KR102336196B1 - A Method for Extracting of Ionospheric Echoes from Oblique Ionograms Observed by VIPIR - Google Patents

Abstract

The present invention relates to a method of extracting a high-resolution ionosphere oblique incident signal to extract a predetermined layer structure of the ionosphere from ionosphere image information of an ionogram. More specifically, a processor executes: a noise reduction step (a) of removing a noise from an ionosphere image according to a frequency and altitude by using image processing to generate a noise reduction image; a segmentation step (b) of generating a segmented image in which a segment is divided into at least one object through grouping of adjacent pixels in the noise reduction image; an ionization layer extraction step (c) of extracting images of an F layer and E layer of the ionosphere from the segment image to separate and generate the images of the F layer and the E layer; and a tracing step (d) of calculating a N^th degree equation corresponding to an F layer structure in the image of the F-layer.

Description

{A Method for Extracting of Ionospheric Echoes from Oblique Ionograms Observed by VIPIR}

The present invention relates to a method for extracting a high-resolution ionospheric incident signal, and more particularly, to a method for extracting a high-resolution ionosphere signal from an ionogram image of the ionosphere calculated from VIPIR (Vertical Incidence Pulsed Ionospheric Radar). .

The ionosphere refers to the atmospheric layer formed at a height of 50 to 600 km above the ground by ionizing the atmospheric layer surrounding the earth by energy radiated from the sun. The ionosphere constantly changes state according to season, time and topography.

In this ionosphere, what can be utilized for radio communication is the D layer, which exists at an altitude of 50 to 100 km, and the E and F layers, which are formed at a height of 100 to 400 km. These layers E and F are shown in FIG. 10 . The influence of the ionosphere on communication is due to the electron density forming the layer, and the density determines the frequency of radio waves used by reflecting the ionosphere. Therefore, the main task of the study of the ionosphere was to observe and analyze the distribution of electron density in the ionosphere and the state of change with time to enable optimal communication.

In general, the shortwave band transmission wave used for ionospheric observation is reflected from the transmission/reception midpoint and arrives at the receiving point. Among them, VIPIR (Vertical Incidence Pulsed Ionospheric Radar), a system that introduces ionospheric incidence technology, has a much higher resolution than the existing Inonsonde system. In addition, while the existing ionosonde system is installed at a fixed point and has a limitation in the installation location as it is directly incident from the sky of the point, VIPIR is installed not only at direct incidence but also at two spaced points, so that it It is also possible to use the oblique angle method to obtain information about the ionosphere, so the installation location can be adjusted according to the observation point, so it is attracting attention as a next-generation ionosphere observation technology.

However, VIPIR can grasp the shape of the oblique and direct incidence from the received data, but has a disadvantage in that noise is severe as the resolution is high.

In addition, since a radio wave transmission point and a reception point are required due to the characteristics of VIPIR, the range of values of the observation data varies depending on the location of the point, and thus there is also a problem in that uncertain observation data is produced.

To solve this, a VIPIR-based dead-incidence tracing technique is required, and a system is needed to derive parameters necessary for ionosphere analysis by matching the N-order equation corresponding to the structure of each layer from the ionosphere observation image.

US Patent Publication No. 2013-0050024 (“BISTATIC RADAR SYSTEM USING SATELLITE-BASED TRANSMITTERS WITH IONOSPHERIC COMPENSATION”, 2012.08.24)

The present invention has been devised to solve the above problems, and through a method of removing noise and accurately extracting only a part corresponding to the signal from the ionogram image of the ionosphere calculated from VIPIR to detect the oblique incident signal, a stable and reliable ionosphere Its purpose is to provide observational data.

In order to solve the above problems, the present invention relates to a method of extracting a specific layer structure of an ionosphere from ionosphere image information, more specifically, a processor, (a) from an ionosphere image according to frequency and altitude, noise using image processing A noise reduction step of generating a noise reduction image by removing (c) extracting the F-layer and E-layer images of the ionosphere from the segment image, respectively, to generate an F-layer image and an E-layer image separately (Extracting), and (d) from the F-layer image to the F-layer structure It may include a tracing step of calculating a corresponding N-order equation.

