CN105654477A - Detecting and positioning method for strip-shaped underground object - Google Patents

Detecting and positioning method for strip-shaped underground object Download PDF

Info

Publication number
CN105654477A
CN105654477A CN201510998761.5A CN201510998761A CN105654477A CN 105654477 A CN105654477 A CN 105654477A CN 201510998761 A CN201510998761 A CN 201510998761A CN 105654477 A CN105654477 A CN 105654477A
Authority
CN
China
Prior art keywords
massif
remote sensing
infrared remote
background
target place
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.)
Granted
Application number
CN201510998761.5A
Other languages
Chinese (zh)
Other versions
CN105654477B (en
Inventor
张天序
马文绚
曹少平
郝龙伟
黄正华
杨柳
桑农
王岳环
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Original Assignee
Huazhong University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology filed Critical Huazhong University of Science and Technology
Priority to CN201510998761.5A priority Critical patent/CN105654477B/en
Publication of CN105654477A publication Critical patent/CN105654477A/en
Application granted granted Critical
Publication of CN105654477B publication Critical patent/CN105654477B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10048—Infrared image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30184—Infrastructure

Abstract

The invention discloses a detecting and positioning method for a strip-shaped underground object. The detecting and positioning method comprises the steps of (1), establishing a simulated mountain background heat radiation field in which the strip-shaped underground object exists; (2) performing linear mapping between the simulated mountain background heat radiation field and a mountain infrared remote sensing image in which the object exists, thereby obtaining a mountain background heat radiation model; and (3) by means of the mountain background heat radiation model obtained in the step (2), filtering the simulated mountain background heat radiation from the mountain infrared remote sensing image in which the object exists, and obtaining a strip-shaped underground object information field. The detecting and positioning method settles problems of only shallow-layer object detection and positioning and low identification rate for the strip-shaped underground object in prior art. In a distorted ultra-weak signal of the strip-shaped underground target after interferences of the mountain and environment heat radiation, the mountain background information field is filtered and the strip-shaped underground object is obtained. The detecting and positioning method has advantages of overcoming difficulty in obtaining multi-temporal infrared image and remarkably improving identification accuracy in strip-shaped underground object detection.

