Communications: directive radio wave systems and devices (e.g. – Transmission through media other than air or free space
Reexamination Certificate
2001-05-31
2003-06-03
Gregory, Bernarr E. (Department: 3662)
Communications: directive radio wave systems and devices (e.g.,
Transmission through media other than air or free space
C342S176000, C342S180000, C342S195000, C702S014000, C702S016000, C367S014000, C367S068000, C367S072000, C367S087000
Reexamination Certificate
active
06573855
ABSTRACT:
TECHNICAL FIELD
The present invention relates to a method and an apparatus for three-dimensional exploration wherein location of an object present within a medium is found by transmitting wave signals by means of an electromagnetic wave or sonic wave into the medium in the course of movement over the surface of the medium and receiving the signals reflected from the object within the medium and processing the received reflected-signals. The invention relates also to a method and an apparatus for displaying three-dimensional voxel data generated in the form of coordinates (x, y, t) consisting of a position (x, y) on the surface of the medium and a reflection time (t) based on a reflected signal intensity of a wave signal transmitted from the surface of medium into this medium.
BACKGROUND ART
For the three-dimensional exploration as noted above, a three-dimensional exploratory apparatus is employed for exploring an object or a hollow space buried or present underground.
A typical conventional art is known from a paper entitled: “Underground Buried Object Exploring Radar System (Part 3); Three-Dimensional Exploration Image Processing” (National Convention of The Institute of Electrical Engineers of Japan, 63
rd
year of Showa, p 1372).
According to this art (first prior art), measured section information obtained by a plurality of cycles of scanning operations is used. If an image of the buried unidentified object is obtained at identical positions in all of the plurality of section images, then the buried object is judged as a pipe. Whereas, if the image of the buried object is obtained at identical positions in fewer than all of the section images, the buried object is then judged as a block object. By determining the connectivity
on-connectivity between the images present in different sections, the three-dimensional structure is obtained.
In the above, only one kind of threshold value is employed in a binary encoding scheme based on which the presence/absence of the object is to be judged.
With the above-described prior art, the symbolic representations of the respective sections (e.g. binary symbol representation representing presence/absence of the object according to the reflected signal intensity) are connected three-dimensionally. Hence, the setting of the threshold value employed for the symbolizing scheme significantly influences the determination of the unidentified object either as a pipe or as a block object. Especially, in the case of underground exploration, the S/N ratio is low and the intensity of the reflected signal from the object can vary significantly with change of the exploring position. Therefore, in binarizing the reflected signal according to its intensity, if the threshold value is significantly lowered to enable detection of the pipe, this will lead to occurrence of a large amount of noise region of e.g. unwanted reflected signals. This is because the information about the neighboring sections is not utilized at all for the binarizing process.
Moreover, this prior art requires that the buried pipe be oriented in a direction perpendicular to the scanning direction of the apparatus. Therefore, although the symbolizing method as above may be useful on such prerequisite, the reliability of exploration is reduced when the buried pipe is not oriented in the direction perpendicular to the scanning direction.
Also, with the three-dimensional exploratory apparatus, the electromagnetic wave is transmitted into the ground at a position (x, y) on the ground surface and the signal reflected from the buried object is received, in the course of which the reflected signal intensity (s) is determined at every reflected time (t) of a predetermined interval. Therefore, if positions (x, y) on the ground surface of the three-dimensional exploratory apparatus are taken in the form of a grating pattern of a predetermined spacing, it is possible to obtain complete three-dimensional voxel data s (x, y, t) having data values (reflected signal intensities (s)) at all of its voxels. However, when the three-dimensional exploratory apparatus is scanned on the ground surface, depending on the actual surface condition of the exploration site such as the ground surface being a road surface, it is not always possible to effect this scanning operation exactly in the form of grating pattern, due to safety and/or time restriction. Then, while the data measurement can be done densely in the direction of the reflection time (t), both voxels having data values and voxels not having any data values coexist in the x-y plane. Here, for such three-dimensional voxel data, a voxel having a data value is defined as a source voxel and a voxel not having any data value is defined as a deficient voxel.
Conventionally, as a method of interpolating such a deficient voxel in such three-dimensional voxel data which are irregularly present in the x-y plane, it is well-known, as a second prior art, to effect weighting of the interpolation according to its distance from a source voxel. According to this prior art, when the distance between a deficient voxel and a source voxel (voxel value (s)) is D, the interpolation for this deficient voxel is done with a value (s′) obtained by the following formula 1:
S
′=(&Sgr;
D
−E
s
)/(&Sgr;
D
−E
) (1)
where, E is provided for adjusting the degree of weighting, which can be a numerical value of 3, 5, etc. This numerical value will be appropriately selected, depending on the density of the data to be interpolated and/or dispersion of the voxel values. This interpolation can be done three-dimensionally. However, since the data are present densely in the direction of the reflection time (t), this interpolation is rather considered as interpolation of two-dimensional data within the x-y plane. Then, by effecting the interpolation at each reflection time (t) interval with using the obtained weighting value for the two-dimensional plane, the amount of calculation needed may be significantly reduced.
If the interpolation of three-dimensional voxel data is effected according to the above-described prior art, whether it is effected three-dimensionally or two-dimensionally, the following three problems occur.
Firstly, when another source voxel is present at a slightly distant location along the same direction from a deficient voxel to be interpolated toward a source voxel and these two source voxels have significantly differing data values (e.g. when their signs are different being positive or negative), the target value for the deficient voxel to be interpolated may be influenced also by the value of the distant source voxel.
Secondly, since the interpolation is effected in accordance with the distance alone, that is, without considering the direction, an extrapolation which generates a value far less reliable than obtained with an interpolation may also take place inadvertently.
Thirdly, if no source voxel at all is present in the vicinity of the deficient voxel, the interpolation will be effected anyway in a forcible manner by using the data value of a very distant source voxel, whereby the precision of the interpolated value will suffer considerably.
As a method devised for solving the first and second problems noted above, there is known a second conventional method. With this method, a Delaunay triangulation diagram is obtained by calculations from the two-dimensional distribution of source voxels in the x-y plane. Then, for a deficient voxel present within each triangle of the diagram, by using a data value of a source voxel present at the apex of this triangle, an interpolation operation is effected for the deficient voxel with a weighting corresponding to its distance therefrom. However, in order to obtain such Delaunay triangulation diagram by calculations, if ‘n’ units of source voxels are present in the two-dimensional distribution thereof within the x-y plane, a vast amount of calculations on the order of n
2
to n
3
will be necessary. Moreover, the third problem remains unsolved.
As a third prior art, the following metho
Hayakawa Hideki
Kawanaka Akira
Takesue Yasuhiro
Gregory Bernarr E.
Osaka Gas Co. Ltd.
Webb Ziesenhiem Logsdon Orkin & Hanson P.C.
LandOfFree
Three-dimensional questing method, three-dimensional voxel... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Three-dimensional questing method, three-dimensional voxel..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Three-dimensional questing method, three-dimensional voxel... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3161267