The step (a) includes: (a1) removing a non-receivable area to generate a first image in which a predetermined area is removed from the ionospheric image; A device noise removal step of generating (a3) a low-output signal removal step of generating a third image by removing a signal of a point whose output is less than or equal to a predetermined value from the second image, (a4) removing the point noise from the third image It may include a point noise removal step of generating a fourth image, and (a5) a vertical noise removal step of generating a noise reduction image by removing vertical noise from the fourth image.

The step (a) includes the steps of: generating an image from which a predetermined area is removed; generating an image from which noise of a predetermined characteristic is removed; The method may include at least one of generating an image from which ? is removed, and generating an image from which vertical noise is removed.

The step (b) includes (b1) a first segmentation step of grouping the points constituting the noise reduction image into segments of at least one object by grouping them in a predetermined size, (b2) dividing the points constituting the noise reduction image in step (b1) The method may include an adjacency deriving step of deriving inter-segment adjacency, and (b3) a secondary segmentation step of merging adjacent segments and generating a segment image by adjusting the group of the plurality of segments based on the derived adjacency.

Between the steps (a) and (b), a masking step of applying a mask in which an effective ionospheric signal pattern is formed from the noise-reduced image, and between the steps (b) and (c), forming a segment A pruning step of removing a value in which the signal strength of each point is lower than a predetermined value may be included.

The step (c) includes (c1) a segment calculation step of calculating the signal strength of each segment, (c2) a segment in which the signal strength of each segment is less than or equal to a predetermined value based on the signal strength calculated in the step (c1) Segment adjustment step of removing as noise, (c3) segment sorting step of arranging each segment in the order of the magnitude of the signal strength calculated in step (c1), (c4) the segment aligned in step (c3) F-layer extraction step of extracting the F-layer image based on the sequence, (c5) determining whether the F-layer and E-layer overlap in each segment of the F-layer image extracted in the step (c4), and extracting the E-layer image It may include an E-layer extraction step.

The step (c1) includes (c1-1) an average signal strength calculation step of calculating the average signal strength of each segment, (c1-2) an N-sigma average signal strength calculation step of calculating an N-sigma average signal strength, (c1) -3) a maximum signal strength calculation step of calculating the maximum signal strength of a segment, and (c1-4) a segment size calculation step of calculating a segment size may be included.

The step (c2) includes (c2-1) a first segment removal step of removing a segment having a size of less than or equal to a predetermined value, and (c2-2) a case in which the maximum signal strength of each segment is less than or equal to a predetermined value. It may include a second segment removing step of removing a segment, and (c2-3) a third segment removing step of removing a segment having an average signal strength or N sigma average signal strength of each segment equal to or less than a predetermined value.

In step (c4), the segment having the largest average signal strength among each segment may be regarded as the F layer.

The step (c5) includes (c5-1) a histogram generation step of generating a histogram for the elevation of each segment in the F-layer image extracted in the step (c4), (c5-2) the step (c5-1) In the histogram generated in , a valley determination step of determining an altitude of a segment having a number of points in the histogram with respect to an altitude equal to or less than a predetermined value as a valley, and (c5-3) in the step (c5-2) According to the determined existence of the valley, the E-layer separation generating step of generating the E-layer image by separating it may be included.

In step (c5-3), when it is determined that a valley exists in step (c5-2), a segment having a high elevation value is regarded as an F floor based on the elevation value of the segment determined as a valley, and the elevation value is A relatively low segment is generated by separating the E-layer image, and when it is determined that there is no valley, the segment corresponding to the F-layer is lower in height and closest to the E-layer image by separating it and generated.

The step (d) is a root extraction step of the equation, (d1) extracting only the points having the average signal intensity of each segment higher than the surrounding value from the F-layer image, (d1-1) the points constituting the segment A weighting step of assigning weights to the points according to a preset range of the values, and (d2) a polynomial modeling step of calculating an Nth-order equation based on the points weighted in step (d1-1). can do.

The step (d2) may be modeled by Equation 1 based on points corresponding to the F-layer signal.

Formula 1: y = ax ² + bx + c (y = frequency, x = altitude)

The step (d) further includes, after the step (d2), (d3) the E-layer modeling step of modeling the elevation of the E-layer in the E-layer image by Equation 2, wherein the step (d3) is the E-layer The elevation corresponding to the segment having the maximum signal strength in the image can be set as the elevation of the E layer.