Description

The detecting and positioning method of a kind of ribbon buried target
Technical field
The invention belongs to the crossing domain of thermophysics, physical geography, pattern recognition, remote sensing and information processing, more specifically, it relates to the method for the Detection location of a kind of ribbon buried target.
Background technology
Ribbon buried target is the three-dimensional target of a kind of ribbon, comprises civil air defense works, subterranean river, tunnel, underground, nature underground cavern etc. There is very important status in tunnel in massif in civilian, especially vcehicular tunnel and railway tunnel can pass massif, the length of the road not only greatly shortened saves the time that people spend during the journey, also reduce and build sky way, the required a large amount of human and material resources of railway, and for automobile, the security in tunnel is far longer than the security of sky way. Subterranean river, naturally descend grottoes detection, find can better protect national resource. Therefore, it is necessary to carry out the research of ribbon underground objects detection, location in the environment of mountain region.
Conventional remote sensing detection only realizes the conditional object on earth's surface or on the water surface, also has the shallow-layer target acquisition of signal process for being only subject to air dielectric decay; But multiple medium faces, compared with deep layer remote sensing detection, the multiple distortion process that target signal is subject to air, solid, the Multiple decrements process of water body medium and the characteristic of medium itself and air, solid, water body medium. Final signal becomes very faint, cannot detect by existing ordinary method at all. Ribbon buried target buried depth has even several hundred meters, tens meters of rice up to a hundred usually, and its signal shows through the modulation of rock soil media, belongs to the buried target of deep layer.
At present, being mostly manually detect on the spot to the method for ribbon buried target mountain region ribbon facility detection both at home and abroad, lack the effective means of remote sensing detection, this kind of method is very time-consuming and needs a large amount of manpowers. Part uses the ribbon underground objects detection technology of infrared imaging, but is mostly directly in the enterprising pedestrian's work interpretation of infrared figure, and the filtering of simple earth's surface interference, lacks model, can not obtain degree of depth information.Overseas utilization machine load medium wave and long-wave infrared sweep the research of sensor detection shallow underground target (such as ancient tomb), but have no the relevant report of degree of depth ribbon buried target when utilizing remote sensing image to carry out mountain region (such as civil air defense works, subterranean river, tunnel, underground, nature underground cavern etc.). Correspondingly, this area is needed badly and is found a kind of method being applicable to be carried out by ribbon buried target under massif background radiation model retrains Detection location.
Summary of the invention
For above defect or the deficiency of prior art, the present invention proposes the method for a kind of ribbon underground objects detection location, overcome the problem that the infrared figure of multidate obtains difficulty, solve existing for the low problem in shallow-layer target detection location and ribbon underground objects detection recognition rate. Simulation softwares and thermal radiation theoretical basis is utilized to set up massif background radiation model, model constrained lower infrared remote sensing image is carried out background filtering process above-mentioned, it is exactly visually successively open stratum, makes the radiation field progressively transparence of ribbon buried target.
For achieving the above object, the present invention proposes the method for a kind of ribbon underground objects detection location, it is characterised in that, described method specifically comprises the following steps: 1, the detecting and positioning method of a kind of ribbon buried target, it is characterized in that, described method comprises the following steps:
(1) the emulation massif background heat radiation field at ribbon buried target place is set up;
(2) described emulation massif background heat radiation field and target place massif infrared remote sensing figure are carried out linear mapping, obtain massif background radiation model;
(3) utilizing the massif background radiation model that step (2) obtains, from the massif infrared remote sensing figure of target place, filtering emulation massif background heat radiation field, obtains ribbon buried target information field.
As preferred further, described step (1) specifically comprises:
(1.1) the sea level elevation data information of ribbon buried target place massif is obtained;
(1.2) according to the sea level elevation data information of step (1.1), extract (x, y, the z) three-dimensional coordinate of the point building massif, build massif realistic model;
(1.3) mountain model that step (1.2) builds is carried out finite element free mesh;
(1.4) mountain model after above-mentioned division is carried out the setting of load boundary condition, thermal conductivity K according to massif and the thermal convection rate �� of massif-air arranges the parameter of heat trnasfer, solve the thermo parameters method estimating to obtain massif, and then, the thermo parameters method of massif is carried out gray-scale map mapping and temperature resolution adjustment, obtains emulation massif background heat radiation field.
As preferred further, described step (2) specifically comprises:
(2.1) target place massif infrared remote sensing figure is carried out cluster analysis, each class carries out emulating the linear mapping of massif background heat radiation field and target place massif infrared remote sensing figure successively;
(2.2) utilize the gray-scale value of the target place massif each class of infrared remote sensing figure after cluster analysis that described emulation massif background heat radiation field is carried out linear mapping treatment for correcting, obtain massif background radiation model.
As further preferred, described target place massif infrared remote sensing figure is carried out cluster analysis, specifically comprises:
(2.1.1) (x, y, the g) of each pixel that choose target place massif infrared remote sensing figure, as a sample, wherein (x, y) represents the ranks coordinate of pixel, and g represents grey scale pixel value;
(2.1.2) the distance ratio of all samples is calculated;
Wherein, i-th sample biDistance ratioS is the number of sample, and d () represents the distance of two samples, i �� j, j �� u;
(2.1.3) chosen distance compares ViMinimum sample biAs first class heart m1, the sequence number q=1 of juxtaposition class;
(2.1.4) to p=1,2 ..., q class, by bi, i=1,2 ..., s is assigned to from its nearest class, and upgrades the class heartNpIt is the sample number of p class, bpiRepresent i-th sample in p class;
(2.1.5) putting q=q+1, if q is greater than k, then algorithm stops; Otherwise continue next step; Wherein k represents category division number total for described target place massif infrared remote sensing figure;
(2.1.6) select to makeMinimum some biAs next class heart mqBest initial center point, proceed to (2.1.4).
As preferred further, described massif background radiation model O is specially:
O = I - I l I h - I l ( O h - O l ) + O l
Wherein, OhFor the maximum brightness gray-scale value of each class in the massif infrared remote sensing figure of target place, OlFor the minimum brightness gray-scale value of each class in the massif infrared remote sensing figure of target place, I is the gray-scale value of emulation massif background heat radiation field, IlFor the minimum brightness gray-scale value of the corresponding each class region of emulation massif background heat radiation field, IhFor the maximum brightness gray-scale value of the corresponding each class region of emulation massif background heat radiation field.
As preferred further, described step (3) specifically comprises:
(3.1) by after accumulation statistics of histogram, the strong jamming background of preliminary filtering target place massif infrared remote sensing figure;
(3.2) utilizing the massif background radiation model that step (2) obtains, the massif background information field of further filtering target place massif infrared remote sensing figure, obtains ribbon buried target information field T (x, y, z, t).
As preferred further, described step (3.1) specifically comprises:
First, pixel discrete gray levels r in estimating target place massif infrared remote sensing figurekProbability density function Pr(rk), have following formula to set up:
P r ( r k ) = n k n ( 0 ≤ r k ≤ l ) k = 0 , 1 , 2 ... l - 1
Wherein, r is pixel grayscale, nkFor there is r in target place massif infrared remote sensing figurekThe pixel count of gray scale, n is pixel count sum in the massif infrared remote sensing figure of target place, and l is the overall number of pixel discrete gray levels;
Further, accumulative grey level histogram, will meet the r of following formula successivelykAlso filtering after finding out, thus complete the preliminary filtering of target place massif infrared remote sensing figure strong jamming background:
Pr(r > rk)��P
Wherein, P% is strong jamming background shared area ratio in the massif infrared remote sensing figure of target place.
As preferred further, described ribbon buried target information field T (x, y, z, t) is specially:
T(x,y,z,t)��BT(x,y,z,t)-k*O
Wherein, the target place massif infrared remote sensing figure that BT (x, y, z, t) is certain moment t target location (x, y, z), O is massif background radiation model, and k is the adjustability coefficients of background radiation field.
In general, according to point of the present invention above technical scheme compared with prior art, mainly possess following technological merit:
1, the present patent application proposes the method for a kind of ribbon underground objects detection location, solve existing for the low problem in shallow-layer target detection location and ribbon underground objects detection recognition rate, grey scale mapping is carried out under the infrared remote sensing figure of shooting instructs, the emulation maximum degree of radiation model is radiated close to real massif background heat, finally target place massif infrared remote sensing image is carried out background filtering process, it is determined that the position of ribbon buried target;
2, in addition, instant invention overcomes the problem that the infrared figure of multidate obtains difficulty, by the method performed in the present invention, visually successively open stratum, make the radiation field progressively transparence of ribbon buried target, the present invention can while guaranteeing that effective ribbon buried target is located, the accuracy that improve ribbon underground objects detection location of maximum possible;
, there is not too much computation complexity in the method 3, proposed according to the present invention, is convenient to manipulation, thus there is certain exploitativeness and practical reference value.
Accompanying drawing explanation
Fig. 1 is schematic flow sheet of the present invention;
Fig. 2 is place, mouth tunnel, Desheng massif infrared remote sensing figure;
Fig. 3 (a) is the three-dimensional elevation data figure of place, mouth tunnel, Desheng massif
Fig. 3 (b) is place, mouth tunnel, Desheng massif level line schematic diagram;
Fig. 3 (c) is place, mouth tunnel, Desheng massif analogous diagram;
Fig. 3 (d) is place, mouth tunnel, Desheng massif stress and strain model figure;
Fig. 4 is place, mouth tunnel, Desheng massif emulation massif background heat radiation field;
Fig. 5 is that the infrared remote sensing figure containing mouth tunnel, Desheng carries out scratching figure according to realistic model;
Fig. 6 is place, mouth tunnel, Desheng infrared remote sensing figure cluster signature;
Fig. 7 for be classify to massif emulate radiation model map correction result figure;
Fig. 8 is the infrared band of each 400m in route left and right, tunnel;
Fig. 9 is the result that true infrared band carries out classified filtering process;
Figure 10 is the signature of part tunnel path.
Embodiment
In order to make the object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated. It is to be understood that specific embodiment described herein is only in order to explain the present invention, it is not intended to limit the present invention.
As shown in Figure 1, the present invention proposes the method for the Detection location of a kind of ribbon buried target, and concrete steps are as follows:
