CN107450054B - A kind of adaptive Coherent Noise in GPR Record denoising method - Google Patents
A kind of adaptive Coherent Noise in GPR Record denoising method Download PDFInfo
- Publication number
- CN107450054B CN107450054B CN201710575039.XA CN201710575039A CN107450054B CN 107450054 B CN107450054 B CN 107450054B CN 201710575039 A CN201710575039 A CN 201710575039A CN 107450054 B CN107450054 B CN 107450054B
- Authority
- CN
- China
- Prior art keywords
- coherent noise
- noise
- data
- singular value
- gpr record
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/28—Details of pulse systems
- G01S7/285—Receivers
- G01S7/292—Extracting wanted echo-signals
- G01S7/2923—Extracting wanted echo-signals based on data belonging to a number of consecutive radar periods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/885—Radar or analogous systems specially adapted for specific applications for ground probing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/28—Details of pulse systems
- G01S7/285—Receivers
- G01S7/292—Extracting wanted echo-signals
- G01S7/2923—Extracting wanted echo-signals based on data belonging to a number of consecutive radar periods
- G01S7/2928—Random or non-synchronous interference pulse cancellers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/35—Details of non-pulse systems
- G01S7/352—Receivers
- G01S7/354—Extracting wanted echo-signals
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a kind of adaptive Coherent Noise in GPR Record denoising methods, by utilizing singular value decomposition at a series of data components corresponding from different singular values Coherent Noise in GPR Record, in order to individually handle data according to unlike signal or noise type；Then the first arrival lineups approximation detected object surface undulation situation of Coherent Noise in GPR Record is utilized, while the root mean square for calculating first arrival lineups is high；Finally using root mean square height as the foundation for judging automatically singular value corresponding with coherent noise, to achieve the purpose that automatically remove coherent noise, the trouble for needing artificially to judge singular value based on SVD decomposition method is avoided, singular value is directly judged according to root mean square high parameter, it is not only accurate but also time saving.
Description
Technical field
The invention belongs to engineering exploration technical field of data processing, and in particular to a kind of adaptive Coherent Noise in GPR Record denoising
Method.
Background technique
Letmbach at the beginning of Ground Penetrating Radar from last century and Lowy propose that concept has been passed by the course in century more than 1 so far, no
Only theoretical mature, practical application also achieves huge achievement.However in practical applications, radar signal inevitably will receive
Various noise jammings, so as to cause signal-to-noise ratio reduction, deep signal is blanked, and imaging effect is poor, by radar signal Inversion Calculation
The accuracy of all kinds of parameters is even more that will receive different degrees of influence, so that end result quality is greatly reduced.
Divide from noise source, Ground Penetrating Radar noise can be mainly divided into noise of instrument, antenna coupled noise, humane industry
Nonisotropic scattering, strong earth surface reflection, relief surface scattering and various multiple reflections in noise, medium etc.；For all kinds of
Noise, existing denoising method mainly include Likelihood ration test technology, parameter system identification technology, WAVELET PACKET DECOMPOSITION technology, son
Space filtering technology, independent component analysis, average value filtering technology, frequency filtering technology etc., although these methods have sufficiently
Theoretical foundation, ideal effect can be also obtained in model data, but be limited to various hypothesis and physical condition, processing made an uproar
Target is generally unattainable when the real data of sound pollution.
Therefore, seeking high-precision, high efficiency, adaptable denoising method is that successful application Ground Penetrating Radar solution is actually asked
The antecedent basis of topic.Singular value decomposition (SVD) is the matrix disassembling method by linear algebra based on theoretical, and core concept is
Higher-dimension complex matrix is expressed as low-dimensional simple matrix, thus the main feature of prominent original matrix；SVD technique is equal in every field
It is widely used, including compression of images, signal denoising, pattern-recognition, feature extraction etc., however these applications all suffer from one altogether
How same problem accurately determines singular value number, and this is the committed step for successfully utilizing SVD technique.For this purpose, each neck
The experts and scholars in domain have developed respective processing method according to specific objective, also achieve certain effect, however there are no which kind of
Method has extensive adaptability.
