KR102217510B1 - Image visualization apparatus using ultra wide band ground penetrating radar - Google Patents

Abstract

A method according to an embodiment of the present disclosure may include: a method for visualizing an image based on an ultra wide band ground penetrating radar, comprising: obtaining data using the ultra wide band ground penetrating radar; Outputting a first value by applying moving average filtering and gain compensation to the acquired data; Detecting an edge of an image by applying a differential operator to the first value; And correcting the image based on the detected edge.

Description

Ultra-wide band ground penetration radar based image visualization device {IMAGE VISUALIZATION APPARATUS USING ULTRA WIDE BAND GROUND PENETRATING RADAR}

The present disclosure relates to a method and apparatus for visualizing an image based on an ultra-wideband ground-transmitting radar.

The GPR (Ground Penetrating Radar) system is an ultra-wideband radar system for imaging a target. In general, a target signal in a parabolic shape is visualized by scanning an antenna. At this time, the target signal is a ringing phenomenon caused by multiple reflections, and the propagation is weakened by the phenomenon that several parabolas are generated downwards, the dielectric constant of the soil, and moisture. Therefore, measures are required to compensate for this.

The disadvantage is that it is not robust to signals other than the target (surrounding environment and system structural noise) such as unevenness of the ground, vibration of the sensor and noise signals. Therefore, there is a problem that the probability of generating a false alarm in which the background signal is recognized as a mine signal and the probability of not detecting an actual buried mine by recognizing the mine signal as a background signal are high.

In addition, the GPR system must be able to quickly and accurately detect landmine signals. Therefore, there is a need for a method for detecting and visualizing landmines, which has a high computational speed and exhibits robust performance for signals other than mines.

The present disclosure provides a method and apparatus capable of having a high computational speed and exhibiting robust performance with respect to signals other than landmines.

The present disclosure provides a method and apparatus capable of being robust to non-target signals, such as ground irregularities, sensor shake and noise signals.

A method according to an embodiment of the present disclosure may include: a method for visualizing an image based on an ultra wide band ground penetrating radar, comprising: obtaining data using the ultra wide band ground penetrating radar; Outputting a first value by applying moving average filtering and gain compensation to the acquired data; Detecting an edge of an image by applying a differential operator to the first value; And correcting the image based on the detected edge.

An apparatus according to an embodiment of the present disclosure is an image visualization apparatus based on an Ultra Wide Band Ground Penetrating Radar, comprising: a transmitter/receiver for transmitting and receiving data; And obtaining the data using the ultra-wide band ground-transmitting radar, outputting a first value by applying a moving average filtering and gain compensation to the obtained data, and applying a differential operator to the first value And a controller for detecting (Edge) and correcting the image based on the detected edge.

A method according to an embodiment of the present disclosure may include: a method for visualizing an image based on an ultra wide band ground penetrating radar, comprising: obtaining data using the ultra wide band ground penetrating radar; Outputting a first value by applying moving average filtering and gain compensation to the acquired data; Detecting an edge of the image by applying a local distance of the image to the first value; And correcting the image based on the detected edge.

An apparatus according to an embodiment of the present disclosure is an image visualization apparatus based on an Ultra Wide Band Ground Penetrating Radar, comprising: a transmitter/receiver for transmitting and receiving data; And obtaining the data using the ultra-wideband ground-transmitting radar, outputting a first value by applying a moving average filtering and gain compensation to the obtained data, and applying a local distance of the image to the first value. And a control unit for detecting (Edge) and correcting the image based on the detected edge.

A method according to an embodiment of the present disclosure may include: a method for visualizing an image based on an ultra wide band ground penetrating radar, comprising: obtaining data using the ultra wide band ground penetrating radar; Outputting a first value by applying moving average filtering and gain compensation to the acquired data; Detecting an edge of the image based on the first value and a feature point of a parabola related to the image; And correcting the image based on the detected edge.

