KR102532732B1 - Apparatus and method for generating a vegetation index - Google Patents

Abstract

The present invention relates to a method for generating a vegetation index. An apparatus for generating a vegetation index according to some embodiments extracts a data collection module for collecting satellite image data, a first image feature associated with vegetation, a wavelength feature of a visible band, and a photosynthesis feature of a near-infrared band from the satellite image data. It includes a feature extraction module and a determination module for generating a vegetation index based on the extracted first image feature, wavelength feature, and photosynthesis feature, wherein the wavelength feature includes first wavelength data for a first wavelength among visible band data and , may be calculated based on second wavelength data for a second wavelength having a relatively shorter wavelength than the first wavelength.

Description

Vegetation index generating device and method {APPARATUS AND METHOD FOR GENERATING A VEGETATION INDEX}

The present invention relates to an apparatus and method for generating a vegetation index.

Currently, hundreds of artificial satellites around the world are performing their own tasks in the earth's air for the purposes of meteorology, communication, broadcasting, agriculture, space development, and military.

Although artificial satellites can observe locations that are difficult for humans to directly explore, such as forest areas, the current image processing technology of artificial satellites is difficult to accurately detect the area of forest areas, and the calculation process is complicated.

An object to be solved by the present invention is to provide a vegetation index generating device.

Another problem to be solved by the present invention is to provide a method for generating a vegetation index.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned above can be understood by the following description and will be more clearly understood by the examples of the present invention. It will also be readily apparent that the objects and advantages of the present invention may be realized by means of the instrumentalities and combinations indicated in the claims.

An apparatus for generating a vegetation index according to some embodiments extracts a data collection module for collecting satellite image data, a first image feature associated with vegetation, a wavelength feature of a visible band, and a photosynthesis feature of a near-infrared band from the satellite image data. It includes a feature extraction module and a determination module for generating a vegetation index based on the extracted first image feature, wavelength feature, and photosynthesis feature, wherein the wavelength feature includes first wavelength data for a first wavelength among visible band data and , may be calculated based on second wavelength data for a second wavelength having a relatively shorter wavelength than the first wavelength.

On the other hand, the data collection module further collects wind speed data as other data, and the feature extraction module, based on the change in seconds of the extracted first image feature when the collected wind speed data is equal to or greater than a predetermined speed, It may be verified whether the extracted first image feature corresponds to the first image feature associated with vegetation.

Meanwhile, the data collection module may further collect seasonal data as other data.

Meanwhile, the feature extraction module may verify whether the extracted first image feature corresponds to a first image feature associated with vegetation, based on a shape change of the extracted first image feature according to a change in seasonal data.

Meanwhile, the feature extraction module may verify whether the extracted first image feature corresponds to the first image feature associated with vegetation based on the extracted wavelength feature and the photosynthetic feature change according to the seasonal data change.

Meanwhile, the data collection module further collects vegetation area data as other data, and the feature extraction module determines an Area of Interest (AoI) based on the vegetation area data, and in the determined area of interest, the first image The occupied area of the feature can be determined.

Meanwhile, the feature extraction module extracts a plurality of first image features from the satellite image data corresponding to the coordinates of the vegetation area data, derives an outline of a set of extracted first image features, and derives an outline of interest based on the derived outline. area can be determined.

Meanwhile, the feature extraction module extracts a second image feature associated with noise from the satellite image data, subtracts the area occupied by the second image feature from the area of the determined region of interest, and determines the area occupied by the first image feature. there is.

Meanwhile, a learning module for learning a deep learning model using training data including at least one of a first image feature, a wavelength feature, and a photosynthesis feature, wherein the determination module includes the first image feature, the wavelength feature, and the photosynthesis feature. At least one of them may be input to a deep learning model to generate a vegetation index.

A vegetation index generation method according to some embodiments includes the step of collecting satellite image data, the feature extraction step of extracting a first image feature associated with vegetation, a wavelength feature of a visible band, and a photosynthetic feature of a near-infrared band from the satellite image data, and Generating a vegetation index based on the extracted first image feature, wavelength feature, and photosynthesis feature, wherein the wavelength feature is relative to first wavelength data for a first wavelength among visible band data and relative to the first wavelength It can be calculated based on the second wavelength data for the second wavelength having a short wavelength.