Remove horizontal regular disturber face in Ground Penetrating Radar, Abujarad etc. simply by corresponding to first singular value at
Divide and be considered as horizontal disturbance, regard remaining data as useful signal, this method is only effective to extremely regular horizontal disturbance noise.
In addition, further including the methods of unusual entropy production, signal-to-noise ratio experience, these methods are all semi-quantitative method, often rely on place
The experience of reason personnel, this not only lowers working efficiencies, and lack objectivity.In response to this problem, the base decomposed in SVD is needed
The method for judging automatically singular value number is studied on plinth, and target component is recombined to realize precise and high efficiency according to singular value
Purpose is denoised, to improve the signal-to-noise ratio of radar data, prominent echo signal feature provides the basis of high quality for subsequent processing
Data guarantee the reliability of result.
Summary of the invention
In view of above-mentioned, the present invention provides a kind of adaptive Coherent Noise in GPR Record denoising methods, can automatically determine relevant
Noise singular number avoids that human-computer interaction is needed to determine singular value number bring trouble for different data.
A kind of adaptive Coherent Noise in GPR Record denoising method, includes the following steps:
(1) for original Coherent Noise in GPR Record, the data in its preceding k nanosecond is intercepted and determine the standard track of data, k is big
In 0 real number；
(2) it for any scanning road, calculates cross-correlation function within preceding k nanosecond of the scanning road and standard track and takes it most
Big value；
(3) all scanning roads are traversed according to step (2), obtains the first arrival being made of each scanning road cross-correlation function maximum value
Curve L, and then calculate the high D of root mean square of first arrival curve L；
(4) singular value number N corresponding with coherent noise is calculated according to the following formula；
N=ROUND { 0.2634D+1.3086 }
Wherein: ROUND { } is the function that rounds up；
(5) singular value decomposition is carried out to original Coherent Noise in GPR Record, the top n data component after abnormal value elimination decomposition,
Denoising process is completed after remaining data ingredient is recombined.
Original Coherent Noise in GPR Record in the step (1) is the data matrix of n × m size, n and m respectively correspond for when
Between and sampling number spatially；The scanning road is the radar data of any sampled point on corresponding detected object surface, described
Standard track is the radar data of a certain sampled point on corresponding detected object surface, which is random or artificial specified.
Cross-correlation function expression formula in the step (2) is as follows:
Wherein: Rj(τ) is that jth scans the cross-correlation function of road and standard track within preceding k nanosecond, x0It (t) is the mark of t moment
Quasi- road, t and τ are time sampling serial number and 1≤τ≤K, K are the sampling number of radar data in preceding k nanosecond, xj *(t- τ) is xj
The conjugate complex number of (t- τ), xj(t- τ) is that the jth at t- τ moment scans road, and j is natural number and 1≤j≤m.
The high D of root mean square of first arrival curve L is calculated in the step (3) according to the following formula:
Wherein: d (j) is that jth scans the cross-correlation function maximum value of road and standard track within preceding k nanosecond,For first arrival song
The average value of line L.
Preferably, k=5.
The present invention is by utilizing singular value decomposition at a series of data corresponding from different singular values Coherent Noise in GPR Record
Ingredient, in order to individually handle data according to unlike signal or noise type；Then the same phase of first arrival of Coherent Noise in GPR Record is utilized
Axis approximation detected object surface undulation situation, while the root mean square for calculating first arrival lineups is high；Finally using root mean square height as certainly
The dynamic foundation for judging singular value corresponding with coherent noise is avoided and is based on to achieve the purpose that automatically remove coherent noise
SVD decomposition method needs artificially judge the trouble of singular value, and singular value is directly judged according to root mean square high parameter, not only accurately but also
It is time saving.