An apparatus according to an embodiment of the present disclosure is an image visualization apparatus based on an Ultra Wide Band Ground Penetrating Radar, comprising: a transmitter/receiver for transmitting and receiving data; And obtaining the data using the ultra-wideband ground-transmitting radar, applying a moving average filtering and gain compensation to the obtained data to output a first value, and based on the first value a feature point of a parabola related to an image. And a controller configured to detect an edge of the image and correct the image based on the detected edge.

The present disclosure has a high computational speed and can exhibit robust performance with respect to signals other than landmines.

The present disclosure can be robust to signals other than the target, such as ground irregularities, sensor shake and noise signals.

The present disclosure can effectively provide a user with a buried location as well as the presence or absence of a landmine.

1 is a flow chart showing a general image visualization method;
2 is a block diagram of a GPR system;
3 is an exemplary view of data acquired by using a ground-transmitting radar after buriing a mine (object) in the ground;
4 is a block diagram of a GPR system according to an embodiment of the present disclosure;
5 is a flowchart illustrating a method of visualizing an image based on an ultra-wideband ground-transmitting radar according to an embodiment of the present disclosure;
6 is a diagram illustrating an example of a super-band ground-transmitting radar-based image according to the first embodiment of the present disclosure;
7 is a diagram illustrating a result of edge detection for each mask according to the first embodiment of the present disclosure;
8 is a diagram illustrating a visualization screen before/after an edge according to the first embodiment of the present disclosure;
9 is a diagram showing a super-band ground-transmitting radar-based image according to a second embodiment of the present disclosure;
10 is a diagram illustrating a super-band ground-transmitting radar-based image according to a third embodiment of the present disclosure;
11 is a diagram showing a super-band ground-transmitting radar-based image to which a FAST method according to a third embodiment of the present disclosure is applied;
12 is a diagram showing an image to which an ultra-wideband ground-transmitting radar-based image visualization method is applied according to an embodiment of the present disclosure;
13 is a flowchart illustrating a method for visualizing an image based on an ultra-wideband ground-transmitting radar according to the first embodiment of the present disclosure;
14 is a flowchart illustrating a method for visualizing an image based on an ultra-wide band ground-transmitting radar according to a second embodiment of the present disclosure; And
15 is a flowchart illustrating a method for visualizing an image based on an ultra-wideband ground-transmitting radar according to a third embodiment of the present disclosure.

Hereinafter, various embodiments of the present disclosure will be described with reference to the accompanying drawings. However, this is not intended to limit the technology described in the present disclosure to a specific embodiment, it is to be understood as including various modifications, equivalents, and/or alternatives of the embodiments of the present disclosure. . In connection with the description of the drawings, similar reference numerals may be used for similar elements.

In the present disclosure, expressions such as "have", "may have", "include", or "may include" are the presence of corresponding features (eg, elements such as numbers, functions, actions, or parts). And does not exclude the presence of additional features.

In this disclosure, expressions such as "A or B", "at least one of A or/and B", or "one or more of A or/and B" may include all possible combinations of items listed together. . For example, “A or B”, “at least one of A and B”, or “at least one of A or B” includes (1) at least one A, (2) at least one B, Or (3) it may refer to all cases including both at least one A and at least one B.

Expressions such as "first", "second", "first", or "second" used in the present disclosure may modify various elements regardless of order and/or importance, and one element may be converted to another. It is used to distinguish it from the component, but does not limit the component. For example, a first user device and a second user device may represent different user devices regardless of order or importance. For example, without departing from the scope of the rights described in the present disclosure, a first component may be referred to as a second component, and similarly, a second component may be renamed to a first component.

Some component (eg, a first component) is "(functionally or communicatively) coupled with/to)" to another component (eg, a second component) or " When referred to as "connected to", it should be understood that any of the above-described components may be directly connected to the other components described above, or may be connected through other components (eg, a third component). On the other hand, when a component (eg, a first component) is referred to as being “directly connected” or “directly connected” to another component (eg, a second component), a component different from a component It may be understood that no other component (eg, a third component) exists between the elements.

