CN106776979A - Vector Electronic Map increased quality automatic mode based on remote sensing - Google Patents

Vector Electronic Map increased quality automatic mode based on remote sensing Download PDF

Info

Publication number
CN106776979A
CN106776979A CN201611109034.XA CN201611109034A CN106776979A CN 106776979 A CN106776979 A CN 106776979A CN 201611109034 A CN201611109034 A CN 201611109034A CN 106776979 A CN106776979 A CN 106776979A
Authority
CN
China
Prior art keywords
electronic map
same place
image
remote sensing
increased quality
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.)
Pending
Application number
CN201611109034.XA
Other languages
Chinese (zh)
Inventor
李先怡
孔祥军
李月华
胡磊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Science Mapuniverse Tchndogy Co Ltd
Original Assignee
China Science Mapuniverse Tchndogy Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Science Mapuniverse Tchndogy Co Ltd filed Critical China Science Mapuniverse Tchndogy Co Ltd
Priority to CN201611109034.XA priority Critical patent/CN106776979A/en
Publication of CN106776979A publication Critical patent/CN106776979A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/29—Geographical information databases

Abstract

The invention discloses a kind of Vector Electronic Map increased quality automatic mode based on remote sensing, it is related to spatial data handling technical field.Methods described：Obtain image Auto-matching same place after raw video and precision improvement；Model is set up on the basis of same place after matching, setting threshold value carries out erroneous point rejecting；Conversion identical point coordinates system, and be a file A by multiple same place Piece file mergences in any one region；Vector Electronic Map is corrected with file A, obtains the Vector Electronic Map after increased quality, and with precision improvement after image set and, complete to be based on the Vector Electronic Map increased quality of remote sensing.The present invention overcomes the artificial operation of existing method, the problem of inefficiency, and without any additional information and manual intervention, high degree of automation, and set and precision meet effect.