(1) the emulation massif background heat radiation field of ribbon buried target place massif is set up. This experiment foundation has mountain model and ribbon target model, and by thermophysics it will be seen that the temperature relative constancy of massif, the temperature of ribbon buried target is higher or lower than the relative constancy temperature of massif. Massif radiation field is a relatively stable amount, it is possible to Modling model obtains this stable amount. First the sea level elevation data of ribbon buried target place massif are obtained by google, utilize level line that sea level elevation data are carried out abstraction, Primary Construction mountain model, utilize simulation software and the whole mountain model of contour data construct, and to its grid division, final condition be set, solve, obtain emulation massif background heat radiation field.
(2) target place massif infrared remote sensing figure instruct under emulation massif background heat radiation field map. (1) the emulation massif background heat radiation field obtained in is 8 infrared figure, under 16 target place massif infrared remote sensing figure instruct, massif background heat radiates 8 and maps to 16 bit linear so that the massif maximum degree of background radiation model obtained after mapping is close to real massif background heat radiation field.
(3) the massif background radiation model obtained after utilizing mapping carries out background filtering. Target place massif infrared remote sensing figure is equivalent to ribbon buried target information field and the superposition of massif background information field, (2) the massif background radiation model obtained after being mapped in, filtering massif background information field from the massif infrared remote sensing figure of target place, obtain ribbon buried target information field, reach the object that detection identifies ribbon buried target.
Preferably, step (1) relates to and utilizes simulation software that ribbon buried target place massif is carried out simulation modeling, specifically comprises:
(1.1) altitude figures of the massif at ribbon buried target place obtains
Calculate the longitude and latitude of ribbon buried target gangway, obtain being obtained the region of altitude figures, write software and download from google, obtain the sea level elevation data information of massif.
(1.2) foundation of mountain model
The structure of ANSYS geometry mountain model is formed with point, line, surface and body, according to above-mentioned altitude information, extract (the x of the point building massif, y, z) three-dimensional coordinate, point is the basic geometry building geometric model, and the establishment of massif generates closed curve by key point, closed curve generates plane, then surrounds geometry massif by the curved surface closed. And geometry massif just constitutes whole mountain model.
(1.3) the FEM meshing step of mountain model
ANSYS FEM meshing carries out the most important step of numerical simulation analysis, owing to the model of massif is not regular, therefore free mesh is carried out, it is free to automatically generate on the whole trilateral or tetrahedral grid, body generates tetrahedral grid automatically, manually carries out the control of smart dimensions simultaneously.
(1.4) mountain model final condition is arranged and solution procedure
Massif after above-mentioned stress and strain model is carried out the setting of load boundary condition, utilize the basic heat trnasfer physical basis of massif thermal conduction and massif-air thermal convection, thermal conductivity K according to massif and the thermal convection rate �� of massif-air arranges the parameter of heat trnasfer, the thermo parameters method calculating massif is solved through Ansys, finally the thermo parameters method of massif is carried out the mapping of gray-scale map and the adjustment of temperature resolution, finally obtain emulation massif background heat radiation field.
Preferably, step (2) relates to emulation massif background heat radiation field from 8 to 16 bit linear mappings. Emulation massif background heat radiation field (8bit image) needs to carry out mapping process with the thermal radiation field (16bit image) of target place massif infrared remote sensing figure, when the thermal radiation field of model and the thermal radiation field changes in distribution of true massif are roughly the same, ensure that model thermal radiation field is more close to the thermal radiation field of true massif. Specifically comprise:
(2.1) massif infrared remote sensing figure in target place is to the differentiation of interference pattern
Due to the true massif background more complicated at ribbon buried target place, compared with emulation massif background heat radiation field, the thermal radiation field of true massif background and target place massif infrared remote sensing figure are subject to the impact of many extraneous factors, such as shadow region, vegetation on massif, the impact of the factors such as exposed rock, can there is relatively big error in overall thermal radiation field and massif infrared remote sensing figure (16bit image) mapping of target place from emulation massif background heat radiation field (8bit image) to massif real background, it is thus desirable to target place massif infrared remote sensing figure is carried out cluster analysis, in each class taking target place massif infrared remote sensing figure be instruct, emulation massif background heat radiation field is mapped, detailed process is as follows:
Each class spatially relative close of target place massif infrared remote sensing figure, roughly the same on gray-scale value, therefore, (x, y, g) can be chosen and as proper vector, it is carried out cluster, wherein (x, y) represents the ranks coordinate of pixel, and g represents the gray-scale value of pixel. The cluster algorithm adopted is as follows:
(2.1.1) (x, y, the g) of each pixel that choose target place massif infrared remote sensing figure, as a sample, wherein (x, y) represents the ranks coordinate of pixel, and g represents grey scale pixel value;
(2.1.1) (x, y, the g) of each pixel that choose target place massif infrared remote sensing figure, as a sample, wherein (x, y) represents the ranks coordinate of pixel, and g represents grey scale pixel value;
(2.1.2) the distance ratio of all samples is calculated;
Wherein, i-th sample biDistance ratioS is the number of sample, and d () represents the distance of two samples, i �� j, j �� u;