Detailed description of the invention
Fig. 1 is the Coherent Noise in GPR Record schematic diagram that practical Noise is obtained in the case of detecting object surfacing.
Fig. 2 is the first arrival Dependence Results schematic diagram that Fig. 1 is calculated automatically from using cross-correlation method.
Fig. 3 is that Fig. 1 utilizes the preceding 50 singular values distribution results schematic diagram obtained after singular value decomposition.
Fig. 4 is the noise schematic diagram that Fig. 1 is extracted using the method for the present invention.
Fig. 5 is that Fig. 1 utilizes the schematic diagram data after the method for the present invention denoising.
Fig. 6 is to detect the Coherent Noise in GPR Record schematic diagram that practical Noise is obtained in the non-smooth situation in object surface.
Fig. 7 is the first arrival Dependence Results schematic diagram that Fig. 6 is calculated automatically from using cross-correlation method.
Fig. 8 is that Fig. 6 utilizes the preceding 50 singular values distribution results schematic diagram obtained after singular value decomposition.
Fig. 9 is the noise schematic diagram that Fig. 6 is extracted using the method for the present invention.
Figure 10 is that Fig. 6 utilizes the schematic diagram data after the method for the present invention denoising.
Specific embodiment
In order to more specifically describe the present invention, with reference to the accompanying drawing and specific embodiment is to technical solution of the present invention
It is described in detail.
Original Coherent Noise in GPR Record is resolved into heterogeneity using singular value decomposition by the present invention, utilizes the equal of first arrival curve
Root height extracts singular value decomposition ingredient other than noise and synthesizes new data as the parameter for determining coherent noise distribution, real
Existing self-adaptive solution purpose.
If A is the original Coherent Noise in GPR Record of n × m rank, wherein n and m is time and spatial sampling points respectively, and A is carried out
Singular value decomposition is shown below:
Wherein: U is n × n rank orthogonal matrix, is often classified as covariance matrix AATFeature vector ui(1≤i≤n), referred to as
The left singular matrix of A；V is m × m rank orthogonal matrix, is often classified as covariance matrix ATThe feature vector v of Aj(1≤j≤m) claims
For the right singular matrix of A；Σ is the n × m rank diagonal matrix being made of r (order that r is A) a element, element be from greatly to
The σ of minispreads(1≤s≤r), and σ1≥σ2≥...≥σr≥0；σsThe referred to as singular value of A, the actually characteristic value of A
Square root；Above formula is unfolded:
Each single item u in formulasvs TIt is the matrix of one with the completely same dimension of A, it represents s-th of characteristic component of A；Above formula table
Bright A can be broken down into a series of different characteristic components, σsIndicate weight factor, it determines character pair ingredient to having constituted
The contribution of whole A.According to a large amount of practical experiences, in most cases, preceding 10% singular value just occupies all unusual
99% or more of the sum of value, for certain special circumstances, preceding 1% singular value almost just controls the master of entire data or image
Want feature；It in other words, can be with greatly approaching former data A using minimal amount of singular value, it may be assumed that
U' in above formulan×k、Σ'k×k、It is by U, Σ, V respectivelyTPreceding k at the matrix being grouped as, k is usually one
Much smaller than the positive integer of dimension n and m.
Ground penetrating radar detection subject surface is more smooth, and first singular value is more prominent, and when denoising, selected singular value number was got over
It is small；Conversely, surface is more uneven, singular value is more uniform, when denoising it is selected singular value number it is more, that is, it is unusual when denoising
The selection of value is related to detected object surface undulation situation.The present invention calculates first arrival curve using cross-correlation method come approximate fluctuating table
First arrival curve fluctuating situation is quantitatively portrayed in face using root mean square height, and detailed process is as follows:
(1) select before original Georadar Data data in 5 nanoseconds windows, one of waveform of simultaneous selection be clear, not by
The data of noise jamming calculate the cross-correlation function of standard track and other each track datas using following formula as standard track:
In formula: j is scanning Taoist monastic name,It is xjThe complex conjugate of (), x0() is to choose to carry out mutually with other each track datas
Relevant standard track, Rj(τ) is corresponding cross-correlation function.Standard track x0() can determine at random, most of under normal circumstances
Road all can serve as standard track.
