CN107818303B - Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory - Google Patents
Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory Download PDFInfo
- Publication number
- CN107818303B CN107818303B CN201710994740.5A CN201710994740A CN107818303B CN 107818303 B CN107818303 B CN 107818303B CN 201710994740 A CN201710994740 A CN 201710994740A CN 107818303 B CN107818303 B CN 107818303B
- Authority
- CN
- China
- Prior art keywords
- image
- difference
- registration
- grid
- ratio
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 24
- 230000011218 segmentation Effects 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 22
- 238000012216 screening Methods 0.000 claims abstract description 13
- 239000013598 vector Substances 0.000 claims description 72
- 238000004364 calculation method Methods 0.000 claims description 31
- 230000008859 change Effects 0.000 claims description 13
- 238000013519 translation Methods 0.000 claims description 12
- 230000000877 morphologic effect Effects 0.000 claims description 10
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 abstract description 14
- 230000002123 temporal effect Effects 0.000 abstract 2
- 239000007789 gas Substances 0.000 description 32
- 238000013459 approach Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000007689 inspection Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000010276 construction Methods 0.000 description 4
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 239000003345 natural gas Substances 0.000 description 2
- 238000006424 Flood reaction Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/32—Determination of transform parameters for the alignment of images, i.e. image registration using correlation-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, a system and a software memory, wherein the method comprises the following steps: s1, acquiring multi-temporal images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle, and cutting an overlapping area of the front temporal image and the rear temporal image according to geographic coordinates; s2, performing brightness equalization processing on the image to be registered and the reference image; s3, segmenting the image to be registered into grids, and registering the grids according to the difference entropy; s4, performing difference enhancement on the result of the grid registration; s5, performing saturation threshold segmentation on the contrast result image; s6, performing binarization segmentation on the grayscale image; s7, vectorizing the denoised binary image, and calculating the barycentric coordinates of the difference polygon; and positioning in the image according to the difference center coordinates, performing manual secondary screening, and judging whether the oil and gas pipeline is damaged at the position. The invention can effectively avoid false detection, avoids fine crushing difference by using a difference clustering merging method and can greatly improve the working efficiency.
Description
Technical Field
The invention relates to the field of processing and analyzing aerial image data of unmanned aerial vehicles, in particular to an automatic contrast analysis method, system and software memory for images of oil and gas pipelines of unmanned aerial vehicles.
Background
Domestic unmanned aerial vehicle technique is used for electric power to patrol line system mostly. The traditional petroleum pipeline inspection work mainly adopts manual inspection, has the defects of high labor intensity, low efficiency, long operation period, low emergency capacity and the like, and particularly can not complete the monitoring task in time in difficult mountainous areas with complex terrain, and the utilization value of historical data of the manual inspection is not high. And the line patrol by using advanced technologies such as emerging unmanned aerial vehicles and the like has the advantages of timely response, high efficiency, no need of manual crossing of difficult areas, objective data, easiness in supervision, reusability and the like. In the fields of natural gas and oil, the unmanned aerial vehicle technology is still in a starting stage. At present, unmanned aerial vehicle patrols linear system and has been equipped with GPS and inertial sensor, possesses the navigation function of independently, can carry out analysis processes by the relevant staff of ground station automatically with the information remote transmission to the ground station of gathering again. For the acquired images, manual interpretation is initially performed, however, the manual interpretation relies on the qualitative analysis of the images by the interpreter, and the interpretation result has a great relationship with the a priori knowledge of the interpreter. In recent years, in order to solve the above problems, various methods for performing image recognition using a computer have been proposed, but the conventional computer aberration recognition method has a certain unsuitability for oil and gas pipeline inspection data.
The traditional image difference identification method generally comprises the following three methods:
1. change detection based on simple algebraic operations
The simple algebraic operation method is used for change detection and is simply described, namely the pixel colors of the two-stage images are correspondingly compared (phase difference, division, directional gradient, edge fitting and the like), and an inherent threshold value is set according to the comparison result, so that small differences are weakened, large differences are enhanced, and a change area is obtained.
The simple algebraic operation method has the advantages of high speed, simple algorithm and stability in many fields, but for aerial images, the method requires the preconditions of strict registration, simple background patterns of target areas and the like.
The image early-stage registration depends on the stability (temperature, humidity and positioning) of aerial shooting equipment and the precision degree of elevation data of a target area, and has certain difficulty.
The background pattern of the target area is simple, and depends on the position of the target area, so that the method is favorable for areas such as gobi deserts and the like, but for jungle areas with luxuriant vegetation or construction sites with exposed ground surfaces brushed by rainwater, a quite large difference is usually obtained, and the false detection rate is increased linearly.
2. Change detection based on image transformation
This method filters out extraneous elements (such as vegetation) based on the direction of emphasis between multiple bands of the image, and then compares the target elements of interest.
This approach is not versatile and requires pre-image registration as is.
3. Feature description based change detection
The method firstly extracts the outline of the image to be compared and compares the outline.
The advantage of this approach is that it is substantially immune to light, and the disadvantages remain numerous, such as the requirement for prior registration of images, and the inability to adapt to changes in the jungle.
