JPH04184185A - Searching for invisible article - Google Patents

Abstract

PURPOSE:To search for an invisible article with high precision by suppressing the peak value of a noise part smaller in comparison with the position of the invisible article by using three kinds of normalization data of (>=1, 0, and +1) as the data for carrying out the synthetic aperture processing. CONSTITUTION:The position data shown by a black round point is normalized to '+1', and the position data shown by a white round point is normalized to '-1', and other points are normalized to '0'. Through the synthetic aperture processing, the points of a hyperbola shown by a broken line 11 are converged at a point 13 shown by the arrow. Accordingly, the value of this point is '+7', and forms a peak. Further, as for the data in a noise part which appears in the lower part of the figure, the points on the hyperbola shown by the broken line 12 are converged at a point 14 shown by the arrow through the synthetic aperture processing, and the value at this point is '-1', and the pseudo conversion of the noise part is suppressed. Accordingly, the position of a conduit has a peak, and the noise point is suppressed by the offset between '+1' and '-1', and the erroneous recognition is prevented.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an invisible object detection method for detecting the position of an invisible object based on collected reflected wave data.

[Conventional technology]

As an example, underground exploration of underground pipes will be explained.

Pulse electromagnetic waves are emitted from the earth's surface into the ground, and the reflected waves are stored. This is performed every fixed scanning distance while moving, and after the scanning is completed, image data of reflected waves (reflected wave data) is obtained, and this data is used as input.

FIG. 4 is an example of collected reflected wave data.

The upper row of the horizontal axis shows the moving distance during moving scanning, the lower row of the horizontal axis shows the amplitude of the reflected wave, and the vertical axis shows the time from pulse signal transmission to reception, which corresponds to the depth direction.

Next, for the collected reflected wave data, a positive sign is assigned to one side of the amplitude direction of the reflected wave data and a negative sign is assigned to the other side, and the point where the reflected wave data changes from a neutral value (0) to a positive value is determined by Normalized data is created by normalizing the amplitude values at the points where the amplitude changes from a negative value to a negative value and from a negative value to a neutral value to "+1", and the other amplitude values to "0". FIG. 5 shows how one line of reflected wave data is normalized.

Then, the synthetic aperture method is applied to the created normalized data. Regarding this synthetic aperture method, patent application No. 2-5
Since it is shown in No. 8173, its details will be omitted.

Since the conduit appears as a hyperbolic image on the normalized data, by performing synthetic aperture processing, the data on the hyperbola are concentrated at the vertex of the hyperbola.

Therefore, image data having a high data value is obtained at the position where the conduit exists, and the observer can recognize the position where the peak of this image data exists as the position of the conduit. Alternatively, it can be automatically calculated using a calculation device that can detect the data level.

[Problem to be solved by the invention]

However, in addition to the large reflected waves from the pipe, the reflected waves include small waves as noise, and therefore the normalized data includes many noise points. The reflected wave from the pipe has continuity in the lateral direction, but the noise part has low correlation in the lateral direction. ro that contains a lot of noise like this
As a result of synthetic aperture processing using binary normalized data of There is a possibility that it will be observed as

FIG. 6 shows the result of performing synthetic aperture processing on the binary normalized data generated from the reflected wave data shown in FIG.

The position data indicated by black dots is normalized to "+1",
All other points are normalized to "0". By the synthetic aperture processing, points on the hyperbola indicated by the broken line 21 converge at a point 23 indicated by the arrow. As a result, the value at this point becomes "+7", which is the peak. Furthermore, when synthetic aperture processing is performed on the data of the noise portion appearing at the bottom of Fig. 6, the broken line 22
The points on the hyperbola converge at point 24 indicated by the arrow, and the value at this point is "+5", which is a fairly large value and can cause erroneous recognition.

[Means to solve the problem]

The present invention was proposed in order to solve such problems, and it assigns a first state code to one side of the collected reflected wave data in the amplitude direction and a second state code to the other side, and changes the collected reflected wave data from the neutral state to the second state code. The amplitude value at the point where the state code changes to 1 and the point where the second state code changes to the neutral state is set to "+1", and the amplitude value at the point where the state code changes from the first state code to the second state code is set to r.
-IJ, other amplitude values are normalized to "0" to create normalized data, and the created normalized data is subjected to synthetic aperture processing.

[Effect]

Therefore, according to the present invention, ternary normalized data of r-1, 0, +IJ is used as data for performing synthetic aperture processing instead of conventional binary normalized data of rO, +IJ. As a result, the noise part that conventionally caused erroneous recognition becomes -1°0, +1 randomly existing on the hyperbola that is the target of addition in synthetic aperture processing, and -1 and +1 cancel each other out. , it is possible to suppress the peak value to be smaller than the position of the invisible object.