By extracting image features, wavelength features, and photosynthesis features and using them together, the vegetation index can be determined more accurately.

In addition to the above description, specific effects of the present invention will be described together while explaining specific details for carrying out the present invention.

1 illustrates a vegetation index generating system according to some embodiments.
2 is a block diagram of an apparatus for generating a vegetation index according to some embodiments.
3 is a block diagram of a feature extraction module in accordance with some embodiments.
4A is a diagram for explaining a process of verifying a first image feature according to some embodiments.
4B is a diagram for explaining a process of verifying a first image feature according to some other embodiments.
5A is a diagram for explaining a process of determining a region of interest according to some embodiments.
5B is a diagram for describing a process of determining a region of interest and a process of determining an area occupied by a first image feature, according to some embodiments.
6 is a flowchart of a method for generating a vegetation index according to some embodiments.
7 is a flowchart of a method for generating a vegetation index according to some other embodiments.

Terms or words used in this specification and claims should not be construed as being limited to a general or dictionary meaning. According to the principle that an inventor may define a term or a concept of a word in order to best describe his/her invention, it should be interpreted as meaning and concept consistent with the technical spirit of the present invention. In addition, the embodiments described in this specification and the configurations shown in the drawings are only one embodiment in which the present invention is realized, and do not represent all of the technical spirit of the present invention, so they can be replaced at the time of the present application. It should be understood that there may be many equivalents and variations and applicable examples.

Terms such as first, second, A, and B used in this specification and claims may be used to describe various components, but the components should not be limited by the terms. Terms are only used to distinguish one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The term 'and/or' includes a combination of a plurality of related listed items or any of a plurality of related listed items.

Terms used in this specification and claims are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. It should be understood that terms such as "include" or "having" in this application do not exclude in advance the possibility of existence or addition of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification. .

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't In addition, each configuration, process, process or method included in each embodiment of the present invention may be shared within a range that does not contradict each other technically.

Also, the term "module" used in the specification means a software or hardware component, and a "module" performs certain roles. However, "module" is not meant to be limited to software or hardware. A "module" may be configured to reside on an addressable storage medium and may be configured to reproduce one or more processors. Thus, as an example, a "module" may include components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. Functionality provided within components and “modules” may be combined into fewer components and “modules” or further separated into additional components and “modules”.

Hereinafter, referring to FIGS. 1 to 8 , an apparatus and method for generating a vegetation index according to some embodiments of the present invention will be described.

1 illustrates a vegetation index generating system 1 according to some embodiments.

The vegetation index generating system 1 may include a satellite image capture device 110 , an external data server 115 and a vegetation index generating device 120 .

The satellite image capture device 110 is a device for capturing satellite image data, and may include, for example, an artificial satellite located on the earth's surface. The orbit and movement method of the satellite image capture device 110 may vary, and for some examples, the satellite image capture device 110 may include a low-orbit satellite, a polar-orbit satellite, and a geostationary orbit satellite.

The satellite image capture device 110 transmits a signal directly from a passive sensor and/or satellite that observes energy re-radiated after reflecting or absorbing solar energy by an object on the earth and a signal reflected back by an object on the earth. It may include an active sensor or the like that measures .

For some examples, the satellite image capture device 110 may include an optical sensor, a multi-spectral sensor, a near-infrared sensor, an infrared sensor, a radar sensor, and the like. Satellite image data captured by the satellite image capture device 110 may include an optical signal, a multi-spectral signal, a near-infrared signal, an infrared signal, a radar image, and the like. In this case, as some examples, the radar image may include a synthetic aperture radar image, that is, a synthetic aperture radar (SAR) image, but is not limited thereto.

Satellite image data photographed by the satellite image capture device 110 may include data of various wavelength bands (bands). For some examples, the satellite image data may include ultraviolet (UV) band data, visible band data, near infrared band data, infrared band data, and the like.

The satellite image capture device 110 may capture satellite image data containing the image of the earth at a predetermined location on the earth's surface, and the captured satellite image data may be transmitted to the vegetation index generating device 120 . At this time, the vegetation index generating system 1 may include a separate network and/or communication network for satellite communication between the satellite image capture device 110 and the vegetation index generating device 120.