(2.1.3) chosen distance compares ViMinimum sample biAs first class heart m1, the sequence number q=1 of juxtaposition class;
(2.1.4) to p=1,2 ..., q class, by bi, i=1,2 ..., s is assigned to from its nearest class, and upgrades the class heartNpIt is the sample number of p class, bpiRepresent i-th sample in p class;
(2.1.5) putting q=q+1, if q is greater than k, then algorithm stops; Otherwise continue next step; Wherein k represents category division number total for described target place massif infrared remote sensing figure;
(2.1.6) select to makeMinimum some biAs next class heart mqBest initial center point, proceed to (2.1.4).
(2.2) matching and correlation of massif background heat radiation field gray scale scope and target place massif infrared remote sensing figure distribution range is emulated
The gray-scale value of emulation massif background heat radiation field carries out linear mapping treatment for correcting according to the gray scale maximum value of each class obtained in (2.1) and minimum gray value. Described massif background radiation model O is specially:
O = I - I l I h - I l ( O h - O l ) + O l
Wherein, OhFor the maximum brightness gray-scale value of each class in the massif infrared remote sensing figure of target place, OlFor the minimum brightness gray-scale value of each class in the massif infrared remote sensing figure of target place, I is the gray-scale value of emulation massif background heat radiation field, IlFor the minimum brightness gray-scale value of the corresponding each class region of emulation massif background heat radiation field, IhFor the maximum brightness gray-scale value of the corresponding each class region of emulation massif background heat radiation field.
Preferably, step (3) relates to and utilizes massif background radiation model, in the information field that ribbon buried target and massif background superposition produce, filtering massif background information field, obtains the fuzzy message field of ribbon buried target after stratum is modulated. Specifically comprise:
(3.1) strong jamming of target place massif infrared remote sensing figure is removed
Object observing place massif infrared remote sensing figure, it is possible to find that grey value profile is strong jamming background in the background area of upper section. Owing to target place massif infrared remote sensing figure is subject to the impact of many extraneous factors, such as shadow region, the vegetation on massif, exposed rock etc., it is thus desirable to carry out region constraint process, pass through statistics with histogram, carrying out threshold process, reject the interference of extraneous factor, detailed process is as follows:
First, pixel discrete gray levels r in estimating target place massif infrared remote sensing figurekProbability density function Pr(rk), have following formula to set up:
P r ( r k ) = n k n ( 0 ≤ r k ≤ l ) k = 0 , 1 , 2 ... l - 1
Wherein, r is pixel grayscale, nkFor there is r in target place massif infrared remote sensing figurekThe pixel count of gray scale, n is pixel count sum in the massif infrared remote sensing figure of target place, and l is the overall number of pixel discrete gray levels;
Further, accumulative grey level histogram, will meet the r of following formula successivelyk(being the strong jamming background area needing filtering) finds out rear also filtering, thus completes the preliminary filtering of target place massif infrared remote sensing figure strong jamming background:
Pr(r > rk)��P
Wherein, P% is strong jamming background shared area ratio in the massif infrared remote sensing figure of target place.
(3.2) target place massif infrared remote sensing figure background filtering transparence step.
Utilize the massif background radiation model that obtains as the background of target place massif infrared remote sensing figure, owing to the target thermal radiation in massif meets the mathematical model of thermal conduction, it is assumed that at certain moment t0Target location (x0,y0,z0) thermal radiation curved surface be BT (x, y, z, t), i.e. target place massif infrared remote sensing figure, be specially:
BT (x, y, z, t)=O+T (x, y, z, t) * Rb(x,y,z,t)+A(x,y,z,t)+��(x,y,z,t)
Wherein, O is multiple dielectric radiation field, i.e. massif background radiation model, T (x, y, z, t) * Rb(x, y, z, t) ribbon buried target information field for having been distorted by multiple dielectric, the radiation quantity that �� (x, y, z, t) scatters and disappears for water body/ground body and air contact face, A (x, y, z, t) is atmospheric environment distortion effects.
Owing to the impact of ribbon buried target information field produces primarily of B (x, y, z, t), therefore there is following formula:
T(x,y,z,t)*Rb(x,y,z,t)��BT(x,y,z,t)-k*O
Wherein T (x, y, z, t) * Rb(x, y, z, t) ribbon buried target information field for having been distorted by multiple dielectric, BT (x, y, z, t) is target place massif infrared remote sensing figure, and O is massif background radiation model, and k is the adjustability coefficients of background radiation field.
And then can obtain, preliminary observable ribbon buried target information field is: T (x, y, z, t) �� BT (x, y, z, t)-k*O.
Thus the image of transparence after finally obtaining the massif infrared remote sensing figure background filtering of target place, it is the ribbon buried target information field of location detection.
Further, the inventive method, for place, mouth tunnel, the Desheng massif infrared remote sensing figure obtained, is described by the present invention, specific as follows:
(1) the background heat radiation field of place, tunnel massif is set up, and comprises following sub-step:
(1.1) altitude figures of the massif at place, tunnel obtains
As shown in Figure 2, for the infrared figure in region at place, mouth tunnel, Desheng, the latitude and longitude information of localized tunnel gangway, write software, downloading the elevation data of respective regions from google, be one piece of more complete massif of ratio in figure in circle, we choose the elevation data in this region, obtain contour data, carry out massif modeling.
(1.2) establishment step of mountain model
(1.2.1) by analyzing place, mouth tunnel, Desheng physical features, in tunnel distribution range, the small hill that one piece of sunkly with one piece of projection is come can be divided into. We choose protruding next small hill and carry out modeling, the long about 1650m of small hill, wide about 1400m, account for ribbon buried target about 2/3 (the long about 2700m in mouth tunnel, Desheng), and its three-dimensional elevation data figure is as shown in Fig. 3 (a).