〔Example〕

Hereinafter, the method for detecting an invisible object according to the present invention will be explained by taking, as an example, an underground probe for a pipe existing underground.

First, reflected wave data is collected in the same manner as before. Next, for the collected reflected wave data, give a positive sign to one side and a negative sign to the other side in the amplitude direction of the reflected wave data (see Figure 4), and change from the neutral value (O) to a positive value. The amplitude value of the point and the point that changes from negative value to neutral value is set to “+1”.
”, the amplitude value at the point where it changes from a positive value to a negative value is r−I
J and other amplitude values are normalized to "0" to create normalized data. Figure 2 shows how one line of reflected wave data is normalized.

Then, synthetic aperture processing is performed on the created three-value normalized data.

FIG. 3 shows the result of synthetic aperture processing performed on the ternary normalized data created from the reflected wave data shown in FIG. 4.

Position data indicated by black dots is normalized to "+1", position data indicated by white dots is normalized to "r-1", and all other points are normalized to "0". With synthetic aperture processing,
Points on the hyperbola indicated by a broken line 11 converge at a point 13 indicated by an arrow.

As a result, the value at this point becomes "+7", which is the peak. Furthermore, when synthetic aperture processing is performed on the data of the noise portion appearing at the bottom of Fig. 3, the points on the hyperbola of the broken line 12 converge at the point 14 indicated by the arrow, and the value of this point becomes "-1". As a result, pseudo-concentration of noise parts can be suppressed. That is, the position of the conduit has a peak, and the noise point is suppressed by the cancellation of +1 and -1, so that there is no possibility of erroneous recognition. FIG. 1 shows the above-mentioned processing flow. Further, the table below shows a comparison with the conventional synthetic aperture processing results shown in FIG.

In addition, in the above-mentioned embodiment, the exploration is performed by emitting electromagnetic waves, but the exploration may be performed by emitting ultrasonic waves, and various other transmission waves can be considered.

Further, when creating the normalized data, the amplitude values to be normalized may be weighted according to the rate of change of the reflected wave data.

In addition, in FIG. 2, the amplitude of the reflected wave data is given a positive sign in the right direction in the figure and a negative sign in the left direction in the figure.
The right direction may be negative and the left direction may be positive. In this case, the amplitude value at the point where the neutral value (0) changes to a negative value and the point where the positive value changes to the neutral value is set to "+1", and the amplitude value at the point where the value changes from a negative value to a positive value. It is assumed that normalized data is created by normalizing the amplitude values to "-1" and the other amplitude values to "0".

〔Effect of the invention〕

As is clear from the above explanation, according to the present invention, [-1, O, +IJ
The three-value normalized data is used, and the peak value of the noise part can be suppressed to be smaller than the position of the invisible object, making it possible to search for the invisible object with high precision.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing a processing flow when performing underground exploration of underground pipes as an application example of the invisible object exploration method according to the present invention, and FIG. 2 is a diagram showing the reflected wave data shown in FIG. 4. Figure 3 is a diagram showing the state of standardization of one line by applying the method of the present invention, Figure 3 is a diagram showing the result of synthetic aperture processing on the ternary normalized data column by applying the method of the present invention, and Figure 4 The figure shows an example of collected reflected wave data, Figure 5 shows how one line is normalized by applying the conventional method to this reflected wave data, and Figure 6 shows how the reflected wave data is normalized by applying the conventional method. FIG. 3 is a diagram showing the results of synthetic aperture processing performed on the binary normalized data of FIG. 11.12... broken line, 13.14... point (convergence point)
. Patent applicant Nippon Telegraph and Telephone Corporation

Claims (2)

[Claims] (1) Emit a transmitted wave, memorize the reflected wave, collect reflected wave data by scanning it every predetermined distance while moving it, and detect the position of an invisible object based on the collected reflected wave data. In the invisible object detection method, a first state code is assigned to one side of the amplitude direction of the reflected wave data and a second state code is assigned to the other side, and a point where the neutral state changes to the first state code and the second state are determined. Add 1 to the amplitude value at the point where the sign changes to the neutral state.
Then, normalized data is created by normalizing the amplitude value at the point where the first state code changes to the second state code to "-1" and the other amplitude values to "0", and this created standard An invisible object detection method characterized in that synthetic aperture processing is performed on converted data. (2) In claim 1, when creating the standardized data,
An invisible object detection method characterized in that the amplitude value to be normalized is weighted according to the rate of change of reflected wave data.