The external data server 115 may be one or more servers that store other data.

Other data include, for example, meteorological data (e.g. climate data, weather data, precipitation data), wind speed data, cloud status data, typhoon data, tidal wave data, fire data, seasonal data, seasonal vegetation status data, vegetation area data and the like. Vegetation area data may include, for example, coordinates in which a forest area is formed over a predetermined range.

The vegetation index generator 120 may receive data from the satellite image capture device 110 and the external data server 115, and store, manage, and/or analyze the received data.

For some examples, the vegetation index generator 120 may generate analysis data associated with vegetation based on received satellite image data and other data.

The detailed configuration and operating method of the vegetation index generator 120 will be described in detail with reference to FIG. 2 .

2 is a block diagram of a vegetation index generating device 120 according to some embodiments.

Referring to FIG. 2 , the vegetation index generator 120 may include a data collection module 210 , a feature extraction module 220 , a determination module 230 and a learning module 240 .

The data collection module 210 may collect data for generating analysis data associated with vegetation.

For some examples, the data collection module 210 may collect satellite image data and other data. At this time, as described above, the satellite image data may be captured by the satellite image capture device 110, and other data may be data stored and managed by the external data server 115.

As described above, satellite image data may include data of various wavelength bands (bands). As some examples, the satellite image data may include ultraviolet (UV) band data, visible band data, near infrared band data, infrared band data, etc., and other data may include , meteorological data (eg, climate data, weather data, precipitation data), wind speed data, seasonal data, seasonal vegetation status data, vegetation area data, and the like.

The feature extraction module 220 may extract features associated with vegetation from satellite image data and/or other data.

As some examples, features associated with vegetation may include image features associated with vegetation, areas occupied by image features, temperature characteristics, wavelength characteristics, photosynthesis characteristics, etc., but are not limited thereto. A detailed configuration of the feature extraction module will be described later with reference to FIGS. 3 to 5 .

The determination module 230 may generate analysis data associated with vegetation based on the extracted features.

For some examples, the determination module 230 may generate analysis data associated with the vegetation index by using a predetermined mathematical formula for the relationship between the extracted features and/or collected other data and the vegetation index.

For example, the determination module 230 inputs a first image feature associated with vegetation, an area occupied by the first image feature, a wavelength feature, a photosynthesis feature, and other collected data into the deep learning model received from the learning module 240. Thus, analysis data related to vegetation can be generated. The deep learning model will be described later in the learning module 240.

Analysis data related to vegetation (hereinafter referred to as “vegetation index”) includes, for example, the ratio of forest area in a certain area, the area of forest area, crop distribution information in forest area, the ratio of living trees and dead trees in a specific area, etc. It may include, but is not limited thereto.

For some examples, the determination module 230 does not determine the vegetation index, and the data collection module ( 210) to collect satellite images again, and/or to control feature extraction module 220 to extract features from satellite images again.

In some other examples, the determination module 230 determines that the feature that does not meet the standard in the deep learning model when at least one of the extracted first image feature, the area occupied by the first image feature, the wavelength feature, and the photosynthesis feature does not meet the predetermined criterion. It is possible to assign a low weight to the corresponding item. However, it is not limited to the above example.

The learning module 240 may perform learning using training data for generating a vegetation index.

For some examples, the learning module 240 may train a deep learning model that generates a vegetation index.

For example, the learning module 240 may train a deep learning model using training data including collected satellite image data, other data, and/or features associated with vegetation. As described above, the feature associated with vegetation may include a first image feature, an area occupied by the first image feature, a wavelength feature, a photosynthesis feature, and the like.

At this time, the learned deep learning model may perform an operation of outputting a vegetation index as output data by using satellite image data, other data, and/or vegetation-related features as input data based on machine learning. .

Deep learning technology, a type of machine learning, learns by going down to a multi-level deep level based on data. That is, deep learning represents a set of machine learning algorithms that extract core data from a plurality of data while stepping up.

The deep learning model may use various known deep learning structures. For example, the deep learning model may use a structure such as a convolutional neural network (CNN), a recurrent neural network (RNN), a deep belief network (DBN), or a graph neural network (GNN).