Equation 2: x = b (x = altitude)

Prior to the step (a), the processor may further include (a0) receiving ionospheric image information from a VIPIR (Vertical Incidence Pulsed Ionospheric Radar).

The present invention according to the above configuration relates to a method for detecting a high-resolution oblique signal from an ionogram image of an ionosphere calculated from VIPIR, and a signal transformed through noise removal and oblique signal masking in a noisy image. After restoring the ionosphere, it is possible to accurately observe and analyze the atmosphere by deriving the ionosphere parameters necessary to analyze the ionospheric state by accurately extracting only the desired dead incident signal by tracing.

1 is a flowchart of a method for extracting a high-resolution ionosphere oblique incident signal of the present invention.
2 shows a flow chart of the noise removal step of the present invention.
3 shows a flowchart of the segmentation step of the present invention.
Figure 4 shows a flow chart of the ionospheric image separation step of the present invention.
5 is a flowchart of another embodiment of the high-resolution ionospheric incident signal extraction method of the present invention.
6 shows a flowchart of the segment calculation step of the present invention.
7 shows a flow chart of the segment adjustment step of the present invention.
Figure 8 shows the E-layer extraction step of the present invention.
9 shows the tracing step of the present invention.
10 shows the shape of a typical ionospheric signal (layer E and layer F).
11 shows an ionosphere image of the denoising step of the present invention.
12 shows an image of the masking step of the present invention.
13 shows an image of the segmentation step of the present invention.
14 shows an image of the tracing step of the present invention.

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and detailed description will be given. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present invention.

When a component is referred to as being “connected” or “connected” to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in between. it should be

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

Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

Hereinafter, the technical idea of the present invention will be described in more detail with reference to the accompanying drawings.

Since the accompanying drawings are only examples shown to explain the technical idea of the present invention in more detail, the technical idea of the present invention is not limited to the form of the accompanying drawings.

The present invention discloses a method of extracting a high-resolution ionospheric signal, more preferably an F-layer signal, from an ionogram image received from VIPIR (Vertical Incidence Pulsed Ionospheric Radar), an ionosphere observation system capable of direct and oblique incidence methods. . This will be described in more detail with reference to FIGS. 1 to 14 .

In the method for extracting the specific layer structure of the ionosphere from the ionosphere image information, as shown in FIG. 1 , the processor (a) removes noise from the ionosphere image according to frequency and altitude using image processing to obtain a noise reduction image A noise reduction step (S100) to generate, (b) a segmentaion step (S200) and , (c) extracting the F-layer and E-layer images of the ionosphere from the segment image, respectively, to generate an F-layer image and an E-layer image separately (S300), and (d) the F-layer from the F-layer image It may be composed of a tracing step (S400) of calculating an N-order equation corresponding to the structure.

In addition, before the step (a), as shown in FIG. 5, the processor may further include a step (S50) of receiving ionospheric image information from a VIPIR (Vertical Incidence Pulsed Ionospheric Radar). VIPIR may consist of a transmitter that transmits a radar signal in any one of direct-incident and oblique-incident to the ionosphere, and a receiver that receives a signal reflected from the ionosphere. VIPIR may image the received patent data to generate patent image information, and transmit it to the processor. The ionosphere image information may be an ionogram image. Such an image may be composed of a set of points, which are the minimum units constituting the image, and the point may be any one of a pixel and a dot, and more preferably a pixel.

Next, the step (a) will be described in detail. The step (a), as shown in FIG. 2 , the step (a) includes: (a1) removing the unreceivable area of generating a first image in which a predetermined area is removed from the ionosphere image (S110), (a2) ) A device noise removal step of generating a second image by removing noise of a predetermined characteristic from the first image (S120), (a3) removing a signal of a point whose output is less than or equal to a predetermined value from the second image to remove the third image A low-output signal removal step (S130) for generating It may include a vertical noise removal step (S150) of generating a noise reduction image by removing the noise.

Meanwhile, FIG. 11-(a) may be a second image, FIG. 11-(b) may be a third image, FIG. 11-(c) may be a fourth image, and FIG. 11-(d) may be a fifth image. .