The expression "configured to (configured to)" used in the present disclosure is, for example, "suitable for", "having the capacity to" depending on the situation. It can be used interchangeably with ", "designed to", "adapted to", "made to", or "capable of". The term "configured to (or set)" may not necessarily mean only "specifically designed to" in hardware. Instead, in some situations, the expression "a device configured to" may mean that the device "can" along with other devices or parts. For example, the phrase “a processor configured (or configured) to perform A, B, and C” means a dedicated processor (eg, an embedded processor) for performing the operation, or by executing one or more software programs stored in a memory device. , May mean a generic-purpose processor (eg, a CPU or an application processor (AP)) capable of performing corresponding operations.

Terms used in the present disclosure are used only to describe a specific embodiment, and may not be intended to limit the scope of other embodiments. Singular expressions may include plural expressions unless the context clearly indicates otherwise. Terms used herein, including technical or scientific terms, may have the same meaning as commonly understood by one of ordinary skill in the technical field described in the present disclosure. Among the terms used in the present disclosure, terms defined in a general dictionary may be interpreted as having the same or similar meaning as the meaning in the context of the related technology, and unless explicitly defined in the present disclosure, an ideal or excessively formal meaning Is not interpreted as. In some cases, even terms defined in the present disclosure cannot be interpreted to exclude embodiments of the present disclosure.

An ultra-wideband ground-transmitting radar-based image visualization apparatus according to various embodiments of the present disclosure includes, for example, a smartphone, a tablet personal computer, a mobile phone, a video phone, and an e-book reader. (e-book reader), desktop personal computer (desktop personal computer), laptop personal computer (laptop personal computer), netbook computer, workstation, server, personal digital assistant (PDA), portable multimedia player (PMP) ), an MP3 player, a mobile medical device, a camera, or a wearable device.

1 is a flow chart showing a conventional image visualization method.

In step 101, the GPR system A scans the GPR data. A scan is a method of expressing data in one dimension using a horizontal axis representing distance or time and a vertical axis representing amplitude, and B scan is a two-dimensional representation of data representing the cross-sectional area of a signal.

In step 103, the GPR system performs pre-processing (matched filter, moving average filter, linear gain curve, etc.) before compensation.

The GPR system detects the target by forming an image in step 105.

The radar signal is compensated using a matched filter, a moving average filter, and a linear gain curve, and then an image is formed to detect the target.

The GPR system is largely divided into vehicle type and portable type. Unlike the vehicle type, the portable has a limitation in that it must perform landmine detection with less power. Therefore, the portable has a lot of simple and computationally small algorithms for fast data processing. However, these algorithms have the disadvantage that they are not robust to signals other than the target such as unevenness of the ground, vibration of the sensor and noise signals. That is, there is a problem in that the probability of generating a false alarm in which the background signal is recognized as a landmine signal and the probability of not detecting an actual buried mine by recognizing the mine signal as a background signal are high.

2 shows a block diagram of a GPR system.

The GPR system 200 includes a transmission/reception control and signal processing unit 210, a transmission unit 212, a transmission antenna 214, a reception unit 216, a reception antenna 218, and the like.

The transmission/reception control and signal processing unit 210 generates ultra-wideband pulses and transmits them to the transmission unit 212. In addition, the transmission/reception control and signal processing unit 210 receives a signal received from the reception unit 216.

The transmission unit 212 transmits the ultra-wideband pulse signal received from the transmission/reception control and signal processing unit 210 to the transmission antenna 214.

The transmission antenna 214 radiates the ultra-wideband pulse signal generated by the transmission/reception control and signal processing unit 210 to the buried object 220. Here, the buried object 220 refers to an object in which a landmine (object) is buried in the ground.

The receiving antenna 218 serves to receive a signal reflected from the buried object 220 after passing through the buried object 220 and returned.

The receiving unit 216 amplifies the signal input from the receiving antenna 218 and transmits the amplified signal to the transmission/reception control and signal processing unit 210.

3 is an exemplary diagram of data obtained by using a ground-transmitting radar after buriing a mine (object) in the ground.

Reflected waves are generated on the surface of an object artificially installed in a medium such as reinforcement inside concrete, tunnel linings, and underground burials, and the acquired reflected wave image is shown in the form of a parabola 310 as shown in FIG. 3. That is, if there is a target underground, a signal in the form of a parabola 310 is obtained in the direction in which the antenna moves.