(1.2.2) structure of ANSYS geometry mountain model is formed with point, line, surface and body, and point (coordinate) is the basis building geometric model. So according to above-mentioned altitude information, the establishment of geometry massif is the curve having crucial dot generation closed, and closed curve generates plane, then surrounds geometry massif by the curved surface closed. And geometry massif just constitutes whole mountain model, as shown in Fig. 3 (b) and Fig. 3 (c).
(1.2) the FEM meshing step of mountain model
Anasy FEM meshing carries out the most important step of numerical simulation analysis, owing to the model of massif is not regular, therefore free mesh is carried out, it is free to automatically generate on the whole trilateral or tetrahedral grid, body generates tetrahedral grid automatically, manually carries out the control of smart dimensions simultaneously. This example adopts two tetrahedron elements (No. 92 unit), ensures to calculate precision, as shown in Fig. 3 (d).
(1.3) mountain model final condition is arranged and solution procedure
Massif after above-mentioned stress and strain model is carried out the setting of load boundary condition, utilize the basic heat trnasfer physical basis of massif thermal conduction and massif-air thermal convection, thermal conductivity K according to massif and the thermal convection rate �� of massif-air arranges the parameter of heat trnasfer, the thermo parameters method calculating massif is solved through Ansys, finally the thermo parameters method of massif is carried out the mapping of gray-scale map and the adjustment of temperature resolution, finally obtain emulation massif background heat radiation field. In this example, K=3.49Kg/m3, ��=3W/ (m^2.C), bottom surface, mountain region temperature 15 degrees Celsius, air themperature 25 degrees Celsius, solving result is as shown in Figure 4.
(2) target place massif infrared remote sensing figure instruct under massif simulation context thermal radiation map
The emulation massif background heat radiation field obtained in (1) is 8 infrared figure, under 16 target place massif infrared remote sensing figure instruct, emulation massif background heat radiation field 8 maps to 16 bit linear so that the maximum degree of massif background radiation model is close to real massif background heat radiation field.
(2.1) massif infrared remote sensing figure in target place is to the differentiation of interference pattern
Carry out scratching figure according to realistic model by the infrared remote sensing figure containing mouth tunnel, Desheng in Fig. 2, such as Fig. 5, it is that target place massif infrared remote sensing figure scratches figure according to realistic model. Again target place massif infrared remote sensing figure is carried out cluster analysis, here 10 classes are divided by target place massif infrared remote sensing figure altogether, such as Fig. 6, it is target place, place, mouth tunnel, Desheng massif infrared remote sensing figure cluster signature, different classes of marks with different gray-scale values.
(2.2) matching and correlation of mountain model thermal radiation gray scale scope and remote sensing figure distribution range
Such as Fig. 7, it is classify massif is emulated radiation model map correction result figure.
(3) the massif background radiation model after mapping is utilized to carry out background filter step
Infrared remote sensing figure containing tunnel is equivalent to tunnel information field and the superposition of massif background information field, (2) the massif background radiation model after being mapped in, for massif background information field, filtering massif background information field from infrared remote sensing figure, obtain the information field of tunnel target, reach the object that detection identifies tunnel target.
(3.1) strong jamming of the infrared figure of remote sensing is removed
Observe infrared remote sensing figure, it is possible to find that grey value profile is strong jamming background in the background area of upper section. Owing to the thermal radiation of true massif background is subject to the impact of many extraneous factors, such as shadow region, the vegetation on massif, exposed rock etc., it is thus desirable to carry out region constraint process, pass through statistics with histogram, carry out threshold process, reject the interference of extraneous factor, specifically comprise:
First, pixel discrete gray levels r in estimating target place massif infrared remote sensing figurekProbability density function Pr(rk), have following formula to set up:
P r ( r k ) = n k n ( 0 ≤ r k ≤ l ) k = 0 , 1 , 2 ... l - 1
Wherein, r is pixel grayscale, nkFor there is r in target place massif infrared remote sensing figurekThe pixel count of gray scale, n is pixel count sum in the massif infrared remote sensing figure of target place, and l is the overall number of pixel discrete gray levels;
Known be difficult to filtering extraneous factor impact and strong jamming to account for shared area ratio in entire image be P%, then have following formula:
Pr(r > rk)��P
Accumulative grey level histogram, finds corresponding r successivelyk, for needing the background area of further filtering, it is possible to preliminary filtering strong jamming background.
(3.2) target place massif infrared remote sensing figure background filtering transparence step.
Here each class being carried out filtering process respectively, k gets ki(i=1,2 ..., 15). By the priori of the gangway coordinate in mouth tunnel, Desheng, it is possible to obtain the roughly distribution range in tunnel, the region getting each 400m in left and right, bar tunnel carries out filtering, such as Fig. 8, is the infrared band of each 400m in route left and right, tunnel, the region being in Fig. 5 two wire tags. Such as Fig. 9, being the result that infrared remote sensing figure carries out classified filtering process, Figure 10 is the signature of tunnel path.
Those skilled in the art will readily understand; the foregoing is only the better embodiment of the present invention; not in order to limit the present invention, all any amendment, equivalent replacement and improvement etc. done within the spirit and principles in the present invention, all should be included within protection scope of the present invention.