Here, the predetermined area of step (a1) may be a section in which reception of the patent right signal is impossible. Accordingly, the first image may be an image from which the reception impossibility section of the patent right signal has been removed. In addition, the noise of the predetermined characteristic in step (a2) may be noise generated due to the characteristics of the VIPIR device. Accordingly, the second image may be an image from which device noise is removed, as shown in FIG. 11-( a ). In addition, the output below a predetermined level of step (a3) may preferably be an output of 75% or less of the average output. Since the output of less than 75% of the average output is more likely to be a noise rather than a dead-incident signal empirically, signals with an output of less than 75% of the average output can be removed. Accordingly, the third image may be an image from which the low output signal is removed, as shown in FIG. 11-(b) . Also, as shown in FIG. 11-(c), the fourth image of step (a4) may be an image from which point noise is removed. In addition, the ionosphere image information received from VIPIR has a lot of vertical noise in the vertical direction. Accordingly, vertical noise is removed in step (a5), and the fifth image may be an image from which vertical noise is removed, as shown in FIG. 11-(d).

Also, according to the present invention, the step (a) may include at least one of steps (a1) to (a6), and the order may not be limited. That is, the step (a) includes the steps of generating an image from which a predetermined area is removed, generating an image from which noise of a predetermined characteristic has been removed, and generating an image in which signals of pixels having an output equal to or less than a predetermined output are removed It may include at least one of the following steps: generating an image from which salt and pepper noise is removed, and generating an image from which vertical noise is removed. A noise reduction image may be generated through at least one of the above steps.

Here, the predetermined area may be a reception unavailable section, and the noise of a predetermined characteristic may be noise generated due to the characteristics of the device. In addition, the output of the predetermined or less may preferably be an output of 75% or less of the average output. As a rule of thumb, a signal with an output of less than 75% of its average output is considered invalid.

On the other hand, in the ionosphere image generated by VIPIR, when the noise removal step (S100) is performed, as shown in FIG. 11-d, a signal in the form of a parabola lying down is derived. In this case, the original ionospheric signal may also be removed in the noise removal step ( S100 ), and the signal shape may be discontinued and not continuous. In this case, if the parameters for the analysis of the patent right are extracted from the signal of the broken form, an error occurs, and it is difficult to analyze the patent right. In order to solve this problem, according to the present invention, a segmentation step (S200) of generating a segment image in which a segment is divided into at least one object through grouping of adjacent points in a noise-reduced image from which noise is removed is performed. Through the segmentation step S200, the signal is discretized, and the signal component included in each segment is expressed as an average signal strength value, so that the derivation of the parameter value may not be limited by some cut off signals.

Specifically, the step (b) is, as shown in FIG. 3, (b1) a primary segmentation step (S210) of grouping the points constituting the noise reduction image to a predetermined size and dividing the points into segments of at least one object (S210). ), (b2) an adjacency deriving step (S220) of deriving the adjacency between segments, and (b3) a second segmentation step of generating a segment image by adjusting the group of the plurality of segments based on the derived adjacency ( S230) may be included. 13-(a) shows a first segmentation step, and FIG. 13-(b) shows a second segmentation step.

Specifically, in the step (b1), that is, the first segmentation step (S210), each of the points constituting the noise reduction image is adjacent to each other by using an adjacent determination function having a predetermined filter size in the noise reduction image. It may be dividing into segments of at least one entity by assigning a preset index to a point of , and grouping at least one point having the same index. The adjacency determination function may assign a preset index to each point according to whether a point constituting an image is adjacent. For example, suppose we have a set of points {p1, p2, p3, p4, p5}. Assume that the adjacency determination function determines whether each point is adjacent to each other with a predetermined filter size, assigns a preset index accordingly, and assigns an index to p1 and p2 as 0x0001, p2 and p3 as 0x0002, and p5 as 0x0003. . In this case, p1 and p2 form one segment, p3 and p4 form one segment, and p5 forms one segment, so that it can be divided into a total of three segments.