Specifically, CNN (Convolutional Neural Network) extracts the basic features of an object when a person recognizes an object, and then performs complex calculations in the brain to recognize the object based on the result. It is a simulated model.

RNN (Recurrent Neural Network) is widely used in natural language processing, etc., and is an effective structure for processing time-series data that changes over time.

DBN (Deep Belief Network) is a deep learning structure composed of multiple layers of RBM (Restricted Boltzman Machine), a deep learning technique. When a certain number of layers is obtained by repeating RBM (Restricted Boltzman Machine) learning, a DBN (Deep Belief Network) having a corresponding number of layers may be configured.

GNN (Graphic Neural Network, hereinafter, GNN) represents an artificial neural network structure implemented in a way to derive similarities and feature points between modeling data using modeling data modeled on the basis of data mapped between specific parameters. .

On the other hand, learning of the artificial neural network of the deep learning model can be performed by adjusting the weight of the connection line between nodes (and adjusting the bias value if necessary) so that a desired output is produced for a given input. In addition, the artificial neural network may continuously update a weight value by learning. In addition, a method such as back propagation may be used to learn the artificial neural network.

In this case, unsupervised learning, semi-supervised learning, supervised learning, and the like may be used as the machine learning method of the artificial neural network. In addition, the deep learning model can be controlled to automatically update the artificial neural network structure for outputting analysis data after learning according to settings.

The deep learning model learned in the learning module 240 may be delivered to the decision module 230 and stored separately or an existing model may be updated.

As another example, the learning module 240 may determine the band of the first wavelength and the second wavelength, which are the basis of the wavelength feature, using training data such as collected satellite image data and other data.

In general, plants can reflect part of the visible band (wavelength range) and absorb part. In some instances, plants may reflect a first wavelength and absorb a second wavelength. At this time, when a vegetation index is generated, plants existing in an area where a vegetation index is generated may be burned due to a fire. When the vegetation is burned, it can be predicted that reflectance and/or absorption of the first wavelength and the second wavelength will be different in the corresponding area compared to before the vegetation index is generated.

Using this principle, the learning module 240 compares data before and after the occurrence of the vegetation index to determine a band of the first wavelength and the second wavelength that causes a difference in reflectance and/or absorbance of predetermined coordinates.

As such, by using the wavelength features extracted using visible band data, it is possible to have a remarkable effect of extracting features related to vegetation and/or generating a vegetation index using only visible band data.

3 is a block diagram of a feature extraction module 220 in accordance with some embodiments.

Referring to FIG. 3 , the feature extraction module 220 may include an image feature extractor 310 , a wavelength feature extractor 320 , and a photosynthetic feature extractor 330 .

The image feature extraction unit 310 may include a first image feature extraction unit 311 , a second image feature extraction unit 312 , a verification unit 313 and an area determination unit 314 . In this case, the first image feature may be an image feature related to vegetation, and the second image feature may mean an image feature related to noise. For some examples, image features associated with noise may refer to image features of objects not included in vegetation, such as roads, waterways, rocks, and the like.

The first image feature extractor 311 may extract first image features associated with vegetation from the collected satellite image data.

For some examples, the first image feature associated with vegetation may include an outline of an object or the like. In this case, the object may include trees, meadows, forests, forests, etc., but is not limited thereto. For convenience of explanation, the first image feature associated with vegetation will be described as a tree image feature.

For some examples, the first image feature extractor 311 may extract tree image features from satellite image data using a deep learning model that recognizes an object from an image learned by the learning module 240 .

As another example, the first image feature extractor 311 may extract tree image features from satellite image data using a predetermined object recognition model including an algorithm for recognizing an object from an image. The object recognition model may include, for example, FCN, Mask R-CNN, U-Net, image contour analysis model, etc., but is not limited thereto.

The second image feature extractor 312 may extract second image features associated with noise from the collected satellite image data.

The second image feature may refer to an image feature associated with noise. For some examples, image features associated with noise may refer to image features of objects not included in vegetation, such as roads, waterways, rocks, and the like.

For some examples, the second image feature extractor 312 may extract image features associated with noise from satellite image data using a deep learning model that recognizes an object from an image learned by the learning module 240 .