The B-scan data includes various signals such as a landmine signal that appears in the form of a parabola 310, a reflection signal on the ground surface, and a signal that appears periodically in the depth direction. The GPR system must be able to quickly and accurately detect landmine signals among these signals. Therefore, there is a need for a method for detecting and visualizing landmines, which has a high computational speed and exhibits robust performance for signals other than mines.

4 is a block diagram of a GPR system according to an embodiment of the present disclosure.

The GPR system or GPR device (hereinafter referred to as a GPR system) includes a GPR data acquisition unit 401, a constant signal removal unit 402, a matched filter unit 403, a moving average filter unit 404, and a linear gain compensation unit ( 405), a nonlinear gain compensation unit 406, a time average application unit 407, a depth average application unit 408, a computer vision unit 410, an energy sum processing unit 420, an image correction unit 430, etc. Include.

The GPR data acquisition unit 401 acquires GPR data. That is, the GPR data acquisition unit 401 acquires data by using the GPR at a location where a landmine or an object is buried.

The constant signal removing unit 402 removes the constant signal from the acquired GPR data.

The matched filter unit 403 and the moving average filter unit 404 filter to remove the background signal from the signal output from the constant signal removing unit 402.

The time average application unit 407 and the depth average application unit 408 respectively apply the time average and the depth average to emphasize the parabolic line before applying the computer vision algorithm.

The computer vision unit 410 according to an embodiment of the present disclosure serves to apply a computer vision algorithm, and a differential mask processing unit 411, an edge detection unit 412, a rectangular window segment processing unit 413, and a local A distance calculation unit 414, a one window segment processing unit 415, a feature point extraction unit 416, and the like are included.

As shown in FIG. 6 to be described later, the differential mask processing unit 411 checks whether an outline exists in the image by the magnitude of the graph slope at the first derivative value, and determines whether the pixel of the outline by the sign of the graph slope at the second derivative value. Check the location of the bright and dark areas. In GPR signal processing, a mask filter is used to play the role of a differential operator. At this time, the sum of all pixels in the mask is set to 0.

Masks used in the differential mask processing unit 411 may include Canny, Log, Prewitt, Roberts, Sobel, Zero-cross, and the like.

The edge detection unit 412 detects an edge of an image based on the confirmation result of the differential mask processing unit 411 according to an embodiment of the present disclosure. Edge refers to a part that exists at a point where the intensity of an image changes from a small value to a high value, or vice versa, and refers to the boundary of an object in the image. Edge contains a number of information, such as the ability to detect the shape and direction. By providing this information to the operator, the clarity of the image can be improved and the detection rate can be improved.

The rectangular window segment processing unit 413 extracts the minimum value min and the maximum value max by forming a 3by3 window.

The local distance calculation unit 414 calculates the local distance based on the difference between the minimum value and the maximum value extracted by the rectangular window segment processing unit 413. For example, the distance is 3 (min=1, max=4). The local distance calculator 414 detects an edge of an image based on the calculated local distance.

The one window segment processing unit 415 extracts keypoints and forms a circle window to generate a corner point.

The feature point extracting unit 416 extracts a feature point based on the corner point generated by the original window segment processing unit 415. In addition, the feature point extraction unit 416 detects an edge of the image based on the extracted feature point.

The energy sum processing unit 420 determines an energy sum based on the edge detected by the computer vision unit 410, and obtains target intensity information based on the determined energy sum.

The image correction unit 430 corrects an image based on the edge detected by the computer vision unit 410 or a result value of the energy sum processing unit 420.

5 is a flowchart illustrating a method for visualizing an image based on an ultra-wide band ground-transmitting radar according to an embodiment of the present disclosure.

Reference numeral 510 denotes GPR data obtained by performing GPR exploration. That is, it represents the result value of the GPR data acquisition unit 401.

Reference number 520 denotes a result of applying a moving average filter to GPR data. That is, it shows the result value of the moving average filter unit 404 of FIG. 4.

Reference numeral 530 denotes a result of performing linear or nonlinear gain compensation. That is, the result of the linear gain compensation unit 405 or the nonlinear gain compensation unit 406 of FIG. 4 is shown.