After the step (b1), the step (b2), that is, the adjacency deriving step (S220), derives the adjacency between the segments divided in the step (b1), and in the step (b2), that is, the second segmentation step (S230) A segment image may be generated by removing a segment having an adjacency of a predetermined size or less based on the derived adjacency, and merging the removed segment with a neighboring segment.

In addition, between the steps (a) and (b), as shown in FIG. 5 , a masking step (S160) of applying a mask having an effective ionospheric signal pattern formed from the noise reduction image may be further included. In addition, between the steps (b) and (c), a pruning step (S250) of removing a signal strength of each of the points constituting the segment is a predetermined value, preferably a value lower than 5 dB may be included.

Fig. 12-(a) shows the mask on which the effective ionospheric signal pattern is formed, and Fig. 12-(b) shows the application of the mask on which the ionospheric signal pattern is formed. Here, the mask may be a criterion for determining whether the received signal is a valid signal or noise before performing the step (b).

With this masking step ( S160 ), it is possible to restore a signal polluted during the noise removal step, and there is an effect of more clearly acquiring an image of a signal shape corresponding to the F-layer and E-layer signal components.

In addition, after the step (b), the predetermined value in the pruning step (S250) may preferably be 5 dB. Empirically, if the signal strength of the point constituting the segment generated in step (b) is 5 dB or less, it is regarded as a meaningless signal, so it is preferable to remove it.

According to the present invention, after the segmentation step (S200), the F-layer and E-layer images are extracted from the signal divided into each segment, and the step (c) of separately generating each, that is, the ionospheric extraction step (S300) is included. To explain this in detail, as shown in FIG. 4 , in the step (c), (c1) calculating the signal strength of each segment (S310), (c2) in the step (c1) Based on the calculated signal strength, segment adjustment step (S320) of removing a segment having a signal strength of each segment below a predetermined level as noise is removed (S320), (c3), respectively, in the order of the magnitude of the signal strength calculated in step (c1) Segment sorting step (S330), (c4) for arranging the segments of the F-layer extraction step (S340), (c5) for extracting the F-layer image based on the segment order sorted in the step (c3) (S340), (c5) The step (c4) It may include an E-layer extraction step (S350) of determining whether the F-layer and the E-layer overlap in each segment of the extracted F-layer image, and extracting the E-layer image.

Steps (c1) to (c3) are steps of adjusting the segments by checking whether each segment has a significant signal strength, removing meaningless signals, and aligning the segments. This will be described in detail below with reference to FIG. 6 .

First, as shown in FIG. 6 , the steps (c1) include (c1-1) calculating the average signal strength of each segment (S311), (c1-2) N-sigma average signal strength N-sigma average signal strength calculation step (S312) for calculating , (c1-3) maximum signal strength calculation step for calculating the maximum signal strength of the segment (S313), and (c1-4) segment size for calculating the segment size It may include a calculation step (S314). Preferably, the order of steps (c1-1) to (c1-4) may not be limited.

Here, N sigma may be a sigma confidence interval. 1 sigma is 68%, 2 sigma is 90%, 3 sigma is 99%, and the average signal strength is given by taking the sigma value. where N is a natural number. N may be arbitrarily selecting a number from which an optimal value can be derived for each image.

The step (c2) is a step of removing insignificant segments having insignificant signal strength based on the signal strength calculated in the step (c1). This will be described in detail with reference to FIG. 7 .

The step (c2) includes, as shown in FIG. 7 , (c2-1) a first segment removal step (S321) of removing segments having a size of each segment equal to or less than a predetermined value; (c2-2) a second segment removal step (S322) of removing a segment having a maximum signal strength of each segment equal to or less than a predetermined value; and (c2-3) a third segment removal step (S323) of removing a segment having an average signal strength or N sigma average signal strength of each segment equal to or less than a predetermined value.

Preferably, the steps (c2-1) to (c2-3) may not be limited in order. Also, the segment size in step (c2-1) above. Also, in step (c2-2), the predetermined value may be 10 dB. Experientially, a segment having a maximum signal strength of 10 dB or less is regarded as a segment consisting of a meaningless signal and removed, and points included in the removed segment may be merged into a neighboring segment. Also, in step (c2-3), the predetermined value may be 7.5 dB. Empirically, a segment having an average signal strength or N sigma average signal strength of 7.5 dB or less is regarded as a segment consisting of a meaningless signal and removed, and points included in the removed segment may be merged into a neighboring segment.