As another example, the second image feature extractor 312 may extract image features associated with noise from satellite image data using a predetermined object recognition model including an algorithm for recognizing an object from an image. The object recognition model may include, for example, FCN, Mask R-CNN, U-Net, image contour analysis model, etc., but is not limited thereto.

The verifier 313 may verify whether the image feature extracted by the first image feature extractor 311 corresponds to the tree image feature.

With satellite images, vegetation (e.g., trees) may not be clearly detected due to limitations in spatial resolution, and it may not be possible to clearly distinguish cases where fires occur, trees are partially cut down, or withered at the end of their lifespan. Accordingly, a process of verifying whether the extracted first image feature corresponds to an image feature related to vegetation (eg, a tree) may be required. The present invention has a remarkable effect that a vegetation index can be further generated by verifying whether or not the first image feature extracted through the method according to some exemplary embodiments corresponds to the tree image feature.

For some examples, the verifier 313 may verify whether the extracted first image feature corresponds to a tree image feature based on a second-unit change of the extracted first image feature. This will be described in detail with reference to FIG. 4A.

4A is a diagram for explaining a process of verifying a first image feature according to some embodiments. (1) and (2) of FIG. 4A show some examples of extracted first image features.

In the case of a general tree, when the wind speed is higher than a certain speed, the leaves and branches may be shaken by the wind, and accordingly, the extracted first image features may be partially different. In (1) of FIG. 4A, a first image feature extracted at a specific time point is shown, and in (2) of FIG. Different extractions in seconds are shown with dotted lines.

That is, the verification unit 313 determines that, among other collected data, when the wind speed data is equal to or greater than a predetermined speed and there is a change in units of seconds in the extracted first image feature, the extracted first image feature is a tree image. It can be identified as a feature.

Due to the verification process through the change in seconds of the extracted first image feature, it has a remarkable effect that non-living trees can be distinguished and, accordingly, a more accurate vegetation index can be generated.

As another example, the verifier 313 may verify whether the extracted first image feature corresponds to a tree image feature based on a change in features according to a change in seasonal data.

For some examples, the verification unit 313 may verify whether the extracted first image feature corresponds to a tree image feature based on a shape change of the first image feature according to a change in seasonal data. It is explained in detail with reference to FIG. 4B.

4B is a diagram for explaining a process of verifying a first image feature according to some other embodiments.

In the case of trees, unlike dead trees, their shapes may be partially different depending on the change of season. For some examples, trees may have luxuriant leaves in spring and summer, but relatively reduce the number of leaves in autumn and winter. In (1) of FIG. 4B, a first image feature extracted when leaves are lush in spring and summer is shown, and in (2) of FIG. 4B, extracted when the number of leaves is relatively reduced in autumn and winter. A first image feature is shown.

That is, the verification unit 313 may perform a change in the extracted first image feature, for example, a change in the number of outlines of the first image feature, when the seasonal data among the collected other data exists. It can be confirmed that the image feature is a tree image feature.

Due to the verification process through the seasonal change of the extracted first image feature, it has a remarkable effect that it is possible to distinguish non-living trees and the like, thereby generating a more accurate vegetation index.

Referring back to FIG. 3 , as another example, the verifier 313 verifies whether the extracted first image feature corresponds to the tree image feature based on the change in wavelength feature and photosynthesis feature according to the change in seasonal data. can

In the case of general trees, unlike dead trees, the color and/or the amount of photosynthesis may be partially different depending on the change of season. In some instances, the leaves of trees may be green and photosynthetic with a higher amount during spring and summer, while they may be relatively brown and photosynthetic with a relatively reduced amount during autumn and winter.

At this time, the verification unit 313, among other data collected, when the seasonal data changes, the wavelength characteristics (eg, the wavelength range of the band with the maximum reflectivity) and / or the photosynthetic characteristics (measured values in the near-infrared band) When this changes, it can be confirmed that the extracted first image feature is a tree image feature.

By verifying the extracted first image feature through a change in wavelength feature and / or photosynthetic feature, it is possible to verify the first image feature using complex feature information, thereby generating a more accurate vegetation index. have a remarkable effect.

The area determiner 314 may determine the area occupied by the first image feature based on the vegetation area data.