(2) after obtaining cross-correlation function, then wherein maximum value d is found out；According to cross-correlation function and first arrival time standard meter
The first arrival time for calculating all roads, using each road first arrival time in order drawing forming curves as the same phase of the first arrival of Coherent Noise in GPR Record
Axis.
D=max (Rj(τ))
D illustrates the offset of current road and standard track, finds out the maximum value of the cross-correlation function of all roads and standard track,
And discharge them in order, just obtain first arrival lineups L, expression formula are as follows:
L=d (j), 1≤j≤m
(3) after obtaining oscillating curves L, the high D of root mean square is calculated, it describes curve in the mean fluctuation journey of vertical direction
Degree, expression formula are as follows:
In formula:For the average value of L.
The high D of root mean square is substituted into following formula:
N=0.2634D+1.3086
(4) it is calculated singular value number N corresponding with coherent noise, then by the top n number after the singular value decomposition of front
It is rejected according to ingredient and remaining data ingredient is recombined into new data, to obtain the data of final removal noise.
Embodiment one
It is as shown in Figure 1 the practical Noise Coherent Noise in GPR Record obtained in the case of detection object surfacing, it can from Fig. 1
To find out much noise preservation in section, the accurate expression of useful signal has been suppressed.It furthermore can be with from the form project of noise
Preliminary judgement, corresponding detection object surface is more smooth, rises and falls there is no serious.Fig. 2 is intercepted on the basis of Fig. 1
5 nanoseconds window data, selecting road where black vertical fine line in Fig. 2 is standard track and other each roads progress cross-correlation meters
It calculates, and obtains all road first arrival times automatically on this basis, obtained first arrival curve shown in black heavy line in Fig. 2.It can be with
Find out, the fluctuations form of curve and the first arrival take-off time in all roads coincide very well.Root mean square is calculated to first arrival curve
High D, and substituted into formula N=0.2634D+1.3086 and round numbers part obtains N=1, i.e., for the data, level is relevant dry
Disturbing corresponding data component is the 1st singular value tie element.Initial data shown in Fig. 1 is subjected to singular value decomposition, and before
30 singular values are drawn in as shown in Figure 3 as a result, the 1st singular value is that noise corresponds to singular value in Fig. 3.It can from Fig. 3
Out, the 1st singular value differs very big with other singular values, and this also illustrates the centralities of noise, need to only correspond to the 1st value
Ingredient removal can reach preferable denoising effect, it is seen that the singular value judged automatically is reasonable.Fig. 4 is unusual by the 1st
Value corresponds to that noise extracts as a result, comparison diagram 1 and Fig. 4 are it is found that most noises are all successfully separated.By remaining data
Data after recombining new data, that is, denoising, as shown in figure 5, useful signal has obtained good reply, horizontal direction is made an uproar
Sound is eliminated well；Whole process executes automatically, determines parameter without artificial.
Embodiment two
Fig. 6 is a more serious real data of noise situations, and useful signal is other than few region is distinguishable, greatly
Partial region is all covered by the noise of strong amplitude.And may determine that from the morphological feature of noise, detection object surface undulation is compared
It is severe.This brings very big difficulty to denoising, and general denoising method can all fail.What Fig. 7 was still intercepted on the basis of Fig. 6
Data in preceding 5 nanoseconds window, black vertical fine line are standard track, and black heavy line is automatic according to standard track and cross-correlation
The first arrival curve of calculating, it is seen that curve also coincide very well with the data first arrival to fluctuate.Root mean square is calculated to first arrival curve
High D, and substituted into formula N=0.2634D+1.3086 and round numbers part obtains N=2, i.e., for the data, level is relevant dry
Disturbing corresponding data component is the 1st, 2 singular value tie element.Fig. 8 is obtained after initial data is carried out singular value decomposition
Preceding 30 singular value distribution situations.Compared with the singular value in embodiment one, first singular value has very big decline, and with it is odd below
The gap of different value substantially reduces, and the distribution situation of noise is reducing with the difference of other signals in this surface initial data, this nothing
It doubts and increases denoising difficulty.But more it is apparent that the difference between the first two singular value and subsequent singular value is still more bright
Aobvious, this also illustrates the differences of the two singular value corresponding data ingredients and follow-up data, actually namely lateral noise, this
Also illustrate that the present embodiment determines that the method for singular value is effective automatically.Fig. 9 is to extract the first two singular value to correspond to noise, with
Initial data shown in Fig. 6 is compared, and most of noise is also extracted.Figure 10 is will be after the synthesis of remaining singular value tie element
Data, useful signal obtained significant reinforcement, and signal originally capped to recognize also is restored.This example is said
Bright, for laterally irregular noise, the present invention also has good adaptability.
The above-mentioned description to embodiment is for that can understand and apply the invention convenient for those skilled in the art.
Person skilled in the art obviously easily can make various modifications to above-described embodiment, and described herein general
Principle is applied in other embodiments without having to go through creative labor.Therefore, the present invention is not limited to the above embodiments, ability
Field technique personnel announcement according to the present invention, the improvement made for the present invention and modification all should be in protection scope of the present invention
Within.
Claims (4)
1. a kind of adaptive Coherent Noise in GPR Record denoising method, includes the following steps:
(1) for original Coherent Noise in GPR Record, the data in its preceding k nanosecond is intercepted and determine the standard track of data, k is greater than 0
Real number；
The original Coherent Noise in GPR Record is the data matrix of n × m size, and n and m respectively correspond sampling for the time and spatially
Points；The scanning road is the radar data of any sampled point on corresponding detected object surface, the i.e. corresponding detection of the standard track
The radar data of a certain sampled point in subject surface, the sampled point are random or artificial specified；
(2) it for any scanning road, calculates cross-correlation function within preceding k nanosecond of the scanning road and standard track and takes its maximum
Value；
(3) all scanning roads are traversed according to step (2), obtains the first arrival curve being made of each scanning road cross-correlation function maximum value
L, and then calculate the high D of root mean square of first arrival curve L；
(4) singular value number N corresponding with coherent noise is calculated according to the following formula；
N=ROUND { 0.2634D+1.3086 }
Wherein: ROUND { } is the function that rounds up；
(5) singular value decomposition is carried out to original Coherent Noise in GPR Record, the top n data component after abnormal value elimination decomposition will remain
Remaining data component completes denoising process after recombining.
2. adaptive Coherent Noise in GPR Record denoising method according to claim 1, it is characterised in that: in the step (2)
Cross-correlation function expression formula it is as follows:
Wherein: Rj(τ) is that jth scans the cross-correlation function of road and standard track within preceding k nanosecond, x0It (t) is the standard track of t moment,
T and τ is time sampling serial number and 1≤τ≤K, K are the sampling number of radar data in preceding k nanosecond, xj *(t- τ) is xj(t-τ)
Conjugate complex number, xj(t- τ) is that the jth at t- τ moment scans road, and j is natural number and 1≤j≤m.
3. adaptive Coherent Noise in GPR Record denoising method according to claim 1, it is characterised in that: in the step (3)
The high D of root mean square of first arrival curve L is calculated according to the following formula:
Wherein: d (j) is that jth scans the cross-correlation function maximum value of road and standard track within preceding k nanosecond,For first arrival curve L's
Average value.
4. adaptive Coherent Noise in GPR Record denoising method according to claim 1, it is characterised in that: k=5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710575039.XA CN107450054B (en) | 2017-07-14 | 2017-07-14 | A kind of adaptive Coherent Noise in GPR Record denoising method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710575039.XA CN107450054B (en) | 2017-07-14 | 2017-07-14 | A kind of adaptive Coherent Noise in GPR Record denoising method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107450054A CN107450054A (en) | 2017-12-08 |
| CN107450054B true CN107450054B (en) | 2019-09-10 |
Family
ID=60488055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710575039.XA Active CN107450054B (en) | 2017-07-14 | 2017-07-14 | A kind of adaptive Coherent Noise in GPR Record denoising method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107450054B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109541586A (en) * | 2018-11-08 | 2019-03-29 | 山东交通学院 | A kind of Ground Penetrating Radar quick detection road engineering concrete cushion thickness analysis method |
| CN112241005B (en) * | 2019-07-19 | 2024-05-31 | 杭州海康威视数字技术股份有限公司 | Compression method, device and storage medium of radar detection data |
| CN110673138B (en) * | 2019-10-09 | 2022-08-23 | 国家电网有限公司 | Ground penetrating radar image processing method based on singular value decomposition and fuzzy C mean value method |
| CN110850409B (en) * | 2019-10-22 | 2022-04-05 | 南京理工大学 | Double-station synthetic aperture radar imaging method based on time reversal |
| CN112285791B (en) | 2020-03-30 | 2022-01-25 | 中国科学院地质与地球物理研究所 | Ground penetrating radar electromagnetic field data processing method, device and medium |
| CN113030954A (en) * | 2021-04-20 | 2021-06-25 | 吉林大学 | Ground penetrating radar data SVD distributed algorithm based on Flink |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105403867A (en) * | 2015-12-04 | 2016-03-16 | 河海大学 | Compression-sensing-based signal reconstruction and de-noising method of ground penetrating radar |
| CN105527617A (en) * | 2016-02-06 | 2016-04-27 | 哈尔滨工业大学 | Ground penetrating radar data background removing method based on robust principal component analysis |
| CN106199532A (en) * | 2016-06-27 | 2016-12-07 | 中国科学院电子学研究所 | Gpr Signal noise-reduction method based on mixing Fourier wavelet analysis |
| WO2016205216A1 (en) * | 2015-06-15 | 2016-12-22 | Humatics Corporation | High precision subsurface imaging and location mapping with time of flight measurement systems |
| US9851438B2 (en) * | 2014-10-23 | 2017-12-26 | King Fahd University Of Petroleum And Minerals | Method and system to identify and estimate relaxation frequencies for ground penetrating radars |
-
2017
- 2017-07-14 CN CN201710575039.XA patent/CN107450054B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9851438B2 (en) * | 2014-10-23 | 2017-12-26 | King Fahd University Of Petroleum And Minerals | Method and system to identify and estimate relaxation frequencies for ground penetrating radars |
| WO2016205216A1 (en) * | 2015-06-15 | 2016-12-22 | Humatics Corporation | High precision subsurface imaging and location mapping with time of flight measurement systems |
| CN105403867A (en) * | 2015-12-04 | 2016-03-16 | 河海大学 | Compression-sensing-based signal reconstruction and de-noising method of ground penetrating radar |
| CN105527617A (en) * | 2016-02-06 | 2016-04-27 | 哈尔滨工业大学 | Ground penetrating radar data background removing method based on robust principal component analysis |
| CN106199532A (en) * | 2016-06-27 | 2016-12-07 | 中国科学院电子学研究所 | Gpr Signal noise-reduction method based on mixing Fourier wavelet analysis |
Non-Patent Citations (2)
| Title |
|---|
| Clutter reduction and detection of landmine objects in ground penetrating radar data using singular value decomposition (SVD);Fawzy Abujarad等;《IEEE UXO&Landmine Detection》;20050531;全文 |