Reference numeral 540 denotes a result value after correcting an image according to an embodiment of the present disclosure. That is, the result value of the image correction unit 430 of FIG. 4 is shown.

In an embodiment of the present disclosure, the operation of the matched filter unit 403, the moving average filter unit 404, the linear gain compensation unit 405, and the nonlinear gain compensation unit 406 on data acquired by an existing ultra-wideband ground-transmitting radar And, additionally, a computer vision algorithm (operation of the computer vision unit 410 of FIG. 4) as shown in FIG. 4 is additionally applied to improve the signal to noise ratio (SNR), thereby making the parabola clear and identifying the exact position of the image. I can. In computer vision algorithms, edge detection, local distance calculation, and feature point extraction of images are used.

6 is a diagram illustrating an example of a super-band ground-transmitting radar-based image according to the first embodiment of the present disclosure.

Reference numeral 610 denotes a waveform of the amount of change in the slope of the X-axis with respect to the change in brightness.

Reference numeral 620 denotes a first-order differential waveform, and reference numeral 630 denotes a second-order differential waveform.

For edge detection, the GPR system checks the existence of an outline in the image by the magnitude of the slope of the graph from the first derivative value as shown in reference number 620. Thereafter, the GPR system checks the positions of the bright and dark portions of the pixels of the outline with the sign of the slope of the graph from the second derivative value, as shown by reference number 630. In GPR signal processing, a mask filter is used to play the role of a differential operator. At this time, the sum of all pixels in the mask is set to 0.

7 is a diagram illustrating a result of edge detection for each mask according to the first embodiment of the present disclosure.

Masks used according to the first embodiment of the present disclosure include a Canny mask, a Log mask, a Prewitt mask, a Roberts mask, a Sobel mask, a zero-cross mask, and the like, and an image after applying the mask is shown in FIG. 7.

Reference number 701 denotes an edge detection result value to which a Sobel mask is applied.

Reference numeral 703 denotes an edge detection result value to which a Canny mask is applied.

Reference numeral 705 denotes an edge detection result value to which a log mask is applied.

Reference numeral 707 denotes an edge detection result value to which a Prewitt mask is applied.

Reference numeral 709 denotes an edge detection result value applying a Roberts mask.

Reference number 711 denotes the edge detection result value applying the zero-cross mask.

When using a canny mask, log mask, or zero-cross mask, the range of the edge is wide, making it difficult for the user to find the center of the target. So, we consider the Sobel mask, Prewitt mask, and Roberts mask, which are masks that detect only very clear edges, and consider using a Prewitt mask that has a fast computational load and is relatively less sensitive to noise.

8 is a diagram illustrating a visualization screen before/after an edge according to the first embodiment of the present disclosure.

Reference numeral 801 denotes a visualization screen before applying the first embodiment of the present disclosure, and reference numeral 803 denotes a visualization screen before applying the first embodiment of the present disclosure. When the first embodiment of the present disclosure is applied, as shown by reference numeral 803, the sharpness of an image may be increased.

9 is a diagram illustrating a super-band ground-transmitting radar-based image according to a second embodiment of the present disclosure.

In the second embodiment of the present disclosure, the GPR system applies a Rangefilt algorithm for extracting an edge using a local distance of an image. The Rangefilt algorithm is an algorithm that extracts a minimum value and a maximum value by forming a 3by3 window as shown in 910 of FIG. 9 and then extracts the difference between the maximum value and the minimum value. In FIG. 9, the distance is 3 (min=1, max=4), and an edge (reference number 920) of the image is detected by this principle. Reference numeral 920 of FIG. 9 denotes a super-band ground-transmitting radar-based image to which the second embodiment of the present disclosure is applied.

10 is a diagram illustrating a super-band ground-transmitting radar-based image according to a third embodiment of the present disclosure. FIG. 11 is a diagram showing a super-band ground-transmitting radar-based image to which the FAST method according to the third embodiment of the present disclosure is applied.