For some examples, the area determiner 314 may determine an area of interest (AoI) based on the vegetation area data, and determine an area occupied by the first image feature in the determined area of interest. This will be described in detail with reference to FIG. 5A.

5A is a diagram for explaining a process of determining a region of interest for region A according to some embodiments.

5A shows regions of interest 501 to 510 determined for region A included in vegetation region data.

The area determining unit 314 extracts a plurality of first image features from the satellite image data corresponding to the coordinates of the vegetation area data, and derives a contour line of a set of extracted first image features to determine the region of interest 501 to 510. can decide That is, the area determining unit 314 derives a plurality of contours so that adjacent first image features are included in the same inner region, and determines the inner regions of the plurality of derived contours as the regions of interest 501 to 510, thereby determining region A. can be subdivided in detail.

Through this process of determining the region of interest, it is possible to generate a vegetation index for more detailed coordinates (region of interest), and thus has a remarkable effect that a more accurate vegetation index can be generated.

For some examples, the area determiner 314 may determine the ROI to include all areas inside the two contour lines when the distance between the contour lines of the set of first image features is equal to or less than a predetermined distance. This will be described in detail with reference to FIG. 5B.

5B is a diagram for describing a process of determining a region of interest and a process of determining an area occupied by a first image feature, according to some embodiments. FIG. 5B shows a result of enlarging the regions of interest 508 and 509 of FIG. 5A.

Referring to FIG. 5B , it can be seen that the distance between the region of interest 508 and the region of interest 509 is rather close. For some examples, the area determiner 314 may determine the ROI to include all areas inside the two contour lines when the distance between the contour lines of the set of first image features is equal to or less than a predetermined distance.

As some examples, as shown in FIG. 5B , a part of the contour L2 of another region of interest 509 is included within a predetermined distance r from an arbitrary point P of the contour L1 of the region of interest 508. In this case, the area determining unit 314 may integrate the area 508 and the area 509 so as to include both inner areas of the two contour lines L1 and L2 and determine one area of interest. At this time, the area determination unit 314 connects the two regions of interest 508 and 509 by generating virtual lines VL1 and VL2 connecting arbitrary points of the two contour lines L1 and L2 to determine the two regions of interest ( 508, 509) can be incorporated.

When the ROI is determined, the area determining unit 314 may determine an area occupied by the first image feature in the ROI.

Forest areas often contain noise (eg, roads, waterways, rocks, etc.) and, as described above, due to the process of integrating the two areas 508 and 509 to determine one area of interest, the area of interest Noise in the inner region of each contour line L1 or L2 and/or noise between the two contour lines L1 or L2 may be partially included. Such noise needs to be excluded in the vegetation index generation process.

For convenience of explanation, a case in which noise between the two contour lines L1 and L2 is included in the process of integrating the two regions 508 and 509 and determining one region of interest will be described. FIG. 5B shows some examples of image features W for a noise channel between two contours L1 and L2.

The area determiner 314 subtracts the area occupied by the image feature (W) associated with the noise within the ROI from the area of the ROI (the ROI in which 508 and 509 are integrated), and then subtracts the first image within the ROI. The occupied area of the feature can be determined. At this time, the image feature related to the noise may be extracted by the second image feature extractor 312 as described above.

In the case of satellite images, due to limitations in spatial resolution, it may not be possible to clearly grasp the boundaries of the area where trees exist. At this time, as shown in FIG. 5B, when it is ambiguous to set the boundary of the area where the tree exists, after integrating the two regions of interest and determining one region of interest, image features corresponding to noise are detected and subtracted through a method. It has a remarkable effect that the efficiency and accuracy of data calculation can be improved.

Referring back to FIG. 3 , the wavelength feature extractor 320 may extract wavelength features associated with vegetation from visible band data among collected satellite image data.

For some examples, the wavelength characteristics may be calculated based on first wavelength data for a first wavelength among visible band data and second wavelength data for a second wavelength having a relatively shorter wavelength than the first wavelength.

For example, the wavelength feature extractor 330 may calculate the wavelength feature using a formula such as Equation 1 below.

