CN109785307A - A kind of unmanned plane image road Damage assessment method based on vector guidance - Google Patents
A kind of unmanned plane image road Damage assessment method based on vector guidance Download PDFInfo
- Publication number
- CN109785307A CN109785307A CN201910020003.4A CN201910020003A CN109785307A CN 109785307 A CN109785307 A CN 109785307A CN 201910020003 A CN201910020003 A CN 201910020003A CN 109785307 A CN109785307 A CN 109785307A
- Authority
- CN
- China
- Prior art keywords
- road
- area
- road surface
- edge
- unmanned plane
- 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
Links
Landscapes
- Image Processing (AREA)
Abstract
The present invention proposes that a kind of unmanned plane image road Damage assessment method based on vector guidance, this method utilize existing road network vector data, collected unmanned plane image data are registrated therewith, buffer area is done using the road being registrated and cuts to image.The marginal information of image is acquired using edge detection operator, then edge is attached with Morphology Algorithm, micronization processes.Seed point is generated using vector data later, is pointed out from seed and unrestrained water filling is carried out to road surface, sequentially generate multiple road surfaces.The number and area for finally counting road surface, are compared with the area of complete road, realize the objective evaluation of road damage.
Description
Technical field
The present invention relates to a kind of unmanned plane image road Damage assessment methods, are based especially on the unmanned plane shadow of vector guidance
As road damage appraisal procedure, belong to road disaster assessment technology field.
Background technique
Road traffic is as main mode of transportation, it carries most flow of personnel, cargo transport task, in state
It is played a crucial role in people's economic development.However road is often subject to the broken of the natural calamities such as earthquake, flood, mud-rock flow
It is bad, serious road damage is caused, brings great loss to national economy.After road is damaged, highway administration portion
Door needs to do a large amount of investigation in roadside, assesses the damage degree of road.
Traditional road Damage assessment mostly uses greatly artificial scene to adjust the mode drawn, and this mode is time-consuming and laborious, can not be quick
Assess the damage situation of road.Especially after disaster generation, road is damaged completely, and personnel cannot be introduced into disaster region at all
Manual research is carried out, road damage appraisal procedure traditional at this time is entirely ineffective.
Remote sensing satellite and conventional photographer's aircraft can provide support for road damage assessment after calamity, utilize remote sensing image
The technology for doing road damage assessment is rapidly developed, but the resolution ratio of remote sensing image is generally lower, and road is past on image
It is past to show as very thin linear target, therefore the precision of road damage assessment is also accordingly restricted.On the other hand, satellite is distant
Sense and conventional aeroplane photography can all be blocked by cloud and mist to be influenced, and the generation of disaster is often all along with overcast and rainy cloud and mist weather
Occur, leads to not the image for obtaining road damage region using remote sensing satellite and conventional photographer's aircraft, therefore this side
The practicability of method is greatly limited.
After unmanned plane occurs, the features such as because of its mobility strong, low-latitude flying not by Influence of cloud, it is well suited for doing road damage
Investigation application.The work most contents that road damage is assessed can be transferred to indoor completion using unmanned plane image, significantly
Improve the working efficiency of road damage assessment.And unmanned plane can reach the region that some can not reach, and make many originals
A possibility that road damage evaluation work to have no idea to complete becomes to complete, and substantially increases road damage assessment.
But it is existing carry out the technology of road damage assessment still to manually visualize based on interpretation using unmanned plane image,
After the way taken is field acquisition unmanned plane image, return to it is indoor human interpretation is carried out to image, hand labeled goes out to damage
Region, and using the area in GIS platform statistics damage region, the damage situations of road are assessed with this.This side
The working efficiency of formula is still not high enough, and the experience of heavy dependence people, as a result also not objective enough.
It is provided by the invention based on vector guidance unmanned plane image road Damage assessment method, using match nobody
Machine image and road network vector data can automatically assess road damage situation, have high degree of automation, speed it is fast,
The objective feature of assessment result.
Summary of the invention
It is low for the degree of automation existing for background technique, working efficiency is not high, as a result not objectively insufficient enough, the present invention
Provide a kind of unmanned plane image road Damage assessment method based on vector guidance.Method proposed by the present invention can be sufficiently sharp
With computerized algorithm, the vector data matched and remote sensing image data are input in computer, are automatically performed by computer
The assessment of road damage.
The technical solution adopted by the invention is as follows:
A kind of unmanned plane image road Damage assessment method based on vector guidance, step include:
Step 1, vector road network data is registrated with unmanned plane image data；
Step 2, unmanned plane image road face edge extracting；
Step 3, Morphological scale-space is carried out to the road surface edge that step 2 is extracted, forms multiple connected domains, it is one of to connect
Logical domain is exactly a complete road surface；
Step 4, seed point is generated under vector guidance, then connected domain is filled based on seed point, form road
Road surface；
Step 5, road damage situation is assessed, according to the number of road surface in step 4, the length of road, road width,
The gross area of road surface assesses the damage of road, calculates the assessed value of road damage.
Further, specific real using the canny operator extraction road surface edge with bilateral threshold property in step 2
Existing mode is as follows,
(2-1) determines the range cut on unmanned plane raw video by the way of buffer area, wherein the width of buffer area
It is arranged according to the width of road；
(2-2) road is cut, and using buffer area as template is cut, will be fallen in the pixel cut in template area and is retained,
It falls in the pixel value cut outside template area and is set to 0, obtain the image after road is cut；
(2-3) to the image gray processing in (2-2), the formula of transformation is as follows:
Gray=(R*299+G*587+B*114+500)/1000 (1)
Wherein R, G, B respectively represent the value of three wave bands of RGB of original color image, the ash after Gray representation transformation
Angle value；
(2-4) carries out median filtering to the image of gray processing, using the window of a 3*3, takes in window 9 pixel values
Median achievees the effect that filtering as the pixel value among window；
(2-5) canny operator extraction road surface edge, implementation process are as follows:
(a) gradient fortune is done to the image after median filtering respectively using two threshold values threshold1 and threshold2
It calculates, and threshold1 > threshold2, retains gradient value and be greater than the pixel of threshold value as candidate edge, threshold1
The edge determined is edge1, and the edge that threshold2 is determined is edge2；
(b) based on edge1, the edge2 being connected with edge1 is incorporated to edge1, and using the edge1 after merging as base
Plinth continues to merge, and repeats this process until the edge2 without being connected with edge1.
Further, the specific implementation of step 3 is as follows,
(3-1) does closed operation to initial road edge with a structural element, and the formula of closed operation is as follows:
In formula, A is original edge image, and B is structural element, and closed operation is exactly first to carry out expansive working to A with B, it
Etching operation is carried out to A with B again afterwards；
(3-2) in order to refine edge more, with another structural element, by following constraint principle by edge thinning
At the filament of single pixel:
I) connectivity will be changed by removing the point；
Ii) removing the point will make line shorten.
Further, the specific implementation steps are as follows for step 4,
The starting kind of (4-1) under the guidance of vector data, every one seed point of m meters of generations, as connected domain filling
Sub- point；
(4-2) carries out unrestrained water filling based on seed point, generates road surface；Since first seed point, successively
Each connected domain is filled, and carries out pixel label by the sequence from 1 to n, and the element marking of the same connected domain is at same
A number；During label, if the pixel where a seed point has been labeled, illustrate connected domain by
It is labeled, then the seed point is skipped, until all seed points all traverse completion；In seed point growth course, setting one
A threshold value prevents from falling in the seed outside road surface and infinitely fills, and after reaching this threshold value, if filling terminates not yet, recognizes
It has been fallen in except road surface for this seed point, all growing point pixels of the seed point has all been reset to 0, and filling process
Turn to next seed point.
Further, the specific implementation of step 5 is as follows,
The assessment of (5-1) road damage point number；The number of road surface is counted, it is equal to the last one connected domain being filled
The mark value of pixel value, the number that the number of road damage point is equal to road surface subtract 1；
The assessment of (5-2) road damage area；By the area of comparison road disaster-stricken preceding area and road after disaster-stricken, indirectly
Evaluate the disaster area of road, the specific steps are as follows:
(a) the disaster-stricken preceding areal calculation of road；Before road is damaged by disaster, road is completely that the area of road is equal to
The product of the length and width of road, calculation formula are as follows:
SBefore calamity=L*d (3)
Wherein, SBefore calamityFor the area before calamity, L is the length of road, and d is the width of road；
(b) road disaster area is assessed；After road is disaster-stricken, road surface is broken into multiple road surfaces, calculates multiple road surfaces
Area be added sum, obtain it is disaster-stricken after path area, be first depending on the area that following formula calculates each road surface:
Si=Ni*s (4)
Wherein, SiFor the area of i-th of road surface, Ni is the number of pixel value contained by i-th of road surface, and s is each picture
The area of element value, equal to square of resolution ratio；
The gross area of road surface after calamity is finally calculated again, and calculation formula is as follows:
SAfter calamity=S1+S2+…+Sn (5)
Wherein, SAfter calamityFor the gross area of the road after impaired, SnFor the area of n-th of road surface；
The assessment of (5-3) road damaged area percentage, the percentage of road damaged area is calculated according to following formula:
Impaired percentage=(SBefore calamity-SAfter calamity)/SBefore calamity (6)。
The present invention has the advantages that:
(1) high degree of automation.
Method proposed by the present invention takes full advantage of computer graphical processing algorithm, and computer is all transferred in a large amount of work
It completes, therefore the degree of automation is compared with art methods height.
(2) estimating velocity is fast.
Road damage evaluation work has been organized into fixed process by method proposed by the present invention, and writes computer program
Most of process is automatically completed, so that the process entirely to work becomes very simple.In such a way that vector guides, only
Image locating for road is handled, the operand of algorithm is greatly reduced, accelerates calculating speed.Therefore, entire road
The work of Damage assessment is very fast.
(3) assessment result is objective.
Existing technical solution is related to many manual operation processes, these manually-operated results rely heavily on
Judge in the subjective experience of people, it is not objective enough.These manual operations are all based on algorithm and automated by technical solution of the present invention
At whole flow process is not related to the subjective judgement of people, therefore the assessment result of road damage is objective.
Detailed description of the invention
Fig. 1 is the method for the present invention overview flow chart；
Fig. 2 is the specific flow chart of step 1 of the present invention；
Fig. 3 is the specific flow chart of step 2 of the present invention；
Fig. 4 is the specific flow chart of step 3 of the present invention；
Fig. 5 is the specific flow chart of step 4 of the present invention；
Fig. 6 is the algorithm flow chart of road surface filling；
Fig. 7 is the specific flow chart of step 5 of the present invention.
Specific embodiment
Understand for the ease of those of ordinary skill in the art and implement the present invention, with reference to the accompanying drawings and embodiments to this hair
It is bright to be described in further detail:
The overall flow of real-time mode of the present invention is shown in Fig. 1: utilizing existing road network vector data, will collect
Unmanned plane image data be registrated therewith, the marginal information of gray level image is acquired using edge detection operator, then calculated with morphology
Method is attached edge, micronization processes.Seed point is generated using vector data later, road surface is carried out from seed point
Unrestrained water filling, generates multiple road surfaces, finally counts the number and area of road surface, be compared with the area of complete road,
Realize the objective evaluation of road damage.It should be pointed out that all steps after registration can be in the side of electronic computer
It helps down and is automatically finished.Illustrate the process of entire embodiment step by step below:
Step 1: the process that road network vector data is registrated with unmanned plane image data is as shown in Figure 2.According to covering for image
Lid range, from road net data cut obtain the vector data of assessment area, if the road to be assessed have accurate name or
Number can also directly inquire from road network data system and obtain the vector data of the road.It finally needs vector data
It is transformed under the same coordinate system with raster data, makes vector data together with raster data registration.
Step 2: the process of road surface edge extracting is as shown in Figure 3.Road surface edge extracting goes out from the essential characteristic of image
Hair, using there is obvious boundary between road surface and background, the gradient value by calculating image realizes road surface edge
Extraction.In order to keep edge extracting more accurate, the present invention uses the canny operator with bilateral threshold property as edge meter
The algorithm of calculation.Illustrate the specific steps of road surface edge extracting below with reference to Fig. 3:
(2-1) buffer generation.It not only include road entity on one secondary unmanned plane raw video, also containing largely and road
Unrelated background atural object needs specially to cut out the road sections on image and, before cutting to accelerate the speed of operation
It needs to be determined that the range cut, the present invention determine the range of cutting by the way of buffer area.The width of buffer area is basis
Width (being obtained by the road network data) setting of road, for example, road width is 5 meters, then the width of buffer area is set as 2.5
Rice, buffering direction are left and right buffering.
(2-2) road is cut.Road is cut using buffer area as template is cut, and is automatically performed cutting on computers,
The principle of cutting is to fall in the pixel cut in template area to retain, and falls in the pixel value cut outside template area and is set to 0.
(2-3) image gray processing.Marginal information is extracted in order to easily carry out gradient algorithm to image, is needed cromogram
As carrying out gray processing, gray processing can there are three the image that the image of wave band becomes only one wave band, the public affairs of transformation by script
Formula is as follows:
Gray=(R*299+G*587+B*114+500)/1000 (1)
Wherein R, G, B respectively represent the value of three wave bands of RGB of original color image, the ash after Gray representation transformation
Angle value.
(2-4) median filtering.It is influenced by unmanned plane imaging sensor precision itself, it can be containing perhaps on unmanned plane image
More noises, if these noises do not filter out, it will form false edge when edge extracting.Median filtering uses a 3*3
Window, take the median of 9 pixel values in window as the pixel value among window, achieve the effect that filtering.
(2-5) canny operator extraction road surface edge.Canny operator is a kind of common arithmetic operators, is implemented
Journey is as follows:
(a) gradient fortune is done to the image after median filtering respectively using two threshold values threshold1 and threshold2
It calculates, and threshold1 > threshold2, retains gradient value and be greater than the pixel of threshold value as candidate edge, threshold1
The edge determined is edge1, and the edge that threshold2 is determined is edge2.
(b) based on edge1, the edge2 being connected with edge1 is incorporated to edge1, and using the edge1 after merging as base
Plinth continues to merge, and repeats this process until the edge2 without being connected with edge1.
Step 3: the process of road surface edge configuration processing is as shown in Figure 4.Road surface side by canny operator extraction
Edge be still it is incomplete, there are many be broken place, road can not be surrounded to a complete road surface.Morphological scale-space can
To connect the edge line disconnected and keep edge line more smooth, the specific steps are as follows:
(3-1) closed operation operation.Using the matrix of a 7*7 as structural element, closed operation is done to initial road edge, closes behaviour
Make the interruption that can diminish narrow and long thin wide gap, eliminates small cavity, and fill up the fracture in contour line.The public affairs of closed operation
Formula is as follows:
In formula, A is original edge image, and B is structural element, and closed operation is exactly first to carry out expansive working to A with B, it
Etching operation is carried out to A with B again afterwards.
(3-2) Refinement operation.It is used in order to which the edge line that farthest will be switched off connects in the closed operation stage
Structural element it is generally all bigger, the edge at edge is generally all relatively thick after processing in this way, in order to keep edge more smart
Refinement, using the matrix of a 3*3 as structural element, by following constraint principle by edge thinning at the filament of single pixel:
I) connectivity will be changed by removing the point
Ii it) removes the point line will be made to shorten and only meet above situation, central point pixel cannot just become 0.
Step 4: road surface filling is as shown in Figure 5.The processing of step 3 is arrived by step 1, road forms multiple connections
Domain, a connected domain are exactly a complete road surface, and road surface filling is to generate seed point under vector guidance, then with seed
Connected domain is filled based on point, forms road surface.Specific step is as follows:
The guidance of (4-1) vector generates seed point.The most important step of filling work to connected domain is the generation of seed point,
Starting seed point under the guidance of vector data, every 1 meter of generation, one seed point, as connected domain filling.
(4-2) carries out unrestrained water filling based on seed point, generates road surface.As shown in fig. 6, being opened from first seed point
Begin, successively each connected domain is filled, and carries out pixel label, the pixel mark of the same connected domain by the sequence from 1 to n
Remember into the same number.During label, if the pixel where a seed point has been labeled, illustrate to be connected to
Domain has been labeled, then skips the seed point, until all seed points all traverse completion.In seed point growth course
In, it needs to set a threshold value, prevents from falling in the seed outside road surface and infinitely fill, after reaching this threshold value, if filling is also
It does not terminate, then it is assumed that this seed point has been fallen in except road surface, and all growing point pixels of the seed point are all reset to
0, and filling process is turned to next seed point.
Step 5: the assessment of road damage situation is as shown in Figure 7.The processing of step 4 is arrived by step 1, road has been extracted
At multiple complete road surfaces, and the pixel in each face has a mark value, by count road surface mark value and
Pixel number can make assessment to road damage situation.Specific steps are as follows:
The assessment of (5-1) road damage point number.The number of road surface is counted, it is equal to the last one connected region being filled
The mark value of domain pixel value, the number that the number of road damage point is equal to road surface subtract 1.For example, there is one to be finally filled
Connected domain element marking value be n, then the road has n-1 damage point.
The assessment of (5-2) road damage area.By the area of comparison road disaster-stricken preceding area and road after disaster-stricken, indirectly
Evaluate the disaster area of road, the specific steps are as follows:
(a) the disaster-stricken preceding areal calculation of road.Before road is damaged by disaster, road is completely that the area of road is equal to
The product of the length and width of road, calculation formula are as follows:
SBefore calamity=L*d (3)
Wherein, SBefore calamityFor the area before calamity, L is the length of road, and d is the width of road.
(b) road disaster area is assessed.After road is disaster-stricken, road surface is broken into multiple road surfaces, calculates multiple road surfaces
Area be added sum, it is available it is disaster-stricken after path area.It is first depending on the area that following formula calculates each road surface:
Si=Ni*s (4)
Wherein, SiFor the area of i-th of road surface, Ni is the number of pixel value contained by i-th of road surface, and s is each picture
The area of element value, equal to square of resolution ratio, such as: unmanned plane image resolution is 0.1m, then s=0.01m2。
The gross area of road surface after calamity is finally calculated again, and calculation formula is as follows:
SAfter calamity=S1+S2+…+Sn (5)
Wherein, SAfter calamityFor the gross area of the road after impaired, SnFor the area of n-th of road surface.
The assessment of (5-3) road damaged area percentage.The percentage of road damaged area is calculated according to following formula:
Impaired percentage=(SBefore calamity-SAfter calamity)/SBefore calamity (6)
It should be understood that the part that this specification does not elaborate belongs to the prior art.
Specific embodiment described herein is only an example for the spirit of the invention.The neck of technology belonging to the present invention
The technical staff in domain can make various modifications or additions to the described embodiments or replace by a similar method
In generation, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Claims (5)
1. a kind of unmanned plane image road Damage assessment method based on vector guidance, which comprises the following steps:
Step 1, vector road network data is registrated with unmanned plane image data；
Step 2, unmanned plane image road face edge extracting；
Step 3, Morphological scale-space is carried out to the road surface edge that step 2 is extracted, forms multiple connected domains, one of connected domain
It is exactly a complete road surface；
Step 4, seed point is generated under vector guidance, then connected domain is filled based on seed point, form road surface；
Step 5, road damage situation is assessed, number, the length of road, the width of road, road according to road surface in step 4
The gross area in face assesses the damage of road, calculates the assessed value of road damage.
2. a kind of unmanned plane image road Damage assessment method based on vector guidance according to claim 1, feature
Be: using the canny operator extraction road surface edge with bilateral threshold property in step 2, specific implementation is as follows,
(2-1) determines the range cut on unmanned plane raw video by the way of buffer area, wherein the width of buffer area according to
The width of road is arranged；
(2-2) road is cut, and using buffer area as template is cut, will be fallen in the pixel cut in template area and is retained, fall in
The pixel value cut outside template area is set to 0, obtains the image after road is cut；
(2-3) to the image gray processing in (2-2), the formula of transformation is as follows:
Gray=(R*299+G*587+B*114+500)/1000 (1)
Wherein R, G, B respectively represent the value of three wave bands of RGB of original color image, the gray value after Gray representation transformation；
(2-4) carries out median filtering to the image of gray processing and takes the middle position of 9 pixel values in window using the window of a 3*3
Number achievees the effect that filtering as the pixel value among window；
(2-5) canny operator extraction road surface edge, implementation process are as follows:
(a) gradient algorithm is done to the image after median filtering respectively using two threshold values threshold1 and threshold2, and
Threshold1 > threshold2 retains pixel of the gradient value greater than threshold value as candidate edge, and threshold1 is determined
The edge come is edge1, and the edge that threshold2 is determined is edge2；
(b) based on edge1, the edge2 being connected with edge1 is incorporated to edge1, and based on the edge1 after merging, after
It is continuous to merge, this process is repeated until the edge2 without being connected with edge1.
3. a kind of unmanned plane image road Damage assessment method based on vector guidance according to claim 1, feature
Be: the specific implementation of step 3 is as follows,
(3-1) does closed operation to initial road edge with a structural element, and the formula of closed operation is as follows:
In formula, A is original edge image, and B is structural element, and closed operation is exactly first to carry out expansive working, Zhi Houzai to A with B
Etching operation is carried out to A with B；
(3-2) in order to refine edge more, with another structural element, by following constraint principle by edge thinning Cheng Dan
The filament of pixel:
I) connectivity will be changed by removing the point；
Ii) removing the point will make line shorten.
4. a kind of unmanned plane image road Damage assessment method based on vector guidance according to claim 1, feature
Be: the specific implementation steps are as follows for step 4,
The starting seed point of (4-1) under the guidance of vector data, every one seed point of m meters of generations, as connected domain filling；
(4-2) carries out unrestrained water filling based on seed point, generates road surface；Since first seed point, successively each company
Logical domain is filled, and carries out pixel label by the sequence from 1 to n, and the element marking of the same connected domain is at same number
Word；During label, if the pixel where a seed point has been labeled, illustrate that connected domain has been labeled
It crosses, then skips the seed point, until all seed points all traverse completion；In seed point growth course, a threshold is set
Value, prevents from falling in the seed outside road surface and infinitely fills, after reaching this threshold value, if filling terminates not yet, then it is assumed that this
A seed point has been fallen in except road surface, all growing point pixels of the seed point is all reset to 0, and filling process is turned to
Next seed point.
5. a kind of unmanned plane image road Damage assessment method based on vector guidance according to claim 1, feature
Be: the specific implementation of step 5 is as follows,
The assessment of (5-1) road damage point number；The number of road surface is counted, it is equal to the last one connected domain pixel being filled
The mark value of value, the number that the number of road damage point is equal to road surface subtract 1；
The assessment of (5-2) road damage area；Pass through the area of comparison road disaster-stricken preceding area and road after disaster-stricken, indirect assessment
The disaster area of road out, the specific steps are as follows:
(a) the disaster-stricken preceding areal calculation of road；Before road is damaged by disaster, road is that completely, the area of road is equal to road
Length and width product, calculation formula is as follows:
SBefore calamity=L*d (3)
Wherein, SBefore calamityFor the area before calamity, L is the length of road, and d is the width of road；
(b) road disaster area is assessed；After road is disaster-stricken, road surface is broken into multiple road surfaces, calculates multiple road surface areas
Be added sum, obtain it is disaster-stricken after path area, be first depending on the area that following formula calculates each road surface:
Si=Ni*s (4)
Wherein, SiFor the area of i-th of road surface, Ni is the number of pixel value contained by i-th of road surface, and s is each pixel value
Area, equal to square of resolution ratio；
The gross area of road surface after calamity is finally calculated again, and calculation formula is as follows:
SAfter calamity=S1+S2+…+Sn (5)
Wherein, SAfter calamityFor the gross area of the road after impaired, SnFor the area of n-th of road surface；
The assessment of (5-3) road damaged area percentage, the percentage of road damaged area is calculated according to following formula:
Impaired percentage=(SBefore calamity-SAfter calamity)/SBefore calamity(6)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910020003.4A CN109785307B (en) | 2019-01-09 | 2019-01-09 | Unmanned aerial vehicle image road damage assessment method based on vector guidance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910020003.4A CN109785307B (en) | 2019-01-09 | 2019-01-09 | Unmanned aerial vehicle image road damage assessment method based on vector guidance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109785307A true CN109785307A (en) | 2019-05-21 |
| CN109785307B CN109785307B (en) | 2020-08-07 |
Family
ID=66500286
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910020003.4A Active CN109785307B (en) | 2019-01-09 | 2019-01-09 | Unmanned aerial vehicle image road damage assessment method based on vector guidance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109785307B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175574A (en) * | 2019-05-28 | 2019-08-27 | 中国人民解放军战略支援部队信息工程大学 | A kind of Road network extraction method and device |
| CN111091049A (en) * | 2019-11-01 | 2020-05-01 | 东南大学 | Road surface obstacle detection method based on reverse feature matching |
| CN111582659A (en) * | 2020-04-16 | 2020-08-25 | 北京航空航天大学青岛研究院 | Mountain land operation difficulty index calculation method |
| WO2021120038A1 (en) * | 2019-12-18 | 2021-06-24 | 深圳市大疆创新科技有限公司 | Unmanned aerial vehicle control method and apparatus, and unmanned aerial vehicle and storage medium |
| CN113807272A (en) * | 2021-09-22 | 2021-12-17 | 应急管理部国家减灾中心 | Method for rapidly extracting disaster situations of disaster damaged roads on remote sensing platform based on vector guidance |
| CN113807271A (en) * | 2021-09-22 | 2021-12-17 | 应急管理部国家减灾中心 | Road damage disaster rapid evaluation system based on remote sensing cloud service platform and application thereof |
| CN114332370A (en) * | 2021-12-28 | 2022-04-12 | 埃洛克航空科技（北京）有限公司 | Road image processing method, device, equipment and storage medium |
| CN114511551A (en) * | 2022-02-22 | 2022-05-17 | 金华高等研究院(金华理工学院筹建工作领导小组办公室) | Ground damage identification system based on machine vision |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101614822A (en) * | 2009-07-17 | 2009-12-30 | 北京大学 | Detect the method for road damage based on post-disaster high-resolution remote sensing image |
| JP4584120B2 (en) * | 2005-11-21 | 2010-11-17 | トヨタ自動車株式会社 | Road marking line detection device, road marking line detection method, road marking line detection program |
| CN103714339A (en) * | 2013-12-30 | 2014-04-09 | 武汉大学 | SAR image road damaging information extracting method based on vector data |
| CN105787937A (en) * | 2016-02-25 | 2016-07-20 | 武汉大学 | OSM-based high-resolution remote sensing image road change detection method |
| CN107944407A (en) * | 2017-11-30 | 2018-04-20 | 中山大学 | A kind of crossing zebra stripes recognition methods based on unmanned plane |
| CN108645342A (en) * | 2018-04-25 | 2018-10-12 | 国交空间信息技术（北京）有限公司 | A kind of road width extracting method based on path locus and high resolution image |
-
2019
- 2019-01-09 CN CN201910020003.4A patent/CN109785307B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4584120B2 (en) * | 2005-11-21 | 2010-11-17 | トヨタ自動車株式会社 | Road marking line detection device, road marking line detection method, road marking line detection program |
| CN101614822A (en) * | 2009-07-17 | 2009-12-30 | 北京大学 | Detect the method for road damage based on post-disaster high-resolution remote sensing image |
| CN103714339A (en) * | 2013-12-30 | 2014-04-09 | 武汉大学 | SAR image road damaging information extracting method based on vector data |
| CN105787937A (en) * | 2016-02-25 | 2016-07-20 | 武汉大学 | OSM-based high-resolution remote sensing image road change detection method |
| CN107944407A (en) * | 2017-11-30 | 2018-04-20 | 中山大学 | A kind of crossing zebra stripes recognition methods based on unmanned plane |
| CN108645342A (en) * | 2018-04-25 | 2018-10-12 | 国交空间信息技术（北京）有限公司 | A kind of road width extracting method based on path locus and high resolution image |
Non-Patent Citations (6)
| Title |
|---|
| RUI GUO 等: "Road Detection Method for Land Consolidation Using Mathematical Morphology from High Resolution Image", 《PROCEEDINGS OF THE 13TH WSEAS INTERNATIONAL CONFERENCE ON APPLIED MATHEMATICS (MATH"08)》 * |
| YUNHE LIU 等: "A Road Extraction Method Based on Region Growing and Mathematical Morphology from Remote Sensing Images", 《JOURNAL OF COMPUTER AND COMMUNICATIONS》 * |
| 丁磊: "矢量数据辅助的高分辨率遥感影像道路自动提取", 《遥感学报》 * |
| 张作昌: "一种GIS辅助下的双线型线状地物提取方法", 《数字技术与应用》 * |
| 徐丰 等: "OpenStreetMap辅助下的高分辨率光学影像道路损毁提取", 《地理空间信息》 * |
| 黄磊 等: "一种基于形态学操作的SAR图像道路检测方法研究", 《福建电脑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175574A (en) * | 2019-05-28 | 2019-08-27 | 中国人民解放军战略支援部队信息工程大学 | A kind of Road network extraction method and device |
| CN111091049A (en) * | 2019-11-01 | 2020-05-01 | 东南大学 | Road surface obstacle detection method based on reverse feature matching |
| CN111091049B (en) * | 2019-11-01 | 2024-02-09 | 东南大学 | Road surface obstacle detection method based on reverse feature matching |
| WO2021120038A1 (en) * | 2019-12-18 | 2021-06-24 | 深圳市大疆创新科技有限公司 | Unmanned aerial vehicle control method and apparatus, and unmanned aerial vehicle and storage medium |
| CN113646722A (en) * | 2019-12-18 | 2021-11-12 | 深圳市大疆创新科技有限公司 | Unmanned aerial vehicle control method and device, unmanned aerial vehicle and storage medium |
| CN111582659A (en) * | 2020-04-16 | 2020-08-25 | 北京航空航天大学青岛研究院 | Mountain land operation difficulty index calculation method |
| CN111582659B (en) * | 2020-04-16 | 2023-09-19 | 北京航空航天大学青岛研究院 | Mountain work difficulty index calculation method |
| CN113807272A (en) * | 2021-09-22 | 2021-12-17 | 应急管理部国家减灾中心 | Method for rapidly extracting disaster situations of disaster damaged roads on remote sensing platform based on vector guidance |
| CN113807271A (en) * | 2021-09-22 | 2021-12-17 | 应急管理部国家减灾中心 | Road damage disaster rapid evaluation system based on remote sensing cloud service platform and application thereof |
| CN114332370A (en) * | 2021-12-28 | 2022-04-12 | 埃洛克航空科技（北京）有限公司 | Road image processing method, device, equipment and storage medium |
| CN114511551A (en) * | 2022-02-22 | 2022-05-17 | 金华高等研究院(金华理工学院筹建工作领导小组办公室) | Ground damage identification system based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109785307B (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109785307A (en) | A kind of unmanned plane image road Damage assessment method based on vector guidance | |
| CN107871119B (en) | Target detection method based on target space knowledge and two-stage prediction learning | |
| CN105528596B (en) | Utilize the high-resolution remote sensing image automatic building extraction method and system of shade | |
| CN112287807B (en) | Remote sensing image road extraction method based on multi-branch pyramid neural network | |
| CN106651886A (en) | Cloud image segmentation method based on superpixel clustering optimization CNN | |
| CN111986099A (en) | Tillage monitoring method and system based on convolutional neural network with residual error correction fused | |
| CN110163213B (en) | Remote sensing image segmentation method based on disparity map and multi-scale depth network model | |
| CN109447160B (en) | Method for automatically matching image and vector road intersection | |
| CN112801230B (en) | Intelligent acceptance method for unmanned aerial vehicle of power distribution line | |
| Hormese et al. | Automated road extraction from high resolution satellite images | |
| CN112560675B (en) | Bird visual target detection method combining YOLO and rotation-fusion strategy | |
| CN111626947A (en) | Map vectorization sample enhancement method and system based on generation of countermeasure network | |
| CN109685045A (en) | A kind of Moving Targets Based on Video Streams tracking and system | |
| Dan et al. | Automatic recognition of surface cracks in bridges based on 2D-APES and mobile machine vision | |
| AU2021100848A4 (en) | A Regional Extraction Method of Ecological Restoration Project in the Grassland Based on the High-resolution Remote Sensing Images | |
| CN110503637B (en) | Road crack automatic detection method based on convolutional neural network | |
| US20210027055A1 (en) | Methods and Systems for Identifying Topographic Features | |
| CN113313107B (en) | Intelligent detection and identification method for multiple types of diseases on cable surface of cable-stayed bridge | |
| WO2022267270A1 (en) | Crack characteristic representation method and system based on multi-fractal spectrum | |
| CN115861409B (en) | Soybean leaf area measuring and calculating method, system, computer equipment and storage medium | |
| CN105023272A (en) | Crop leaf insect pest detection method and system | |
| CN114565675A (en) | Method for removing dynamic feature points at front end of visual SLAM | |
| CN117197686A (en) | Satellite image-based high-standard farmland plot boundary automatic identification method | |
| CN104318240A (en) | Flower bud discriminating method based on computer vision | |
| CN113077438B (en) | Cell nucleus region extraction method and imaging method for multi-cell nucleus color image |
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 |