In the third embodiment of the present disclosure, the GPR system applies an image feature point extraction algorithm. The image feature point extraction algorithm is the most commonly used method when tracking or recognizing a target in an image, and recognizes by extracting a key point from an image.

The advantage of a good image feature is that it can be easily identified even if the shape, size, or location of the target changes, and the corresponding point in the image can be easily found even if the intensity or brightness changes. The best feature that satisfies these conditions is the Corner Point. Most feature point extraction algorithms are based on this corner point detection. In land mine image visualization, image feature point extraction algorithm was performed after making the boundary point prominent by applying a mask to form a corner point. Image feature point extraction algorithms include Harris Corner, Shi-Tomasi, SIFT-Dog, and FAST methods. In general, image feature point extraction algorithms have a large amount of computation and may cause problems in real-time processing. However, the FAST method, a method developed by Edward Rosten of the University of Cambridge in England, in 2006, is a feature point extraction algorithm that pursues extremely fast and can be processed in real time. In addition, the reason FAST is a very excellent algorithm is that the quality of the feature points (repeatability: the degree to which the same feature points are repeatedly detected even in various image changes) is also better than that of the existing methods (Harris, DoG, etc.). great. Therefore, feature points were extracted using the FAST method for portable mine detectors.

In the FAST algorithm, a pixel p equal to 1010 is selected from an image. Assume a circle of 16 pixels around pixel p.

In the FAST algorithm, as shown in FIG. 10, whether a point p is a corner is determined by looking at the value of 16 pixels on a circle having a radius of 3 around p. So, if n or more pixels are consecutively brighter than p (>p+t) or n or more dark (<p-t) pixels are consecutively more than a certain value, p is determined as a corner point. The visualized image of the portable mine detector using the FAST method is shown in FIG. 11.

12 is a diagram illustrating an image to which an ultra-wideband ground-transmitting radar-based image visualization method is applied according to an embodiment of the present disclosure.

12 is a visualized image to which a computer vision algorithm is applied, and SNR can be improved by sharpening a parabolic image using three computer vision algorithms (edge detection, local distance calculation, and image feature point extraction algorithm).

13 is a flowchart illustrating a method of visualizing an image based on an ultra-wide band ground-transmitting radar according to the first embodiment of the present disclosure.

The GPR system acquires GPR data in step 1301.

The GPR system removes ground reflection by applying a moving average filter to the GPR data acquired in step 1303.

In step 1305, the GPR system compensates the gain for the result value to which the moving average filter is applied to compensate for attenuation of the medium.

In step 1307, the GPR system takes an average by time or by depth to emphasize the parabolic line before applying the computer vision algorithm. It goes without saying that the operation of step 1307 can be omitted.

The GPR system applies a differential mask according to the first embodiment in step 1309. A detailed description of the differential mask is based on an operation description of the differential mask processing unit 411 of FIG. 4.

In step 1311, the GPR system detects an edge of an image according to the first embodiment. A detailed description of edge detection is based on an operation description of the edge detection unit 412 of FIG. 4.

The GPR system corrects the image based on the result of applying the computer vision algorithm in step 1313. A detailed description of the image correction is based on the operation description of the image correction unit 430 of FIG. 4.

14 is a flowchart illustrating a method for visualizing an image based on an ultra-wide band ground-transmitting radar according to a second embodiment of the present disclosure.

The GPR system acquires GPR data in step 1401.

The GPR system removes ground reflection by applying a moving average filter to the GPR data acquired in step 1403.

In step 1405, the GPR system compensates the gain for the result of applying the moving average filter in order to compensate for attenuation of the medium.

In step 1407, the GPR system takes an average by time or by depth to emphasize the parabolic line before applying the computer vision algorithm. It goes without saying that the operation of step 1407 can be omitted.

In step 1409, the GPR system processes the rectangular window segment according to the second embodiment. A detailed description of the rectangular window segment processing is based on the operation description of the rectangular window segment processing unit 413 of FIG. 4.

The GPR system detects the edge according to the second embodiment in step 1411. A detailed description of edge detection is based on an operation description of the local distance calculator 414 of FIG. 4.