In summary, the conventional image difference identification method has the following disadvantages:
1. to reduce false detection rates, highly accurate registration is required for images to be compared
2. The requirement on the consistency of the illumination condition is high
3. The traditional method is not easy to avoid false detection on bushy forests with luxuriant vegetation, construction sites with exposed ground surfaces and the like, but the practical engineering application is that the problems of images to be compared are inherent, the problems that the images need to be compared pay large labor cost and the cost is high are solved, the efficiency caused by automatic comparison of the images which are completely rolled is improved, and the related requirements cannot be met.
The traditional satellite remote sensing image and the digital elevation model have the defects of low resolution, difficult interpretation and the like, and can not meet the high-precision requirement on spatial data under certain conditions. With the continuous development of unmanned aerial vehicle technology, unmanned aerial vehicle high resolution images begin to be applied to the fields of natural gas and oil. For the acquired unmanned aerial vehicle aerial data, because human eyes have better resolution capability for the images, the small images are interpreted by people quickly and accurately. But for large images the manual interpretation is cumbersome. Meanwhile, for a large number of multi-temporal images in the same area and at different time, the repeated labor is more.
Disclosure of Invention
The invention aims to solve the technical problems that in the prior art, the false detection rate is high, and the defects that a bush with luxuriant vegetation and a construction site with an exposed earth surface are difficult to identify are overcome, and provides an unmanned aerial vehicle oil and gas pipeline image automatic comparison analysis method, an unmanned aerial vehicle oil and gas pipeline image automatic comparison analysis system and a software memory.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the invention provides an automatic contrast analysis method for images of an oil and gas pipeline of an unmanned aerial vehicle, which comprises the following steps:
s1, acquiring multi-time-phase images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle, wherein the images all contain corresponding geographic coordinates, cutting overlapping areas of front and rear time-phase images according to the geographic coordinates, taking the overlapping area of the current time phase as an image to be registered, and taking the overlapping area of the previous time phase as a reference image;
s2, performing brightness equalization processing on the image to be registered and the reference image;
s3, dividing the image to be registered into grids, carrying out template registration on each grid and the reference image, and recording registration parameter vectors between each grid and the registration grid in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are difference entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
s4, respectively carrying out ratio calculation on the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, changing the sequence of the grids, recalculating the ratio once, and solving a union set of the results of the two calculations to obtain a difference-enhanced ratio result image;
s5, performing saturation threshold segmentation on the ratio result image to enable the gray value of the changed area in the ratio result image to be unchanged, setting the gray value of the unchanged area to be 0, and graying the result of the saturation threshold segmentation;
s6, performing binarization segmentation on the gray level image, and removing image noise through morphological opening operation;
s7, vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon vector, namely the difference central coordinate; and positioning in the image according to the difference center coordinates, performing manual secondary screening, and judging whether the oil and gas pipeline is damaged at the position.
Further, the method for performing the image luminance equalization processing in step S2 of the present invention is:
converting an image to be registered and a reference image from an RGB space to an HSV space, and respectively carrying out histogram equalization on brightness components of the two images, wherein the formula is as follows:
wherein L is 256 gray levels, h (x)i) Representing the number of pixels of each gray level in the histogram, wherein w and h respectively represent the width and height of the image;
and performing inverse transformation on the image after the histogram equalization to convert the image from the HSV space to the RGB space.
Further, the method for template registration of the grid in step S3 of the present invention specifically includes:
dividing an image to be registered into grids, wherein the size of the grids is 256 to 256 pixels, cutting out a region with the area 9 times that of the grids in a reference image by taking the grids as a center, taking each grid as a template, and performing template registration in the cut region in the reference image; calculating the similarity between the template and the cutting area, wherein the formula is as follows:
T'(x',y')=T(x',y')-1/(w·h)∑x”,y”T(x”,y”)
I'(x+x',y+y')=I(x+x',y+y')-1/(w·h)∑x”,y”I(x+x”,y+y”)
wherein w and h are the width and height of the template image, T (x ', y') is the pixel gray value of the template image at (x ', y'), and I (x + x ', y + y') is the pixel gray value of the reference image at (x + x ', y + y'); calculating the average value of the similarity of the three wave bands of r, g and b as the final template similarity;
and forming an array by the obtained similarity, forming a single-channel gray image by the array, and recording a grid corresponding to the brightest point in the gray image as a registration grid.
Further, the method for calculating the ratio in step S4 of the present invention is:
calculating the ratio between the mesh and the registration mesh according to the following formula:
where X1 and X2 are the pixel values in the two image grids being compared, respectively.
Further, the method for performing saturation threshold segmentation in step S5 of the present invention includes:
converting the ratio result image from an RGB space to an HSV space, extracting a saturation component, and setting a saturation threshold, wherein the region with the saturation component larger than the saturation threshold is a change region, and the gray value of the change region is unchanged; the area with the saturation component smaller than the saturation threshold is an unchanged area, the gray value of the unchanged area is set to be 0, the result of the saturation threshold segmentation is grayed, and the graying formula is as follows:
Y=0.3r+0.59g+0.11b
wherein r, g and b are pixel gray values of red, green and blue wave bands respectively.
Further, the method for performing binary segmentation in step S6 of the present invention is:
and setting a binarization threshold value for the grayscale image, setting the gray value of the pixel in the grayscale image larger than the binarization threshold value as 1, and setting the gray value of the pixel smaller than the binarization threshold value as 0, thereby completing the binarization segmentation of the grayscale image.