Claims (8)

1. the detecting and positioning method of a ribbon buried target, it is characterised in that, described method comprises the following steps:
(1) the emulation massif background heat radiation field at ribbon buried target place is set up;
(2) described emulation massif background heat radiation field and target place massif infrared remote sensing figure are carried out linear mapping, obtain massif background radiation model;
(3) utilizing the massif background radiation model that step (2) obtains, from the massif infrared remote sensing figure of target place, filtering emulation massif background heat radiation field, obtains ribbon buried target information field.
2. the method for claim 1, it is characterised in that, described step (1) specifically comprises:
(1.1) the sea level elevation data information of ribbon buried target place massif is obtained;
(1.2) according to the sea level elevation data information of step (1.1), extract (x, y, the z) three-dimensional coordinate of the point building massif, build massif realistic model;
(1.3) mountain model that step (1.2) builds is carried out finite element free mesh;
(1.4) mountain model after above-mentioned division is carried out the setting of load boundary condition, thermal conductivity K according to massif and the thermal convection rate �� of massif-air arranges the parameter of heat trnasfer, solve the thermo parameters method estimating to obtain massif, and then, the thermo parameters method of massif is carried out gray-scale map mapping and temperature resolution adjustment, obtains emulation massif background heat radiation field.
3. method as claimed in claim 1 or 2, it is characterised in that, described step (2) specifically comprises:
(2.1) target place massif infrared remote sensing figure is carried out cluster analysis, each class carries out emulating the linear mapping of massif background heat radiation field and target place massif infrared remote sensing figure successively;
(2.2) utilize the gray-scale value of the target place massif each class of infrared remote sensing figure after cluster analysis that described emulation massif background heat radiation field is carried out linear mapping treatment for correcting, obtain massif background radiation model.
4. method as claimed in claim 3, it is characterised in that, described target place massif infrared remote sensing figure is carried out cluster analysis, specifically comprise:
(2.1.1) (x, y, the g) of each pixel that choose target place massif infrared remote sensing figure, as a sample, wherein (x, y) represents the ranks coordinate of pixel, and g represents grey scale pixel value;
(2.1.2) the distance ratio of all samples is calculated;
Wherein, i-th sample biDistance ratioS is the number of sample, and d () represents the distance of two samples, i �� j, j �� u;
(2.1.3) chosen distance compares ViMinimum sample biAs first class heart m1, the sequence number q=1 of juxtaposition class;
(2.1.4) to p=1,2 ..., q class, by bi, i=1,2 ..., s is assigned to from its nearest class, and upgrades the class heartP=1,2 ..., q, NpIt is the sample number of p class, bpiRepresent i-th sample in p class;
(2.1.5) putting q=q+1, if q is greater than k, then algorithm stops; Otherwise continue next step; Wherein k represents category division number total for described target place massif infrared remote sensing figure;
(2.1.6) select to makeMinimum some biAs next class heart mqBest initial center point, proceed to (2.1.4).
5. method as claimed in claim 3, it is characterised in that, described massif background radiation model O is specially:
O = I - I l I h - I l ( O h - O l ) + O l
Wherein, OhFor the maximum brightness gray-scale value of each class in the massif infrared remote sensing figure of target place, OlFor the minimum brightness gray-scale value of each class in the massif infrared remote sensing figure of target place, I is the gray-scale value of emulation massif background heat radiation field, IlFor the minimum brightness gray-scale value of the corresponding each class region of emulation massif background heat radiation field, IhFor the maximum brightness gray-scale value of the corresponding each class region of emulation massif background heat radiation field.
6. method as claimed in claim 1 or 2, it is characterised in that, described step (3) specifically comprises:
(3.1) by after accumulation statistics of histogram, the strong jamming background of preliminary filtering target place massif infrared remote sensing figure;
(3.2) utilizing the massif background radiation model that step (2) obtains, the massif background information field of further filtering target place massif infrared remote sensing figure, obtains ribbon buried target information field T (x, y, z, t).
7. method as claimed in claim 6, it is characterised in that, described step (3.1) specifically comprises:
First, pixel discrete gray levels r in estimating target place massif infrared remote sensing figurekProbability density function Pr(rk), have following formula to set up:
P r ( r k ) = n k n ( 0 ≤ r k ≤ l )
K=0,1,2...l-1
Wherein, r is pixel grayscale, nkFor there is r in target place massif infrared remote sensing figurekThe pixel count of gray scale, n is pixel count sum in the massif infrared remote sensing figure of target place, and l is the overall number of pixel discrete gray levels;
Further, accumulative grey level histogram, will meet the r of following formula successivelykAlso filtering after finding out, thus complete the preliminary filtering of target place massif infrared remote sensing figure strong jamming background:
Pr(r > rk)��P
Wherein, P% is strong jamming background shared area ratio in the massif infrared remote sensing figure of target place.
8. method as claimed in claim 6, it is characterised in that, described ribbon buried target information field T (x, y, z, t) is specially:
T(x,y,z,t)��BT(x,y,z,t)-k*O
Wherein, the target place massif infrared remote sensing figure that BT (x, y, z, t) is certain moment t target location (x, y, z), O is massif background radiation model, and k is the adjustability coefficients of background radiation field.
CN201510998761.5A 2015-12-25 2015-12-25 A kind of detecting and positioning method of ribbon buried target Active CN105654477B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510998761.5A CN105654477B (en) 2015-12-25 2015-12-25 A kind of detecting and positioning method of ribbon buried target