In addition, only one selected from the average signal strength or the N sigma average signal strength may be calculated, and a segment may be removed based on the selected one.

Meanwhile, in the step (c3), each segment may be arranged in the order of the magnitude of the average signal intensity calculated in the step (c1) or the N-sigma average signal intensity.

In step (c4), the segment having the largest average signal strength among each segment may be regarded as the F layer.

Meanwhile, in the F-layer image extracted in step (c4), a region overlapping the E-layer image may exist. It is preferable to create these overlapping regions separately. Accordingly, according to the present invention, a method for separating a region overlapping an E-layer image in an F-layer image is disclosed. This will be described in detail with reference to FIG. 8 below.

The step (c5) is, as shown in FIG. 8, (c5-1) a histogram generation step of generating a histogram for the elevation of each segment in the F-layer image extracted in the step (c4), (c5-2) ) in the histogram generated in step (c5-1), a valley determination step of determining the altitude of a segment having a number of points in the histogram with respect to an altitude equal to or less than a predetermined value as a valley, and (c5-3) According to the existence of the valley determined in the step (c5-2), the E-layer separation generation step of separating and generating the E-layer image may be included.

Preferably, in the step (c5-1), the histogram may be generated with the number of points of each segment with respect to the elevation. Also, in the step (c5-2), the predetermined value may be 40. That is, a segment having a small size in which the number of points of the histogram for altitude is 40 or less may be determined as a valley, and when there are several valleys, the segment having the smallest size may be determined as a valley. In addition, the altitude having the lowest number of points while lower than the maximum power height among the altitudes having 40 or less points of the histogram for the altitude is determined as a valley. That is, in the segment, a valley means an abrupt change in the image, and a portion where the image change is abrupt may mean that the F-layer signal and the E-layer signal are superimposed. Therefore, it is preferable to separate the F-layer image and the E-layer image when there is a valley. This will be described below.

In step (c5-3), when it is determined that a valley exists in step (c5-2), a segment having a high elevation value is regarded as an F floor based on the elevation value of the segment determined as a valley, and the elevation value is A relatively low segment is generated by separating the E-layer image, and when it is determined that there is no valley, the segment corresponding to the F-layer is lower in height and closest to the E-layer image by separating it and generated.

With this configuration, even when the segments corresponding to the E and F layers are overlapped, the E-layer and F-layer images can be reliably separated and generated.

The step (d) is, as shown in FIG. 9, the step (d) is (d1) of the equation of extracting only the points having the average signal intensity of each segment higher than the surrounding value in the F-layer image. The root extraction step, although not shown in the drawing, (d1-1) a weighting step of assigning weights to the points according to a preset range of points constituting the segment, and (d2) the step (d1-1) It may include a polynomial modeling step of calculating an Nth-order equation based on the weighted points in .

On the other hand, as shown in FIG. 14, it is an object of the present invention to provide a reliable method for extracting signals of the F-layer and E-layer in order to receive and analyze the ionospheric incident signal. In the end, it is most important to reliably extract the F-layer image. At this time, the F-layer image is usually in the shape of a parabola, and the area around the vertex of the parabola, that is, the most curved area, becomes an important parameter in the study after signal extraction. Therefore, in step (d1-1), a greater weight is given to the curved region than the less curved region so that a partial region, preferably the most curved region, can be enlarged and analyzed for easy analysis of the ionospheric oblique incident signal. can be given In this case, the size of the weight may be arbitrarily changed by the user according to the patent image information received from VIPIR.

Also, preferably, in the step (d), a method of approximating to an Nth-order equation by regression may be performed. where N is a natural number.

The step (d2) may be modeled by Equation 1 based on at least three elevations corresponding to the F-layer signal. Preferably, it can be modeled as a quadratic equation as shown in Equation 1 below.

Equation 1: y = ax ² + bx + c (y = frequency, x = altitude)

Here, the root and the number of roots modeled by Equation 1 may be arbitrarily set by the user, or at least three points may be set in order of increasing signal strength among points corresponding to the F-layer signal. Preferably, more points can be set as roots in the curved portion than in the less curved portion.