Further, the formula for calculating the coordinates of the center of difference in step S7 of the present invention is:
wherein x isi,yiIs the coordinate of pixel i.
The invention provides an unmanned aerial vehicle oil gas pipeline image automatic contrast analysis system, which comprises:
the unmanned aerial vehicle image acquisition unit is used for shooting multi-temporal images along the oil and gas pipeline by the unmanned aerial vehicle;
the image processing and positioning unit is used for processing multi-temporal images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle and detecting the positions of the difference points in the images; the system comprises the following modules:
the overlap region cutting module is used for acquiring multi-time phase images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle, wherein the images all comprise corresponding geographic coordinates, the overlap regions of the front and rear time phase images are cut according to the geographic coordinates, the overlap region of the current time phase is used as an image to be registered, and the overlap region of the previous time phase is used as a reference image;
the brightness balance processing module is used for carrying out brightness balance processing on the image to be registered and the reference image;
the image registration module is used for segmenting an image to be registered into grids, performing template registration on each grid and a reference image, and recording registration parameter vectors between each grid and the registration grids in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are difference entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
the difference enhancement module is used for respectively carrying out ratio calculation on the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, changing the grid sequence to recalculate the ratio once, and solving a union set of the results of the two calculations to obtain a ratio result image for enhancing the difference;
the saturation threshold segmentation module is used for performing saturation threshold segmentation on the ratio result image to enable the gray value of a change area in the ratio result image to be unchanged, the gray value of an unchanged area to be set as 0, and graying the result of the saturation threshold segmentation;
the binary segmentation module is used for carrying out binary segmentation on the gray level image and removing image noise through morphological opening operation;
the difference clustering module is used for vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon, namely the difference central coordinate;
and the damage judging unit is used for positioning in the image according to the difference center coordinates, performing manual secondary screening and judging whether the oil and gas pipeline is damaged at the position.
The invention provides a memory storing unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis software, which executes the following programs:
acquiring multi-time-phase images of the oil and gas pipeline along the way shot by an unmanned aerial vehicle, wherein the images all comprise corresponding geographic coordinates, cutting overlapping areas of front and rear time-phase images according to the geographic coordinates, taking the overlapping area of the current time phase as an image to be registered, and taking the overlapping area of the previous time phase as a reference image;
carrying out brightness equalization processing on the image to be registered and the reference image;
dividing an image to be registered into grids, performing template registration on each grid and a reference image, and recording registration parameter vectors between each grid and the registration grids in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are differential entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
respectively calculating the ratio of the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, recalculating the ratio once by changing the grid sequence, and solving a union set of the results of the two calculations to obtain a ratio result image for enhancing the difference;
performing saturation threshold segmentation on the ratio result image to enable the gray value of a changed area in the ratio result image to be unchanged, setting the gray value of an unchanged area to be 0, and graying the result of the saturation threshold segmentation;
carrying out binarization segmentation on the gray level image, and removing image noise through morphological opening operation;
vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon, namely the difference central coordinate; and positioning in the image according to the difference center coordinates, performing manual secondary screening, and judging whether the oil and gas pipeline is damaged at the position.
The invention has the following beneficial effects: according to the automatic comparison and analysis method, system and software memory for the images of the oil and gas pipelines of the unmanned aerial vehicle, on the basis of the orthoimage obtained by the unmanned aerial vehicle, a difference clustering merging method is adopted to avoid the detailed and fragmented difference in reality, and the method is more suitable for image difference comparison of complicated terrain areas such as jungles with luxurious vegetation and construction sites with bare earth surfaces; the problems of high manual identification cost and low efficiency are solved; the novel image contrast analysis technology of the unmanned aerial vehicle is characterized in that aerial photos are identified by a computer, on the basis of a traditional image difference identification method, a difference entropy rule is adopted to avoid false detection, a difference clustering merging method is used to avoid fine and broken differences in reality, suspicious points are found out and then are handed to manual interpretation and screening, and the working efficiency can be greatly improved.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flowchart of an ortho image contrast analysis calculation based on difference entropy according to an embodiment of the present invention;
FIG. 2 is a segmented mesh image of an embodiment of the present invention;
FIG. 3 is a diagram of a cropped area in a reference image according to an embodiment of the present invention;
FIG. 4 is a template matching overlap array of an embodiment of the present invention;
FIG. 5 is a luminance component histogram distribution of an embodiment of the present invention;
FIG. 6 is a histogram distribution of the stretched luminance components of an embodiment of the present invention;
FIG. 7 is a difference detection color patch of an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, the method for automatically comparing and analyzing images of an oil and gas pipeline of an unmanned aerial vehicle in the embodiment of the invention comprises the following steps:
s1, acquiring multi-time-phase images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle, wherein the images all contain corresponding geographic coordinates, cutting overlapping areas of front and rear time-phase images according to the geographic coordinates, taking the overlapping area of the current time phase as an image to be registered, and taking the overlapping area of the previous time phase as a reference image;
s2, performing brightness equalization processing on the image to be registered and the reference image;
s3, dividing the image to be registered into grids, carrying out template registration on each grid and the reference image, and recording registration parameter vectors between each grid and the registration grid in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are difference entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
each table cell of the disparity entropy vector table is a pair of grid-matched registration parameters, which is of the form:
CA=CB*R*S*T;
wherein, CAI.e. the left graph grid, CBIs a right graph grid, R is a rotation amount, S is a zoom amount, T is a translation amount, and the meaning of the above formula is that C isBAfter R rotation amount, S zooming amount and T translation amount are carried out, the R rotation amount, the S zooming amount and the T translation amount can be just equal to CAAnd (4) overlapping.
If so, there is a difference entropy vector:
V=(R*S*T)
and combining a plurality of difference vectors V into a table, namely the difference entropy vector.
It is emphasized that the table of difference entropy vectors does not need to be completely built because each entropy in the table is not linked to other entropy in the table, and therefore, the cell calculation of each table is completed, and then the difference comparison can be performed on the corresponding matching grids according to V.
S4, respectively calculating the ratio of the corresponding grid and the registration grid thereof according to each unit value V in the difference entropy vector table, changing the grid sequence to recalculate the ratio once again, and solving the union set of the results of the two calculations to obtain a difference-enhanced ratio result image;
s5, performing saturation threshold segmentation on the ratio result image to enable the gray value of the changed area in the ratio result image to be unchanged, setting the gray value of the unchanged area to be 0, and graying the result of the saturation threshold segmentation;
s6, performing binarization segmentation on the gray level image, and removing image noise through morphological opening operation;
s7, vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon vector, namely the difference central coordinate; and positioning in the image according to the difference center coordinates, performing manual secondary screening, and judging whether the oil and gas pipeline is damaged at the position.
In another embodiment of the invention:
the first step is as follows: clipping front and rear time phase data overlapping area according to geographic coordinates
And cutting the front and rear time phase overlapping areas according to the coordinate range in the coordinate file.
The second step is that: front and rear time phase image brightness balance
The image is converted from the RGB space to the HSV space according to equations (1) - (3), and the histogram distribution of the transformed luminance components is shown in fig. 5.
v=max (3)
In the formula, r, g and b are gray values of three wave bands of red, green and blue of the image respectively.
Histogram equalization is performed on the luminance component according to equation (4), and the histogram equalized luminance component is shown in fig. 6:
wherein L is the gray level (256), h (x)i) Representing the number of pixels per gray level in the histogram, w and h represent the width and height of the image, respectively.
After stretching, the image is converted from the HSV space back to the RGB space according to inverse transformation equations (5) - (10).
p=v×(1-s) (7)
q=v×(1-f×s) (8)
t=v×(1-(1-f)×s) (9)
The third step: grid matching
The image to be registered is divided into grids (with a grid size of 256 × 256, which can be flexibly configured according to the definition of the aerial photo) (fig. 2), and a larger area with an area 9 times that of the grid is cut in the reference image with the grid as the center (fig. 3). And (3) taking each grid as a template, and performing template matching in the region of the reference image cropping, wherein the formula (11-13) is an image similarity calculation formula.
T'(x',y')=T(x',y')-1/(w·h)∑x”,y”T(x”,y”) (12)
I'(x+x',y+y')=I(x+x',y+y')-1/(w·h)∑x”,y”I(x+x”,y+y”) (13)
Where w, h is the template image width and height, T (x ', y') is the pixel gray scale value of the template image at (x ', y'), and I (x + x ', y + y') is the pixel gray scale value of the reference image at (x + x ', y + y'). And the similarity mean value of the three wave bands is the final template similarity.
The fourth step: difference calculation
And carrying out ratio calculation on the grid after registration according to the following formula:
where X1 and X2 are the pixel values in the two image grids being compared, respectively.
For pixels with approximately the same pixel value, X1/X2 should approach 1. For pixels with larger pixel values, X1/X2 should be close to 0 or close to infinity. When X1/X2 approaches 0, X2/X1 approaches infinity. Since the ratio is performed for each pixel, the result of the ratio will also be a picture, here named T1.
Due to the great difference between X1 and X2, the ratio may approach to 0, and in order to simplify the "saturation threshold segmentation" of the next step, only the pixels with the ratio approaching infinity are regarded as the changed pixels, and for this reason, the image sequence needs to be changed once, specifically, the left and right images to be compared are exchanged, the ratio is performed again, and a second ratio image is obtained, which is named as T2. . .
The results of the two ratios are then subjected to union operation, i.e., T1 and T2 are subjected to union calculation. The calculation method is as follows:
T0x=max(T1x,T2x)
a union image T0 is obtained, i.e. the ratio result image, and the gray or the areas close to gray are all unchanged areas. The results are shown in FIG. 7.
The fifth step: saturation threshold segmentation
Converting the ratio result image from the RGB space to the HSV space according to the formulas (1) - (3), extracting a saturation component S, setting a threshold, wherein the area larger than the threshold is a changed area, the gray value of the area is unchanged, the area smaller than the threshold is an unchanged area, the gray value of the area is set to be 0, and the formula (15) is a saturation threshold segmentation calculation formula. And after the segmentation is finished, converting the image from the HSV space to the RGB space according to the formulas (5) - (10), and graying according to the formula (16).
In the formula TsIs the saturation threshold.
Y=0.3r+0.59g+0.11b (16)
In the formula, r, g and b are pixel gray values of red, green and blue wave bands respectively.
And setting a gray value binarization segmentation threshold, setting the gray value of the pixel larger than the threshold as 1, and setting the gray value of the pixel smaller than the threshold as 0, completing image binarization, and performing morphological opening operation on the binarization result image to remove noise.
The seventh step: difference detection result vectorization
Vectorizing the result image of the last step, extracting the outline of the white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon according to a formula (17), wherein the barycentric coordinate is the difference center coordinate.
In the formula xi,yiIs the coordinate of pixel i.
The automatic comparison and analysis system for the unmanned aerial vehicle oil and gas pipeline images, provided by the embodiment of the invention, is used for realizing the automatic comparison and analysis method for the unmanned aerial vehicle oil and gas pipeline images, and comprises the following steps:
the unmanned aerial vehicle image acquisition unit is used for shooting multi-temporal images along the oil and gas pipeline by the unmanned aerial vehicle;
the image processing and positioning unit is used for processing multi-temporal images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle and detecting the positions of the difference points in the images; the system comprises the following modules:
the overlap region cutting module is used for acquiring multi-time phase images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle, wherein the images all comprise corresponding geographic coordinates, the overlap regions of the front and rear time phase images are cut according to the geographic coordinates, the overlap region of the current time phase is used as an image to be registered, and the overlap region of the previous time phase is used as a reference image;
the brightness balance processing module is used for carrying out brightness balance processing on the image to be registered and the reference image;
the image registration module is used for segmenting an image to be registered into grids, performing template registration on each grid and a reference image, and recording registration parameter vectors between each grid and the registration grids in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are difference entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
the difference enhancement module is used for respectively carrying out ratio calculation on the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, changing the grid sequence to recalculate the ratio once, and solving a union set of the results of the two calculations to obtain a ratio result image for enhancing the difference;
the saturation threshold segmentation module is used for performing saturation threshold segmentation on the ratio result image to enable the gray value of a change area in the ratio result image to be unchanged, the gray value of an unchanged area to be set as 0, and graying the result of the saturation threshold segmentation;
the binary segmentation module is used for carrying out binary segmentation on the gray level image and removing image noise through morphological opening operation;
the difference clustering module is used for vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon, namely the difference central coordinate;
and the damage judging unit is used for positioning in the image according to the difference center coordinates, performing manual secondary screening and judging whether the oil and gas pipeline is damaged at the position.
The memory storing the unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis software of the embodiment of the invention executes the following programs:
acquiring multi-time-phase images of the oil and gas pipeline along the way shot by an unmanned aerial vehicle, wherein the images all comprise corresponding geographic coordinates, cutting overlapping areas of front and rear time-phase images according to the geographic coordinates, taking the overlapping area of the current time phase as an image to be registered, and taking the overlapping area of the previous time phase as a reference image;
carrying out brightness equalization processing on the image to be registered and the reference image;
dividing an image to be registered into grids, performing template registration on each grid and a reference image, and recording registration parameter vectors between each grid and the registration grids in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are differential entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
respectively calculating the ratio of the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, recalculating the ratio once by changing the grid sequence, and solving a union set of the results of the two calculations to obtain a ratio result image for enhancing the difference;
performing saturation threshold segmentation on the ratio result image to enable the gray value of a changed area in the ratio result image to be unchanged, setting the gray value of an unchanged area to be 0, and graying the result of the saturation threshold segmentation;
carrying out binarization segmentation on the gray level image, and removing image noise through morphological opening operation;
vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon, namely the difference central coordinate; and positioning in the image according to the difference center coordinates, performing manual secondary screening, and judging whether the oil and gas pipeline is damaged at the position.
A novel image contrast analysis technology of an unmanned aerial vehicle is combined with images shot twice or many times, on the basis of a traditional image difference identification method, a difference entropy rule is adopted to further avoid false detection, a difference clustering merging method is utilized to avoid real fine and broken differences, main differences are identified, the differences are interpreted as hidden dangers and then submitted to manual secondary screening, and the method has high accuracy.
The method fully considers the problem faced by difference identification in the real-time inspection situation of the unmanned aerial vehicle for oil and gas pipelines, namely, the pipelines are often positioned in complex terrain areas such as forests with luxuriant vegetation, the misdetection of the traditional image difference identification method is avoided by adopting a difference entropy rule, and the detailed and fragmented differences in reality are avoided by using a difference clustering merging method.
This technique is based on unmanned aerial vehicle patrols and examines image and POS data of shooing, finds out various earth's surface changes that have danger to the oil gas pipeline and has dangerous possible artificial facilities. Surface modification may result from natural disasters (floods, landslides, collapses, etc.) or from man-made damage; dangerous manual facilities may be engineering machinery (excavators, tank trucks), illegal buildings and the like. Because the oil gas pipeline is very big through region area, can produce a large amount of photos of taking photo by plane when unmanned aerial vehicle patrols and examines, change through these earth's surfaces of manual identification, with high costs, inefficiency. The novel image contrast analysis technology of the unmanned aerial vehicle is characterized in that aerial photos are identified by a computer, on the basis of a traditional image difference identification method, a difference entropy rule is adopted to avoid false detection, a difference clustering merging method is used to avoid fine and broken differences in reality, suspicious points are found out and then are handed to manual interpretation and screening, and the working efficiency can be greatly improved.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.
Claims (9)
1. An automatic contrast analysis method for images of oil and gas pipelines of unmanned aerial vehicles is characterized by comprising the following steps:
s1, acquiring multi-time-phase images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle, wherein the images all contain corresponding geographic coordinates, cutting overlapping areas of front and rear time-phase images according to the geographic coordinates, taking the overlapping area of the current time phase as an image to be registered, and taking the overlapping area of the previous time phase as a reference image;
s2, performing brightness equalization processing on the image to be registered and the reference image;
s3, dividing the image to be registered into grids, carrying out template registration on each grid and the reference image, and recording registration parameter vectors between each grid and the registration grid in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are difference entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
s4, respectively carrying out ratio calculation on the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, changing the sequence of the grids, recalculating the ratio once, and solving a union set of the results of the two calculations to obtain a difference-enhanced ratio result image;
the method for calculating the union set comprises the following steps:
respectively carrying out ratio calculation on the corresponding grids and the registration grids thereof to obtain a ratio image named T1, interchanging the left and right images to be compared, carrying out ratio again to obtain a second ratio image named T2; and performing union operation on the results of the two ratios, namely performing union calculation on T1 and T2, wherein the calculation mode is as follows:
T0x=max(T1x,T2x)
wherein, the union image is T0, which is the ratio result image;
s5, performing saturation threshold segmentation on the ratio result image to enable the gray value of the changed area in the ratio result image to be unchanged, setting the gray value of the unchanged area to be 0, and graying the result of the saturation threshold segmentation;
s6, performing binarization segmentation on the gray level image, and removing image noise through morphological opening operation;
s7, vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon vector, namely the difference central coordinate; and positioning in the image according to the difference center coordinates, performing manual secondary screening, and judging whether the oil and gas pipeline is damaged at the position.
2. The unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method of claim 1, wherein the method for performing image brightness equalization processing in step S2 comprises:
converting an image to be registered and a reference image from an RGB space to an HSV space, and respectively carrying out histogram equalization on brightness components of the two images, wherein the formula is as follows:
wherein L is 256 gray levels, h (x)i) Representing the number of pixels of each gray level in the histogram, wherein w and h respectively represent the width and height of the image;
and performing inverse transformation on the image after the histogram equalization to convert the image from the HSV space to the RGB space.
3. The unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method of claim 1, wherein the method for template registration of the grid in step S3 specifically comprises:
dividing an image to be registered into grids, wherein the size of the grids is 256 to 256 pixels, cutting out a region with the area 9 times that of the grids in a reference image by taking the grids as a center, taking each grid as a template, and performing template registration in the cut region in the reference image; calculating the similarity between the template and the cutting area, wherein the formula is as follows:
T'(x',y')=T(x',y')-1/(w·h)∑x”,y”T(x”,y”)
I'(x+x',y+y')=I(x+x',y+y')-1/(w·h)∑x”,y”I(x+x”,y+y”)
wherein w and h are the width and height of the template image, T (x ', y') is the pixel gray value of the template image at (x ', y'), and I (x + x ', y + y') is the pixel gray value of the reference image at (x + x ', y + y'); calculating the average value of the similarity of the three wave bands of r, g and b as the final template similarity;
and forming an array by the obtained similarity, forming a single-channel gray image by the array, and recording a grid corresponding to the brightest point in the gray image as a registration grid.
4. The unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method of claim 1, wherein the method for calculating the ratio in step S4 is as follows:
calculating the ratio between the mesh and the registration mesh according to the following formula:
wherein, X1And X2The pixel values in the two image grids to be compared are respectively.
5. The unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method of claim 1, wherein the method for performing saturation threshold segmentation in step S5 comprises:
converting the ratio result image from an RGB space to an HSV space, extracting a saturation component, and setting a saturation threshold, wherein the region with the saturation component larger than the saturation threshold is a change region, and the gray value of the change region is unchanged; the area with the saturation component smaller than the saturation threshold is an unchanged area, the gray value of the unchanged area is set to be 0, the result of the saturation threshold segmentation is grayed, and the graying formula is as follows:
Y=0.3r+0.59g+0.11b
wherein r, g and b are pixel gray values of red, green and blue wave bands respectively.
6. The unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method according to claim 1, wherein the method for performing binarization segmentation in step S6 comprises the following steps:
and setting a binarization threshold value for the grayscale image, setting the gray value of the pixel in the grayscale image larger than the binarization threshold value as 1, and setting the gray value of the pixel smaller than the binarization threshold value as 0, thereby completing the binarization segmentation of the grayscale image.
8. The utility model provides an automatic contrastive analysis system of unmanned aerial vehicle oil gas pipeline image which characterized in that includes:
the unmanned aerial vehicle image acquisition unit is used for shooting multi-temporal images along the oil and gas pipeline by the unmanned aerial vehicle;
the image processing and positioning unit is used for processing multi-temporal images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle and detecting the positions of the difference points in the images; the system comprises the following modules:
the overlap region cutting module is used for acquiring multi-time phase images of the oil and gas pipeline along the way shot by the unmanned aerial vehicle, wherein the images all comprise corresponding geographic coordinates, the overlap regions of the front and rear time phase images are cut according to the geographic coordinates, the overlap region of the current time phase is used as an image to be registered, and the overlap region of the previous time phase is used as a reference image;
the brightness balance processing module is used for carrying out brightness balance processing on the image to be registered and the reference image;
the image registration module is used for segmenting an image to be registered into grids, performing template registration on each grid and a reference image, and recording registration parameter vectors between each grid and the registration grids in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are difference entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
the difference enhancement module is used for respectively carrying out ratio calculation on the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, changing the grid sequence to recalculate the ratio once, and solving a union set of the results of the two calculations to obtain a ratio result image for enhancing the difference;
the method for calculating the union set comprises the following steps:
respectively carrying out ratio calculation on the corresponding grids and the registration grids thereof to obtain a ratio image named T1, interchanging the left and right images to be compared, carrying out ratio again to obtain a second ratio image named T2; and performing union operation on the results of the two ratios, namely performing union calculation on T1 and T2, wherein the calculation mode is as follows:
T0x=max(T1x,T2x)
wherein, the union image is T0, which is the ratio result image;
the saturation threshold segmentation module is used for performing saturation threshold segmentation on the ratio result image to enable the gray value of a change area in the ratio result image to be unchanged, the gray value of an unchanged area to be set as 0, and graying the result of the saturation threshold segmentation;
the binary segmentation module is used for carrying out binary segmentation on the gray level image and removing image noise through morphological opening operation;
the difference clustering module is used for vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon, namely the difference central coordinate;
and the damage judging unit is used for positioning in the image according to the difference center coordinates, performing manual secondary screening and judging whether the oil and gas pipeline is damaged at the position.
9. The utility model provides a store has memory of unmanned aerial vehicle oil gas pipeline image automatic contrastive analysis software which characterized in that, this software execution following procedure:
acquiring multi-time-phase images of the oil and gas pipeline along the way shot by an unmanned aerial vehicle, wherein the images all comprise corresponding geographic coordinates, cutting overlapping areas of front and rear time-phase images according to the geographic coordinates, taking the overlapping area of the current time phase as an image to be registered, and taking the overlapping area of the previous time phase as a reference image;
carrying out brightness equalization processing on the image to be registered and the reference image;
dividing an image to be registered into grids, performing template registration on each grid and a reference image, and recording registration parameter vectors between each grid and the registration grids in the reference image, wherein the registration parameter vectors comprise translation parameters, rotation parameters and scaling parameters, and the vectors are differential entropy vectors; forming a difference entropy vector table after finishing the registration of all the grid templates;
respectively calculating the ratio of the corresponding grid and the registration grid thereof according to each unit value in the difference entropy vector table, recalculating the ratio once by changing the grid sequence, and solving a union set of the results of the two calculations to obtain a ratio result image for enhancing the difference;
the method for calculating the union set comprises the following steps:
respectively carrying out ratio calculation on the corresponding grids and the registration grids thereof to obtain a ratio image named T1, interchanging the left and right images to be compared, carrying out ratio again to obtain a second ratio image named T2; and performing union operation on the results of the two ratios, namely performing union calculation on T1 and T2, wherein the calculation mode is as follows:
T0x=max(T1x,T2x)
wherein, the union image is T0, which is the ratio result image;
performing saturation threshold segmentation on the ratio result image to enable the gray value of a changed area in the ratio result image to be unchanged, setting the gray value of an unchanged area to be 0, and graying the result of the saturation threshold segmentation;
carrying out binarization segmentation on the gray level image, and removing image noise through morphological opening operation;
vectorizing the denoised binary image, extracting the outline of a white area to obtain a difference polygon vector, and calculating the mass moment of the difference polygon to obtain the barycentric coordinate of the difference polygon, namely the difference central coordinate; and positioning in the image according to the difference center coordinates, performing manual secondary screening, and judging whether the oil and gas pipeline is damaged at the position.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710994740.5A CN107818303B (en) | 2017-10-23 | 2017-10-23 | Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710994740.5A CN107818303B (en) | 2017-10-23 | 2017-10-23 | Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107818303A CN107818303A (en) | 2018-03-20 |
CN107818303B true CN107818303B (en) | 2021-06-15 |
Family
ID=61608372
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710994740.5A Active CN107818303B (en) | 2017-10-23 | 2017-10-23 | Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107818303B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110832495A (en) * | 2018-07-13 | 2020-02-21 | 深圳市大疆创新科技有限公司 | Wave identification method and device, computer readable storage medium and unmanned aerial vehicle |
CN109447946B (en) * | 2018-09-26 | 2021-09-07 | 中睿通信规划设计有限公司 | Overhead communication optical cable abnormality detection method |
CN109538197B (en) * | 2018-11-01 | 2020-06-05 | 中国石油大学(北京) | Method and device for determining drilling track of oil and gas reservoir and storage medium |
CN110321770B (en) * | 2019-03-25 | 2022-05-31 | 西安长城数字软件有限公司 | Pipeline monitoring method, device, equipment and storage medium |
JP6788759B1 (en) * | 2020-02-12 | 2020-11-25 | Dmg森精機株式会社 | Information processing equipment and information processing system |
CN112307533A (en) * | 2020-10-30 | 2021-02-02 | 深圳集智数字科技有限公司 | Control and regulation graph generation method and related equipment |
CN113177444A (en) * | 2021-04-16 | 2021-07-27 | 云南云能科技有限公司 | Intelligent identification method and system for unmanned aerial vehicle image oil and gas transmission pipeline |
CN114998159B (en) * | 2022-08-04 | 2022-10-28 | 邹城市天晖软件科技有限公司 | Design image self-adaptive enhancement method |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103646395A (en) * | 2013-11-28 | 2014-03-19 | 中南大学 | A high-precision image registering method based on a grid method |
CN104704424A (en) * | 2012-08-21 | 2015-06-10 | 视觉智能有限合伙公司 | Infrastructure mapping system and method |
CN105023014A (en) * | 2015-08-21 | 2015-11-04 | 马鞍山市安工大工业技术研究院有限公司 | Method for extracting tower target in unmanned aerial vehicle routing inspection power transmission line image |
CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
CN105741309A (en) * | 2016-03-18 | 2016-07-06 | 武汉大学 | Remote sensing image change detection method based on chi-square transformation and sample selection |
US9626568B2 (en) * | 2013-11-26 | 2017-04-18 | Rowan University | Use of spatially structured light for dynamic three dimensional reconstruction and reality augmentation |
CN106681331A (en) * | 2017-02-24 | 2017-05-17 | 王红军 | Grid-path planning method based on geodesic theory |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB201514643D0 (en) * | 2015-08-18 | 2015-09-30 | Micromass Ltd | Mass Spectrometer data acquisition |
-
2017
- 2017-10-23 CN CN201710994740.5A patent/CN107818303B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104704424A (en) * | 2012-08-21 | 2015-06-10 | 视觉智能有限合伙公司 | Infrastructure mapping system and method |
US9626568B2 (en) * | 2013-11-26 | 2017-04-18 | Rowan University | Use of spatially structured light for dynamic three dimensional reconstruction and reality augmentation |
CN103646395A (en) * | 2013-11-28 | 2014-03-19 | 中南大学 | A high-precision image registering method based on a grid method |
CN105023014A (en) * | 2015-08-21 | 2015-11-04 | 马鞍山市安工大工业技术研究院有限公司 | Method for extracting tower target in unmanned aerial vehicle routing inspection power transmission line image |
CN105551028A (en) * | 2015-12-09 | 2016-05-04 | 中山大学 | Method and system for dynamically updating geographic space data based on remote sensing image |
CN105741309A (en) * | 2016-03-18 | 2016-07-06 | 武汉大学 | Remote sensing image change detection method based on chi-square transformation and sample selection |
CN106681331A (en) * | 2017-02-24 | 2017-05-17 | 王红军 | Grid-path planning method based on geodesic theory |
Non-Patent Citations (2)
Title |
---|
"Earth Observation Remote Sensing and GIS Services for Monitoring of Integration Systems";Sefer Kurnaz等;《2007 3rd International Conference on Recent Advances in Space Technologies》;20070808;268-270页 * |
"中石化智能化管线系统720度全景影像制作";黄代军;《江汉石油职工大学学报》;20160731;第29卷(第4期);71-73页 * |
Also Published As
Publication number | Publication date |
---|---|
CN107818303A (en) | 2018-03-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107818303B (en) | Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory | |
CN109615611B (en) | Inspection image-based insulator self-explosion defect detection method | |
US10635929B2 (en) | Saliency-based method for extracting road target from night vision infrared image | |
Liu et al. | Multiscale U-shaped CNN building instance extraction framework with edge constraint for high-spatial-resolution remote sensing imagery | |
CN109598794B (en) | Construction method of three-dimensional GIS dynamic model | |
US10592754B2 (en) | Shadow removing method for color image and application | |
CN108876723B (en) | Method for constructing color background of gray target image | |
CN107451982B (en) | High-canopy-density forest stand crown area acquisition method based on unmanned aerial vehicle image | |
Ghaffarian | Automatic building detection based on supervised classification using high resolution Google Earth images | |
CN105678318B (en) | The matching process and device of traffic sign | |
CN117036641A (en) | Road scene three-dimensional reconstruction and defect detection method based on binocular vision | |
Zakharov et al. | Automatic building detection from satellite images using spectral graph theory | |
Chaloeivoot et al. | Building detection from terrestrial images | |
CN114581307A (en) | Multi-image stitching method, system, device and medium for target tracking identification | |
Manandhar et al. | Segmentation based building detection in high resolution satellite images | |
Sidike et al. | Automatic building change detection through adaptive local textural features and sequential background removal | |
Salehi et al. | Object-based land cover classification of urban areas using VHR imagery and photogrammetrically-derived DSM | |
van de Wouw et al. | Hierarchical 2.5-d scene alignment for change detection with large viewpoint differences | |
CN111833384A (en) | Method and device for quickly registering visible light and infrared images | |
Yu et al. | Advanced approach for automatic reconstruction of 3d buildings from aerial images | |
JP2005173128A (en) | Contour shape extractor | |
Lavigne et al. | Automatic building detection and 3D shape recovery from single monocular electro-optic imagery | |
Zou et al. | Flood Depth Assessment with Location-Based Social Network Data and Google Street View-A Case Study with Buildings as Reference Objects | |
Sohn et al. | 3-D building extraction using IKONOS multispectral images | |
Balamurugan et al. | A novel framework for extraction of landscape areas and automatic building detection in satellite images |
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 |