<Equation 1>

Here, WF means a wavelength characteristic associated with vegetation, and Q1 and Q2 are data for a first wavelength and data for a second wavelength, respectively, among visible band data, that is, measured reflectivity of the first and second wavelengths can mean

For some examples, bands (wavelength ranges) of the first wavelength and the second wavelength may be previously determined among visible light bands. For some examples, the bands of the first wavelength and the second wavelength may be set in advance by a learning process of the learning module (FIG. 2, 240).

In general, plants can reflect part of the visible band (wavelength range) and absorb part. In some instances, plants may reflect a first wavelength and absorb a second wavelength. At this time, when a vegetation index is generated, plants existing in an area where a vegetation index is generated may be burned due to a fire. When the vegetation is burned, it can be predicted that reflectance and/or absorption of the first wavelength and the second wavelength will be different in the corresponding area compared to before the vegetation index is generated.

Using this principle, the learning module 240 compares data before and after the occurrence of the vegetation index to determine a band of the first wavelength and the second wavelength that causes a difference in reflectance and/or absorbance of predetermined coordinates.

As such, by using the wavelength features extracted using visible band data, it is possible to have a remarkable effect of extracting features related to vegetation and/or generating a vegetation index using only visible band data.

The photosynthetic feature extractor 330 may extract photosynthetic features associated with vegetation from near-infrared band data among collected satellite image data.

When vegetation photosynthesizes, the reflectance in the near-infrared band is higher than when photosynthesis does not occur. The photosynthetic feature extractor 330 may extract photosynthetic features including the number/degree of photosynthesis from NIR band data based on a predetermined model and/or a learned model using these properties.

6 is a flowchart of a method for generating a vegetation index according to some embodiments. Each step of FIG. 6 may be performed by the vegetation index generator 120 of FIGS. 1 and 2 . Hereinafter, redundant contents are omitted and briefly described.

First, satellite image data may be collected (S110).

In this case, the satellite image data may include data of various wavelength bands (bands) as described above. For example, satellite image data includes ultraviolet (UV) band data, visible band data, near infrared band data, infrared band data, specific components (eg, carbon dioxide, carbon monoxide, etc.) ) may include data of a band sensitive to .

Subsequently, first image features associated with vegetation are extracted from the collected satellite image data (S120), wavelength features associated with vegetation are extracted from visible band data among the collected satellite image data (S130), and collected satellite image data Photosynthetic characteristics associated with the vegetation index may be extracted from the data of the mid-near infrared band (S140).

For some examples, image features associated with vegetation may be extracted from satellite image data using an object recognition model including an algorithm for recognizing an object from an image. A detailed explanation is omitted.

For some examples, the wavelength characteristics may be calculated based on first wavelength data for a first wavelength among visible band data and second wavelength data for a second wavelength having a relatively shorter wavelength than the first wavelength. A detailed explanation is omitted.

Subsequently, a vegetation index may be generated based on the extracted first image feature, temperature feature, and wavelength feature (S150).

For some examples, a vegetation index may be generated by inputting a first image feature, an area occupied by the first image feature, a wavelength feature, a photosynthetic feature, and the like into a deep learning model to be learned. A detailed explanation is omitted.

7 is a flowchart of a method for generating a vegetation index according to some other embodiments. Each step of FIG. 7 may be performed by the vegetation index generator 120 of FIGS. 1 and 2 .

First, satellite image data may be collected (S110). Step S110 has been described above with reference to FIG. 6, and is omitted here.

Subsequently, other data may be collected (S111). Other data include, for example, meteorological data (e.g. climate data, weather data, precipitation data), wind speed data, cloud status data, typhoon data, tidal wave data, fire data, seasonal data, seasonal vegetation status data, vegetation area data and the like. Vegetation area data may include, for example, coordinates in which a forest area is formed over a predetermined range. A detailed explanation is omitted.

Subsequently, first image features associated with vegetation are extracted from the collected satellite image data (S120), wavelength features associated with vegetation are extracted from visible band data among the collected satellite image data (S130), and collected satellite image data Photosynthetic characteristics associated with the vegetation index may be extracted from the data of the mid-near infrared band (S140).

Steps S120 to S140 have been described above with reference to FIG. 6 and will be omitted here.

Subsequently, it may be verified whether the extracted first image feature corresponds to an image feature associated with vegetation (S121).