In the step (d), as shown in FIG. 9, after the step (d2), (d3) E modeling the elevations of at least two E layers corresponding to the E layer signals in the E layer image with Equation 2 The method may further include extracting the root of the layer equation, wherein step (d3) may set the elevation corresponding to the segment having the maximum signal intensity in the layer E image as the elevation of the layer E.

Equation 2: x = b (x = altitude)

14-(a) shows the F-layer image extracted in step (c) before tracing through step (d), and in FIG. 14-(b), Equation 1 (secondary) after tracing through step (d) The F-layer signal matching the equation) is shown.

With this configuration, it is possible to more clearly extract the ionospheric incident signal and match it to the equation to derive the parameters required for the ionosphere analysis.

The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the gist of the present invention as claimed in the claims.

S100: Noise Removal Step
S200: Segmentation step
S300: Ionospheric extraction step
S400: Tracing Step
S50: receiving the patent image information from VIPIR
S110: Receiving area removal step
S120: Device noise removal step
S130: Low output signal removal step
S140: point noise removal step
S150: Vertical Noise Removal Step
S210: 1st segmentation step
S220: adjacency derivation step
S230: 2nd segmentation step
S250 : Pruning step
S160: masking step
S310: segment calculation step
S320: Segment Adjustment Step
S330: Segment alignment step
S340: F layer extraction step
S350: E-layer extraction step
S311: Average signal strength calculation step
S312: N sigma average signal strength calculation step
S313: Maximum signal strength calculation step
S314: Segment size calculation step
S321: first segment removal step
S322: second segment removal step
S323: Third segment removal step
S351: Histogram generation step
S352: Valley judgment stage
S353: E-layer separation generation step
S410: Root extraction step of the equation
S420: polynomial modeling step
S430: Root extraction step of E-layer equation

Claims (15)