The GPR system corrects the image based on the result of applying the computer vision algorithm in step 1413. A detailed description of the image correction is based on the operation description of the image correction unit 430 of FIG. 4.

15 is a flowchart illustrating a method for visualizing an image based on an ultra-wide band ground-transmitting radar according to a third embodiment of the present disclosure.

The GPR system acquires GPR data in step 1501.

The GPR system removes ground reflection by applying a moving average filter to the GPR data acquired in step 1503.

In step 1505, the GPR system compensates the gain for the result of applying the moving average filter to compensate for the attenuation of the medium.

In step 1507, the GPR system takes an average by time or by depth to emphasize the parabolic line before applying the computer vision algorithm. It goes without saying that the operation of step 1507 can be omitted.

In step 1509, the GPR system processes the original window segment according to the third embodiment. A detailed description of the one window segment processing is based on the operation description of the one window segment processing unit 415 of FIG. 4.

The GPR system detects the edge according to the third embodiment in step 1511. A detailed description of edge detection is based on the operation description of the feature point extraction unit 416 of FIG. 4.

The GPR system corrects the image based on the result of applying the computer vision algorithm in step 1513. A detailed description of the image correction is based on the operation description of the image correction unit 430 of FIG. 4.

Computer vision algorithms applied in an embodiment of the present disclosure include edge detection, local distance calculation (Rangfilt), and image feature point extraction algorithms. Its effect is as follows.

First, edge detection using a Prewitt mask has a fast computational load and is relatively robust to noise. When implemented with a systematic tense visualization algorithm, a clear image can be obtained by emphasizing the edge of a parabolic in real time.

In addition, the Rangefilt algorithm also emphasizes the edge of the parabolic using the local distance of the image, so that even a target with a small RCS (Radar Cross Section) can be easily detected by the user, the detection performance can be improved.

In addition, when a specific point extraction algorithm, which finds and extracts feature points of a parabolic line even when the shape, size, and location of the target changes, is applied to a portable mine detector, it is possible to provide the user with a very effective visual method of not only the existence of a mine, but also the buried location. As a result, the detection rate can be increased by improving the judgment of the presence or absence of a landmine. As described above, after removing the clutter, the SNR can be improved by making the parabolic image clear using three computer vision algorithms.

The above-described contents may be modified and modified without departing from the essential characteristics of the present disclosure by those of ordinary skill in the technical field to which the present disclosure belongs. Accordingly, the embodiments of the present disclosure are not intended to limit the technical idea, but to describe it, and the scope of the technical idea of the present disclosure is not limited by these embodiments. The scope of protection of the present disclosure should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present disclosure.

Claims (8)

In an ultra wide band ground penetrating radar-based image visualization device,
Transmitting/receiving unit for transmitting and receiving data; And
Acquires B-scan image data using the ultra-wideband ground-transmitting radar through the transmitter/receiver, and outputs B-scan image compensation data by applying moving average filtering and gain compensation to the B-scan image data, In order to emphasize a parabolic signal representing a land mine in the B-scan image compensation data, a B-scan image enhancement data is output by applying a time average and a depth average to the B-scan image compensation data, and the B-scan image enhancement By applying a differential operator to data, an edge is detected in the B-scan image enhancement data, the B-scan image enhancement data is corrected based on the detected edge, and the corrected B-scan image enhancement data In the case where the parabolic signal is present, a control unit for determining that the landmine has been detected,
The control unit generates a first differential waveform by first differentiating the waveform of the amount of change in the slope of the X-axis relative to the change in brightness from the B-scan image enhancement data using a first differential operator, and based on the magnitude of the slope of the first differential waveform. To check the existence of the edge in the B-scan image emphasis data, and generate a second differential waveform by second differentiating the first differential waveform using a second differential operator, and the slope of the second differential waveform It characterized in that the position of the light or dark part at the edge is checked based on the sign,
The B-scan image data is data representing a surface area of a signal in two dimensions.
delete The method of claim 1,
The differentiation operator uses a mask filter,
The mask filter is one of a Sobel mask filter, a Prewitt mask filter, and a Roberts mask filter. delete delete delete delete delete

Images