Description

Vector Electronic Map increased quality automatic mode based on remote sensing
Technical field
The present invention relates to spatial data handling technical field, more particularly to a kind of Vector Electronic Map quality based on remote sensing Lifting automatic mode.
Background technology
GIS geographical spatial datas are the foundation stones of spatial Information Service platform, and its quality of data directly affects user to space The use feeling of information service platform.There is certain error skew in annual updating the data, cause existing base map between initial data There is mismatch in data, directly affect industry production and service application with electronic map data is updated.Therefore the quality of data is lifted It is the basic business of GIS geographical spatial datas.
Existing Vector Electronic Map increased quality technology path, the image based on spatial accuracy lifting carries out vector data Subregional spatial accuracy lifting.When vector data and remote sensing image data not exclusively set and when, flow of rectifying a deviation：First existing Choose same place on image after image and precision improvement by hand, image data is same after the existing vector data of generation and precision improvement The error line of famous cake；Then according to error line calculation error distance and direction of error, based on error distance and direction of error Carry out space clustering and form error homogenous area；To different zones, geometric correction is carried out according to error line.
Traditional vector image set and method are processed in ARCGIS softwares, and artificial operation repeats to import image, with vision Experience chooses same place by hand, and then utilization space is corrected function and vector data is rectified a deviation.Obvious existing method is existed Automaticity is low, and manual intervention is more, the defect such as artificial disturbance factor is big, it is difficult to improve production efficiency.
The content of the invention
It is an object of the invention to provide a kind of Vector Electronic Map increased quality automatic mode based on remote sensing, so that Solve foregoing problems present in prior art.
To achieve these goals, the Vector Electronic Map increased quality automatic mode based on remote sensing of the present invention, Methods described includes：
S1, obtains image Auto-matching same place after raw video and precision improvement；
S2, sets up model on the basis of the same place after matching, setting threshold value carries out erroneous point rejecting；
S3, conversion identical point coordinates system, and be a file A by multiple same place Piece file mergences in any one region；
S4, is corrected with file A to Vector Electronic Map, obtains the Vector Electronic Map after increased quality, and with essence Image set and Vector Electronic Map increased quality of the completion based on remote sensing after degree lifting.
Preferably, step S1, automatically generates match point, specially using based on area grayscale matching process：Compare original The gray value of image same local area after image and precision improvement, judges whether gray value similarity degree meets what is pre-set Threshold value, if it is, obtaining match point, continues more next regional area；If it is not, then to find match point, continuing to compare More next regional area, finally gives multipair same place pair.
Preferably, step S1, specifically realizes as steps described below：
S11, using image after raw video and precision improvement one as left image, another is used as right image；
S12, selectes any one point to be located W as impact point in the left image, with impact point as terminal, chooses ash Degree array is used as target area；
S13, sets up a gray scale array more than the target area as the field of search on right image；Then exist successively The gray scale array big with the target area etc. is chosen in the field of search as search window, the search window and the mesh is calculated The similarity measure in region is marked, when similarity measure A is the maximum in all similarity measures, with the similarity measure The center pixel of the corresponding search window A of A as point to be located W same place.
It is highly preferred that it is correlation coefficient process that the measure of similarity measure is adopted.
Preferably, in step S1, carrying out the parameter that Auto-matching same place is related to includes：Matching process, same place Several threshold value, least correlativing coefficient, moving window size and search box size.
Preferably, step S2, specifically realizes as steps described below：
S21, sets up the transformation model between image after raw video and precision improvement, on the basis of same place, uses Least square multinomial, uses inverse matrix computation model coefficient；
Idiographic flow：Using image after raw video and precision improvement one as left image, another as right image, The same place of left and right two width image, to being expressed as (x, y), (x ', y '), is formula (1) according to least square multinomial, is used Inverse matrix computation model coefficient：
a1, b1, c1, a2, b2, c2It is the deformation coefficient of transformation model；
S22, calculates future position position, i.e., according to the deformation coefficient a of the transformation model for calculating by transformation model1, b1, c1, a2, b2, c2, and formula (2) is combined, left image identical point coordinates (x, y) is converted into right image, as future position position (x0,y0)；
S23, future position (x is calculated using Euclidean distance formula (3)0,y0) and actual match same place (x ', y ') error Apart from dij；
S24, sets the threshold value of error distance, when certain is more than advance threshold value to same place to the error distance being calculated When, remove this pair of same place；Remaining point set of the same name is required point set of the same name.
Preferably, step S3, specifically realizes as steps described below：
S31, the pixel planes coordinate transformation of same place is the earth of the longitude and latitude of same place by the conversion of identical point coordinates system Coordinate system；
S32, multiple dot files of the same name are spliced into a region-wide dot file of the same name, specially：Programming realization is opened many Individual dot file of the same name, reads file content, and reading of content is written in a newly-built file one by one.
Preferably, step S4, with the rubber drawing process registration vector data coordinate in file A free-air corrections and adjustment Shape is adopted and Vector Electronic Map is corrected, and the Vector Electronic Map after correction is covered and consistent with the image of precision improvement.
Preferably, the Vector Electronic Map engineer's scale is not limited.
The beneficial effects of the invention are as follows：
The present invention proposes a kind of automation side of the electronic map quality lifting for Vector Electronic Map and image set sum Method, a whole set of technological process of production uses programming realization, overcomes the artificial operation of existing method, the problem of inefficiency, without appointing What additional information and manual intervention, high degree of automation, and set and precision meet effect, its result region be directly facing application.
Brief description of the drawings
Fig. 1 is the schematic flow sheet one of the Vector Electronic Map increased quality automatic mode based on remote sensing；
Fig. 2 is the schematic flow sheet two of the Vector Electronic Map increased quality automatic mode based on remote sensing.
Specific embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, below in conjunction with accompanying drawing, the present invention is entered Row is further described.It should be appreciated that specific embodiment described herein is only used to explain the present invention, it is not used to Limit the present invention.
Vector Electronic Map increased quality automatic mode based on remote sensing described in the present embodiment, methods described includes：
Image Auto-matching same place after S1, raw video and precision improvement；
Match point is automatically generated using based on area grayscale matching process, specially：Compare raw video and precision improvement The gray value of image same local area, judges whether gray value similarity degree meets the threshold value for pre-setting afterwards, if it is, Match point is obtained, continues more next regional area；If it is not, then to find match point, continuing more next partial zones Domain, finally gives multipair same place pair.
S2, sets up model on the basis of the same place after matching, setting threshold value carries out erroneous point rejecting；
S3, conversion identical point coordinates system, and be a file A by multiple same place Piece file mergences in any one region；
S4, is corrected with file A to Vector Electronic Map, obtains the Vector Electronic Map after increased quality, and with essence Image set and Vector Electronic Map increased quality of the completion based on remote sensing after degree lifting.
Explanation is explained in more detail is：
(1) step S1, specifically realizes as steps described below：
S11, using image after raw video and precision improvement one as left image, another is used as right image；
S12, selectes any one point to be located W as impact point in the left image, with impact point as terminal, chooses ash Degree array is used as target area；
S13, sets up a gray scale array more than the target area as the field of search on right image；Then exist successively The gray scale array big with the target area etc. is chosen in the field of search as search window, the search window and the mesh is calculated The similarity measure in region is marked, when similarity measure A is the maximum in all similarity measures, with the similarity measure The center pixel of the corresponding search window A of A as point to be located W same place.It is phase relation that the measure of similarity measure is adopted Number method, i.e.,：
Wherein
Normalization canonical correlation coefficient formula (2) is derived on the basis of formula (1)：
In formula (2)WithThe average of vector x and y is represented respectively.
In step S1, carrying out the parameter that Auto-matching same place is related to includes：The threshold of matching process, same place number Value, least correlativing coefficient, moving window size and search plain window size.
(2) step S2, specifically realizes as steps described below：
S21, sets up the transformation model between image after raw video and precision improvement, on the basis of same place, uses Least square multinomial, uses inverse matrix computation model coefficient；
Idiographic flow：Using image after raw video and precision improvement one as left image, another as right image, The same place of left and right two width image, to being expressed as (x, y), (x ', y '), is formula (3) according to least square multinomial, is used Inverse matrix computation model coefficient：
a1, b1, c1, a2, b2, c2It is the deformation coefficient of transformation model；
Because same place is larger to data volume and there is certain error, the thought of fitting is introduced so that in all data points The residual error at place is all smaller, i.e., all error combinations are minimum, i.e.,Determine that fitting is more using mentioned above principle The method of item formula is least square method fitting of a polynomial, determines that above-mentioned polynomial process namely determines the coefficient in multinomial Process.
S22, calculates future position position, i.e., according to the deformation coefficient a of the transformation model for calculating by transformation model1, b1, c1, a2, b2, c2, and formula (4) is combined, left image identical point coordinates (x, y) is converted into right image, as future position position (x0,y0)；
S23, future position (x is calculated using Euclidean distance formula (5)0,y0) and actual match same place (x ', y ') error Apart from dij；
S24, sets the threshold value of error distance, when certain is more than advance threshold value to same place to the error distance being calculated When, remove this pair of same place；Remaining point set of the same name is required point set of the same name.
(3) step S3, specifically realizes as steps described below：
S31, the pixel planes coordinate transformation of same place is the earth of the longitude and latitude of same place by the conversion of identical point coordinates system Coordinate system；
S32, multiple dot files of the same name are spliced into a region-wide dot file of the same name, specially：Programming realization is opened many Individual dot file of the same name, reads file content, and reading of content is written in a newly-built file one by one.
Each pair image same place is preserved in step S3, all Image Matching points produce a file；Due to whole provinces and regions or The image framing difference Corresponding matching generation same place in bigger region, so needing to ensure that whole region is a same place text Part.
(4) step S4, with rubber drawing process registration vector data coordinate and adjustment shape in file A free-air corrections Shape is adopted and Vector Electronic Map is corrected, and the Vector Electronic Map after correction is covered and consistent with the image of precision improvement.Institute Vector Electronic Map engineer's scale is stated not limit.
(5) image set and consistency check after vector and precision improvement after correcting；
Vector image covers and consistency check, and using remote sensing image as control base map, artificial visual travels through figure layer inspection.By In the even element vector data wide coverage in whole provincial region, vector element complexity is various, in order to accurate evaluation saves overall vector The quality of data, the method that random sampling is taken in inspection.
VectorLayer space covers and accuracy checking main contents include：1. figure layer key element is overall inclined with respect to remote sensing image Move；2. irregular skew of the figure layer key element relative to remote sensing image.Inspection result is that Vector Electronic Map covers substantially with remote sensing image With.
By using above-mentioned technical proposal disclosed by the invention, following beneficial effect has been obtained：
The present invention proposes a kind of automation side of the electronic map quality lifting for Vector Electronic Map and image set sum Method, a whole set of technological process of production uses programming realization, overcomes the artificial operation of existing method, the problem of inefficiency, without appointing What additional information and manual intervention, high degree of automation, and set and precision meet effect, its result region be directly facing application.
The above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should Depending on protection scope of the present invention.