| 小波变换在探地雷达弱信号去噪中的研究;王超等;《物探与化探》;20150430;第39卷(第2期);全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107450054A (en) | 2017-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107450054B (en) | A kind of adaptive Coherent Noise in GPR Record denoising method | |
| CN106443802B (en) | A kind of quantitatively characterizing method and system of the Electrical imaging reservoir fracture hole body based on morphologic filtering | |
| CN106468770B (en) | Nearly optimal radar target detection method under K Distribution Clutter plus noise | |
| CN101727662B (en) | SAR image nonlocal mean value speckle filtering method | |
| Huang et al. | A novel method for speckle noise reduction and ship target detection in SAR images | |
| CN102722892B (en) | SAR (synthetic aperture radar) image change detection method based on low-rank matrix factorization | |
| CN102609701B (en) | Remote sensing detection method based on optimal scale for high-resolution SAR (synthetic aperture radar) | |
| CN106093890A (en) | The method setting up residual clutter suppression radar residual spur | |
| CN103971364B (en) | Remote sensing image variation detecting method on basis of weighted Gabor wavelet characteristics and two-stage clusters | |
| CN105510964A (en) | Seismic recognition method of low-order strike-slip faults in complex structural areas | |
| CN102830404B (en) | Method for identifying laser imaging radar ground target based on range profile | |
| CN104361582A (en) | Method of detecting flood disaster changes through object-level high-resolution SAR (synthetic aperture radar) images | |
| CN104155638A (en) | Tree species classification method based on LiDAR (Light Detection and Ranging) false-vertical waveform model | |
| CN105866776B (en) | The high quality dynamic PS of ground SAR a kind of clicks selection method | |
| CN101329402A (en) | Multi-dimension SAR image edge detection method based on improved Wedgelet | |
| CN101685158B (en) | Hidden Markov tree model based method for de-noising SAR image | |
| CN103400383A (en) | SAR (synthetic aperture radar) image change detection method based on NSCT (non-subsampled contourlet transform) and compressed projection | |
| CN101587186A (en) | Characteristic extraction method of radar in-pulse modulation signals | |
| CN110110675A (en) | A kind of wavelet field of combination of edge information divides shape infrared cirrus detection method | |
| CN104200471A (en) | SAR image change detection method based on adaptive weight image fusion | |
| CN105675986A (en) | Estimation of arrival angle of narrowband frequency modulation signal based on time frequency analysis during data loss | |
| CN105160648A (en) | Radar target and shadow segmentation method based on wavelet and constant false alarm rate | |
| Li et al. | Filtering airborne lidar data by modified white top-hat transform with directional edge constraints | |
| CN106156758A (en) | A kind of tidal saltmarsh method in SAR coast chart picture | |
| CH716789B1 (en) | Method for extracting hidden geological structural information from a remote sensing image pair. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: No. 928, yuhangtang Road, Xihu District, Hangzhou City, Zhejiang Province 310011 Patentee after: Zhejiang shuzhijiaoyuan Technology Co.,Ltd. Address before: No. 928, yuhangtang Road, Xihu District, Hangzhou City, Zhejiang Province 310011 Patentee before: ZHEJIANG PROVINCIAL INSTITUTE OF COMMUNICATIONS PLANNING DESIGN & RESEARCH Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP03 | Change of name, title or address |
Address after: No. 928, yuhangtang Road, Xihu District, Hangzhou City, Zhejiang Province 310011 Patentee after: ZHEJIANG PROVINCIAL INSTITUTE OF COMMUNICATIONS PLANNING DESIGN & RESEARCH Co.,Ltd. Address before: 310006 No. 89 West Ring Road, Hangzhou, Zhejiang, Xihu District Patentee before: ZHEJIANG PROVINCIAL INSTITUTE OF COMMUNICATIONS PLANNING, DESIGN & RESEARCH |
|
| CP03 | Change of name, title or address |