For some examples, when the wind speed is equal to or greater than a predetermined speed, whether the extracted first image feature corresponds to the tree image feature may be verified based on the second-unit change of the extracted first image feature. Skip the detailed explanation.

As another example, it may be verified whether the extracted first image feature corresponds to a tree image feature based on a change in features according to a change in seasonal data. A detailed explanation is omitted.

Subsequently, a vegetation index may be generated based on the extracted first image feature, temperature feature, and wavelength feature (S150). Step S150 has been described above with reference to FIG. 6, and is omitted here.

The above description is merely an example of the technical idea of the present embodiment, and various modifications and variations can be made to those skilled in the art without departing from the essential characteristics of the present embodiment. Therefore, the present embodiments are not intended to limit the technical idea of the present embodiment, but to explain, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of this embodiment should be construed according to the claims below, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of rights of this embodiment.

Claims (10)

A data collection module for collecting satellite image data;
a feature extraction module for extracting a first image feature related to vegetation, a wavelength feature of a visible band, and a photosynthesis feature of a near infrared band from the satellite image data; and
A determination module for generating a vegetation index based on the extracted first image feature, wavelength feature, and photosynthesis feature,
The wavelength characteristic is calculated based on first wavelength data for a first wavelength among data of the visible band and second wavelength data for a second wavelength having a relatively shorter wavelength than the first wavelength,
The data collection module further collects wind speed data as other data;
The feature extraction module,
determining whether the collected wind speed data is equal to or greater than a predetermined speed;
When the collected wind speed data is equal to or greater than a predetermined speed, determining whether there is a change in the second unit of the extracted first image feature;
Vegetation index generation device for verifying that the extracted first image feature corresponds to a first image feature associated with vegetation when a change in seconds of the extracted first image feature exists.
delete According to claim 1,
The data collection module,
Vegetation index generator for further collecting seasonal data as the other data.
According to claim 3,
The feature extraction module,
Vegetation index generator for verifying whether the extracted first image feature corresponds to a first image feature associated with vegetation, based on a shape change of the extracted first image feature according to a change in the seasonal data.
According to claim 3,
The feature extraction module,
Vegetation index generating device for verifying whether the extracted first image feature corresponds to a first image feature associated with vegetation based on a change in the extracted wavelength feature and photosynthesis feature according to a change in the seasonal data.
According to claim 1,
The data collection module,
Further collecting vegetation area data as the other data,
The feature extraction module,
Determining an area of interest (AoI) based on the vegetation area data;
Vegetation index generator for determining an area occupied by the first image feature in the determined region of interest.
According to claim 6,
The feature extraction module,
Extracting a plurality of first image features from satellite image data corresponding to the coordinates of the vegetation area data;
Derive an outline of the set of extracted first image features;
Vegetation index generating device for determining the region of interest based on the derived contour.
According to claim 6,
The feature extraction module,
Extracting a second image feature associated with noise from the satellite image data;
Vegetation index generator for determining the area occupied by the first image feature by subtracting the area occupied by the second image feature from the determined area of the region of interest.
According to claim 1,
Further comprising a learning module for learning a deep learning model using training data including at least one of the first image feature, the wavelength feature, and the photosynthesis feature,
The decision module,
Vegetation index generating device for generating the vegetation index by inputting at least one of the first image feature, wavelength feature, and photosynthesis feature to the deep learning model.
Collecting satellite image data;
a feature extraction step of extracting a first image feature related to vegetation, a wavelength feature of a visible band, and a photosynthesis feature of a near infrared band from the satellite image data; and
Generating a vegetation index based on the extracted first image feature, wavelength feature, and photosynthesis feature,
The wavelength characteristic is calculated based on first wavelength data for a first wavelength among data of the visible band and second wavelength data for a second wavelength having a relatively shorter wavelength than the first wavelength,
Collecting the satellite image data further collects wind speed data as other data,
In the feature extraction step,
determining whether the collected wind speed data is equal to or greater than a predetermined speed;
When the collected wind speed data is equal to or greater than a predetermined speed, checking whether there is a change in the second unit of the extracted first image feature;
and verifying that the extracted first image feature corresponds to a first image feature associated with vegetation when there is a change in seconds of the extracted first image feature.

Images