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510998761.5A CN105654477B (en) 2015-12-25 2015-12-25 A kind of detecting and positioning method of ribbon buried target

Publications (2)

Publication Number Publication Date
CN105654477A true CN105654477A (en) 2016-06-08
CN105654477B CN105654477B (en) 2017-11-28

Family

ID=56477979

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510998761.5A Active CN105654477B (en) 2015-12-25 2015-12-25 A kind of detecting and positioning method of ribbon buried target

Country Status (1)

Country Link
CN (1) CN105654477B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108053411A (en) * 2017-12-21 2018-05-18 华中科技大学 A kind of Subaqueous tunnel remote sensing localization method under border heat exchange constraint
CN108090955A (en) * 2017-11-09 2018-05-29 珠海金山网络游戏科技有限公司 The system and method that a kind of virtual scene simulates true mountain model
CN108305257A (en) * 2017-12-27 2018-07-20 华中科技大学 A kind of seabed tunnel remote sensing localization method under thermal background emission constraint
CN111798407A (en) * 2020-05-15 2020-10-20 国网浙江省电力有限公司嘉兴供电公司 Electrified equipment fault diagnosis method based on neural network model
CN114112069A (en) * 2022-01-27 2022-03-01 华中科技大学 Geological-constrained infrared imaging detection method and system for urban deep-buried strip channel

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104484577A (en) * 2014-12-30 2015-04-01 华中科技大学 Detection method based on ridge energy correction for ribbon underground target in mountain land
CN104536009A (en) * 2014-12-30 2015-04-22 华中科技大学 Laser infrared composite ground building recognition and navigation method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104484577A (en) * 2014-12-30 2015-04-01 华中科技大学 Detection method based on ridge energy correction for ribbon underground target in mountain land
CN104536009A (en) * 2014-12-30 2015-04-22 华中科技大学 Laser infrared composite ground building recognition and navigation method

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108090955A (en) * 2017-11-09 2018-05-29 珠海金山网络游戏科技有限公司 The system and method that a kind of virtual scene simulates true mountain model
CN108053411A (en) * 2017-12-21 2018-05-18 华中科技大学 A kind of Subaqueous tunnel remote sensing localization method under border heat exchange constraint
CN108053411B (en) * 2017-12-21 2020-05-19 华中科技大学 Remote sensing detection positioning method for river bottom tunnel under boundary heat exchange constraint
CN108305257A (en) * 2017-12-27 2018-07-20 华中科技大学 A kind of seabed tunnel remote sensing localization method under thermal background emission constraint
CN111798407A (en) * 2020-05-15 2020-10-20 国网浙江省电力有限公司嘉兴供电公司 Electrified equipment fault diagnosis method based on neural network model
CN114112069A (en) * 2022-01-27 2022-03-01 华中科技大学 Geological-constrained infrared imaging detection method and system for urban deep-buried strip channel
CN114112069B (en) * 2022-01-27 2022-04-26 华中科技大学 Geological-constrained infrared imaging detection method and system for urban deep-buried strip channel

Also Published As

Publication number Publication date
CN105654477B (en) 2017-11-28

Similar Documents

Publication Publication Date Title
CN104484577B (en) The detection method of banding buried target in a kind of mountain region based on ridge energy correction
CN105654477A (en) Detecting and positioning method for strip-shaped underground object
CN104637073B (en) It is a kind of based on the banding underground structure detection method for shining upon shadow compensation
CN105426881B (en) Mountain background thermal field model constrained underground heat source daytime remote sensing detection locating method
Jiao et al. Evaluation of four sky view factor algorithms using digital surface and elevation model data
CN105893590B (en) One kind being used for digital Terrain Analysis modeling knowledge case automatic processing method
Clark Small unmanned aerial systems comparative analysis for the application to coastal erosion monitoring
Badapalli et al. Assessment of aeolian desertification near Vedavathi river channel in Central part of Andhra Pradesh: remote sensing approach
Ali et al. Monitoring degradation of wetland areas using satellite imagery and geographic information system techniques.
Shaw-Faulkner et al. Delineation and classification of Karst depressions using LiDAR: fort hood military installation, texas
CN110111376B (en) Coal mining subsidence water area calculation method
Vasconcelos et al. Analysis of spatial–temporal patterns of water table change as a tool for conjunctive water management in the Upper Central Plain of the Chao Phraya River Basin, Thailand
Arrofi et al. Delineating reservoir area using geologic remote sensing analysis, Jizan (Saudi Arabia): Implications for geothermal exploration
CN114112069B (en) Geological-constrained infrared imaging detection method and system for urban deep-buried strip channel
Gokdemir et al. A framework for assessing tunnel drainage-induced impact on terrestrial vegetation
He et al. An integrated use of topography with RSI in gully mapping, Shandong Peninsula, China
CN105606224A (en) Underwater heat source detection and location method based on water body hierarchical filtering
Margarint et al. Assessing the spatial variability of weights of landslide causal factors in different regions from Romania using logistic regression
Karmakar et al. Land Use/Land Cover Change and Environmental Impact Analysis of Ramgarh-Naudiha Region in Uttar Pradesh, India through Geospatial Technology
Elimy et al. Land Use/Land Cover Change Detection Analysis for Eastern Nile Delta Fringes, Egypt
CN108053411A (en) A kind of Subaqueous tunnel remote sensing localization method under border heat exchange constraint
Wang et al. Quantitative Attribution Analysis of the Spatial Differentiation of Gully Erosion in the Black Soil Region of Northeast China
Anyadiegwu et al. Assessment of flood hazard in Owerri West, Imo State, Nigeria using Analytical Hierarchical Process and Index Based Approach
Razak et al. The development of a site suitability map for RBF location using remote sensing and GIS techniques
Liu et al. Identification of Terrace Boundaries from DEMs Using Multidirectional Hill-Shading

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant