conversion scaling, etc. in the process of performing communication, which are obvious to those skilled in the art, so further description is given. should be omitted.

The control unit 910 is in charge of overall control operations of the communication interface unit 900, the coastline detection unit 920, and the storage unit 930. For example, when the third party device 130 provides a previously produced program according to an embodiment of the present invention, the controller 910 performs an operation to load the program into the coastline detection unit 920 . In addition, the controller 910 may execute a program for detecting the shoreline by controlling the shoreline detection unit 920 to detect the shoreline.

The control unit 910 performs an operation for time-series shoreline detection by utilizing multi-temporal satellite images and artificial intelligence according to an embodiment of the present invention in conjunction with the coastline detection unit 920. To this end, the control unit 910 receives satellite images according to multiple times, that is, time changes, stores them in the storage unit 930, retrieves them, and provides them to the coastline detection unit 920. In addition, the control unit 910 communicates with the coastline detection unit 920 to systematically classify and store in the DB 120a of FIG. A communication operation of the interface unit 900 is controlled.

The coastline detection unit 920 performs a time-series coastline detection operation using multi-temporal satellite images and artificial intelligence. At this time, the coastline detector 920 detects N time-series coastlines by applying an artificial intelligence technique to N multi-temporal satellite images acquired using the same sensor such as an optical camera of the satellite 100 and generates them in a GIS shape file format. can do.

More specifically, the coastline detection unit 920 sets the main land cover area in the satellite image, acquires brightness value information of pixels located in the main land cover from individual multispectral bands, and defines this as a training sample for each land cover. It can be seen that various land covers such as water, soil, and vegetation are present in the satellite image, and training samples are obtained from the location and area of the training samples located there. Next, the coastline detection unit 920 trains the artificial intelligence-based DNN algorithm on the training samples for each land cover and applies the learned model to all satellite images to create (or generate) a coastal land cover image. In addition, the coastline detection unit 920 merges the land cover from the coastal land cover image to finally produce a binary image divided into water and land. In binary images, topographical dilation and erosion operations are applied to remove errors located around the boundary between water and land in binary images. In addition, the coastline detection unit 920 extracts the boundary lines in the GIS shape file format from the binary image from which errors are removed, defines (or sets, designates) a line object with the longest length among them as a coastline, and in the same coastal area Detect N time-series coastlines from N multi-period satellite images obtained using the same sensor, and calculate (or create) them as individual GIS shape file formats. Through this, the coastline detecting unit 920 recognizes changes in the coastal topography caused by natural and artificial causes such as coastal erosion and port construction in comparison with multi-period satellite images.

The storage unit 930 may temporarily store various types of data processed under the control of the controller 910 . The storage unit 930 may temporarily store the satellite image according to the request of the control unit 910, and output the satellite image at the request of the control unit 910 so that it is provided to the coastline detection unit 920. In addition, the storage unit 930 may store the analysis result of the satellite image of the coastline detection unit 920 in a designated format, and the analysis result is provided to the DB 120a of FIG. 1 under the control of the control unit 910 to systematically can be classified and stored.

In addition to the above, the communication interface unit 900, control unit 910, coastline detection unit 920, and storage unit 930 of FIG. 9 can perform various operations, and other details have been sufficiently described above. want to replace them with

The communication interface unit 900, the control unit 910, the coastline detection unit 920, and the storage unit 930 of FIG. 9 according to an embodiment of the present invention are composed of physically separated hardware modules, but each module is internally It will be possible to store and execute software for performing the above operation in . However, since the corresponding software is a set of software modules, and each module can be formed of hardware, it will not be particularly limited to the configuration of software and hardware. For example, the storage unit 930 may be hardware storage or memory. However, since it is possible to store information in a software manner (repository), the above content will not be particularly limited.

Meanwhile, as another embodiment of the present invention, the control unit 910 may include a CPU and a memory, and may be formed as a single chip. The CPU includes a control circuit, an arithmetic unit (ALU), a command interpreter, and a registry, and the memory may include RAM. The control circuit performs control operations, the operation unit performs operation of binary bit information, and the instruction interpretation unit performs an operation of converting high-level language into machine language and machine language into high-level language, including an interpreter or compiler. , the registry may be involved in software data storage. According to the above configuration, for example, at the beginning of the operation of the shoreline detection device 120, the program stored in the shoreline detection unit 920 is copied, loaded into a memory, that is, RAM, and then executed, thereby speeding up data operation processing speed can increase In the case of a deep learning model, it can be loaded into GPU memory rather than RAM, and can be executed by accelerating the execution speed using the GPU.

10 is a flowchart illustrating a driving process of the shoreline detecting device of FIG. 1

Referring to FIG. 10 together with FIG. 1 for convenience of explanation, the coastline detection device 120 according to an embodiment of the present invention receives multi-time satellite images according to a change in time (t) for the same coastline area ( S1000).

In addition, the coastline detecting device 120 compares and analyzes the received satellite images of multiple periods according to the time change by applying artificial intelligence, and recognizes a change in coastal topography occurring in the same coastline area based on the analysis result (S1010).

In this process, the coastline detection device 120 may detect N time-series coastlines by applying an artificial intelligence technique to N multi-temporal satellite images obtained using the same sensor such as an optical camera of the satellite 100, and at this time, GIS It is created in the shape file format and can perform actions such as defining, that is, setting or designating the line object with the longest length as a coastline. In addition, the coastline detecting device 120 converts the coastal land cover image into a binary image and then removes errors from the converted binary image. To this end, topographic dilation and erosion operations are applied to binary images.

In addition to the above, the shoreline detection device 120 of FIG. 1 may perform various operations, and since other details have been sufficiently described above, they will be replaced with those contents.

On the other hand, even if all the components constituting the embodiment of the present invention are described as being combined or operated as one, the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all of the components may be selectively combined with one or more to operate. In addition, although all of the components may be implemented as a single independent piece of hardware, some or all of the components are selectively combined to perform some or all of the combined functions in one or a plurality of pieces of hardware. It may be implemented as a computer program having. Codes and code segments constituting the computer program may be easily inferred by a person skilled in the art. Such a computer program may implement an embodiment of the present invention by being stored in a computer-readable non-transitory computer readable media, read and executed by a computer.

Here, the non-transitory readable recording medium means a medium that stores data semi-permanently and can be read by a device, not a medium that stores data for a short moment, such as a register, cache, or memory. . Specifically, the above-described programs may be provided by being stored in a non-transitory readable recording medium such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, or ROM.

Although the preferred embodiments of the present invention have been shown and described above, the present invention is not limited to the specific embodiments described above, and is common in the art to which the present invention pertains without departing from the gist of the present invention claimed in the claims. Of course, various modifications and implementations are possible by those with knowledge of, and these modifications should not be individually understood from the technical spirit or perspective of the present invention.

100: artificial satellite 110: communication network
120: coastline detection device 130: third party device
900: communication interface unit 910: control unit
920: coastline detection unit 930: storage unit

Claims (12)

A communication interface unit for receiving multi-time satellite images according to time (t) change for the same coastline area; and
A control unit that compares and analyzes satellite images of multiple periods according to the received time change by applying artificial intelligence (AI) and recognizes a change in the coastal topography occurring in the same coastline area based on the analysis result;
When the control unit compares and analyzes the multi-time satellite image by applying artificial intelligence (AI), it checks pre-stored specification information related to the performance of the artificial satellite providing the satellite image and captures the image captured by the same sensor or camera. It determines whether it is a satellite image and performs a comparative analysis operation based on the determination result,
The control unit obtains the pixel brightness value information from the monochromatic image of each band from individual multispectral bands of the satellite image in which several monochromatic images of the same scene are collected;
The control unit generates a coastal land cover image corresponding to the entire satellite image of a plurality of coastline areas, and generates a binary image divided into water and land by merging land cover of soil and vegetation in the generated coastal land cover image. In the vivid binary image, errors located around the boundary between water and land are removed, and dilation and erosion operations of geomorphology are applied to the binary image to determine the difference between water and land in the binary image. Eliminate errors located around the boundary,
The erosion operation is an operation in which, when a mask is placed on each pixel of the input binary image, the result value is 255 only when the input image also has the brightness value information 255 for all pixel positions in which the mask has the brightness value information 255. , Time-series coastline detection device using multi-temporal satellite images and artificial intelligence, wherein the dilation operation represents an operation to brighten all pixels in the effective area of the mask, as opposed to the erosion operation. According to claim 1,
The control unit sets a learning area for applying the artificial intelligence to land cover including water, soil or vegetation in the received satellite image, and pixel brightness value information about the location of the set land cover learning area. A time-series coastline detection device using multi-period satellite imagery and artificial intelligence to perform the comparative analysis operation by setting as a training sample for each land cover. delete delete According to claim 1,
The control unit extracts boundary lines from the error-removed binary image in a GIS (Geographic Information System) shape file format, and sets a line object having the longest length among the boundary lines of the extracted GIS shape file format as a coastline. A time-series coastline detection device using multi-period satellite images and artificial intelligence used for detection. delete Receiving, by a communication interface unit, multi-time satellite images according to time (t) changes with respect to an arbitrary same coastline area; and
Recognizing, by the control unit, the coastal topography change occurring in the same coastline area based on the analysis result by comparing and analyzing the received multi-period satellite images according to the time change by applying artificial intelligence (AI);
The step of recognizing is
When the multi-period satellite image is compared and analyzed by applying artificial intelligence (AI), pre-stored specification information related to the performance of the artificial satellite providing the satellite image is checked to determine whether the satellite image is captured by the same sensor or camera. determining and performing a comparative analysis operation based on the determination result;
obtaining the pixel brightness value information from the monochromatic image of each band from individual multispectral bands of the satellite image in which several monochromatic images of the same scene are collected; and
A coastal land cover image corresponding to the entire satellite image of a plurality of coastline areas is generated, and a binary image divided into water and land is generated by merging land cover of soil and vegetation in the generated coastal land cover image, and generating the coastal land cover image. removing errors located around the boundary between water and land in a binary image, and removing errors located around the boundary between water and land in the binary image by applying topographic dilation and erosion operations to the binary image; and
The erosion operation is an operation in which, when a mask is placed on each pixel of the input binary image, the result value is 255 only when the input image also has the brightness value information 255 for all pixel positions in which the mask has the brightness value information 255. , The dilation operation is a method of driving a time-series coastline detection device using multi-temporal satellite images and artificial intelligence, which represents an operation to brighten all pixels in the effective area of the mask, as opposed to the erosion operation. According to claim 7,
The step of recognizing is
setting a learning area for applying the artificial intelligence to land cover including water, soil, or vegetation in the received satellite image; and
setting pixel brightness value information for the location of the set land cover learning area as a training sample for each land cover and performing the comparative analysis operation;
A method for operating a time-series coastline detection device using multi-period satellite images and artificial intelligence. delete delete According to claim 7,
The step of recognizing is
Multi-temporal satellite imagery and artificial intelligence used when detecting a coastline are extracted from the binary image from which the error has been removed in a GIS shape file format, and a line object having the longest length among the boundary lines of the extracted GIS shape file format is set as a coastline. A driving method of a time-series coastline detection device using intelligence. delete

Images