In the method of extracting the specific layer structure of the ionosphere from the ionosphere image information,
processor,
(a) removing noise from the ionosphere image according to frequency and altitude using image processing to generate a noise reduction image;
(b) a segmentation step of generating a segment image in which segments are divided into at least one object through grouping of adjacent points in the noise reduction image;
(c) extracting the F-layer and E-layer images of the ionosphere from the segment image, respectively, and separately generating the F-layer image and the E-layer image (Extracting); and
(d) a tracing step of calculating an Nth-order equation corresponding to the F-layer structure in the F-layer image;
High-resolution ionospheric incidence signal extraction method, characterized in that.
The method of claim 1,
The step (a) is,
(a1) an unreceivable area removal step of generating a first image in which a predetermined area is removed from the ionospheric image;
(a2) a device noise removal step of generating a second image by removing noise of a predetermined characteristic from the first image;
(a3) a low-output signal removing step of generating a third image by removing a signal of a point having an output equal to or less than a predetermined value in the second image;
(a4) removing the point noise from the third image to generate a fourth image; and
(a5) a vertical noise removal step of removing vertical noise from the fourth image to generate a noise-reduced image;
High-resolution ionospheric incidence signal extraction method, characterized in that.
The method of claim 1,
The step (a) is,
generating an image from which a predetermined area is removed;
generating an image from which noise of a predetermined characteristic has been removed;
generating an image from which a signal of a point having an output equal to or less than a predetermined value is removed;
generating an image from which point noise is removed; and
generating an image from which vertical noise has been removed; comprising at least one step of
High-resolution ionospheric incidence signal extraction method, characterized in that.
The method of claim 1,
The step (b) is,
(b1) a first segmentation step of grouping the points constituting the noise reduction image to a predetermined size and dividing it into segments of at least one object;
(b2) an adjacency deriving step of deriving adjacency between the segments divided in step (b1); and
(b3) a secondary segmentation step of merging adjacent segments and generating a segment image by adjusting the group of the plurality of segments based on the derived adjacency;
High-resolution ionospheric incidence signal extraction method, characterized in that.
The method of claim 1,
Between step (a) and step (b),
A masking step of applying a mask in which an effective ionospheric signal pattern is formed from the noise reduction image;
Between step (b) and step (c),
A pruning step of removing a value lower than a predetermined value of each signal strength of the points constituting the segment; containing
High-resolution ionospheric incidence signal extraction method, characterized in that.
The method of claim 1,
The step (c) is,
(c1) a segment calculation step of calculating the signal strength of each segment;
(c2) a segment adjustment step of removing a segment having a signal strength of each segment equal to or less than a predetermined value as noise based on the signal strength calculated in step (c1);
(c3) a segment sorting step of arranging each segment in the order of magnitude of the signal strength calculated in step (c1);
(c4) an F-layer extraction step of extracting an F-layer image based on the sequence of segments arranged in step (c3);
(c5) in each segment of the F-layer image extracted in step (c4), determining whether the F-layer and the E-layer overlap or not, and extracting the E-layer image
High-resolution ionospheric incidence signal extraction method, characterized in that.
7. The method of claim 6,
The step (c1) is,
(c1-1) an average signal strength calculation step of calculating an average signal strength of each segment;
(c1-2) an N-sigma average signal strength calculation step of calculating an N-sigma average signal strength;
(c1-3) a maximum signal strength calculation step of calculating the maximum signal strength of the segment; and
(c1-4) a segment size calculation step of calculating the size of the segment;
High-resolution ionospheric incidence signal extraction method, characterized in that.
7. The method of claim 6,
The step (c2) is,
(c2-1) a first segment removal step of removing segments whose size is less than or equal to a predetermined value; and
(c2-2) a second segment removal step of removing a segment having a maximum signal strength of each segment equal to or less than a predetermined value; and
(c2-3) a third segment removal step of removing a segment having an average signal strength or N sigma average signal strength of each segment equal to or less than a predetermined value;
High-resolution ionospheric incidence signal extraction method, characterized in that.
7. The method of claim 6,
The step (c4) is,
Considering the segment with the largest average signal strength among each segment as the F-layer
High-resolution ionospheric incidence signal extraction method, characterized in that.
7. The method of claim 6,
The step (c5) is,
(c5-1) a histogram generating step of generating a histogram for the elevation of each segment in the F-layer image extracted in step (c4);
(c5-2) in the histogram generated in step (c5-1), a valley determination step of determining an altitude of a segment in which the number of points in the histogram with respect to altitude is less than or equal to a predetermined value as a valley; and
(c5-3) according to the existence of the valley determined in the step (c5-2), the E-layer separation generation step of separating and generating the E-layer image;
High-resolution ionospheric incidence signal extraction method, characterized in that.
11. The method of claim 10,
The step (c5-3) is,
When it is determined in step (c5-2) that a valley exists, a segment with a high elevation value is regarded as an F layer based on the elevation value of the segment determined as a valley, and a segment with a relatively low elevation value is regarded as an E layer image. separate and create
When it is judged that there is no valley, the segment corresponding to the F-layer is lower than the segment corresponding to the F-layer and the closest segment is separated and created as an E-layer image.
High-resolution ionospheric incidence signal extraction method, characterized in that.
The method of claim 1,
The step (d) is,
(d1) extracting the root of the equation from the F-layer image, extracting only the points having the average signal intensity of each segment higher than the surrounding value; and
(d2) a polynomial modeling step of calculating an Nth-order equation based on the points in step (d1);
High-resolution ionospheric incidence signal extraction method, characterized in that.
13. The method of claim 12,
The step (d2) is,
Modeling with Equation 1 based on the points corresponding to the F-layer signal
High-resolution ionospheric incidence signal extraction method, characterized in that.
Formula 1: y = ax ² + bx + c (y = frequency, x = altitude)
13. The method of claim 12,
The step (d) is, after the step (d2),
(d3) E-layer modeling step of modeling the elevation of E-layer in Equation 2 in the E-layer image; further comprising,
The step (d3) is to set the altitude corresponding to the segment having the maximum signal strength in the E-layer image as the E-layer altitude
High-resolution ionospheric incidence signal extraction method, characterized in that.
Equation 2: x = b (x = altitude)
The method of claim 1,
Prior to step (a),
The processor is
(a0) receiving the ionosphere image information from VIPIR (Vertical Incidence Pulsed Ionospheric Radar); further comprising
High-resolution ionospheric incidence signal extraction method, characterized in that.

Images