Claims (9)

1. a kind of Vector Electronic Map increased quality automatic mode based on remote sensing, it is characterised in that methods described includes：
S1, obtains image Auto-matching same place after raw video and precision improvement；
S2, sets up model on the basis of the same place after matching, setting threshold value carries out erroneous point rejecting；
S3, conversion identical point coordinates system, and be a file A by multiple same place Piece file mergences in any one region；
S4, is corrected with file A to Vector Electronic Map, obtains the Vector Electronic Map after increased quality, and carry with precision Image set and Vector Electronic Map increased quality of the completion based on remote sensing after rising.
2. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 1, it is characterised in that step Rapid S1, automatically generates match point, specially using based on area grayscale matching process：Compare shadow after raw video and precision improvement As the gray value of same local area, judge whether gray value similarity degree meets the threshold value for pre-setting, if it is, obtaining Match point, continues more next regional area；If it is not, then to find match point, continuing more next regional area, most Multipair same place pair is obtained eventually.
3. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 1, it is characterised in that step Rapid S1, specifically realizes as steps described below：
S11, using image after raw video and precision improvement one as left image, another is used as right image；
S12, selectes any one point to be located W as impact point in the left image, with impact point as terminal, chooses gray scale battle array Row are used as target area；
S13, sets up a gray scale array more than the target area as the field of search on right image；Then successively in search The gray scale array big with the target area etc. is chosen in area as search window, the search window and the target area is calculated The similarity measure in domain, when similarity measure A is the maximum in all similarity measures, with similarity measure A pairs The center pixel of the search window A for answering as point to be located W same place.
4. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 3, it is characterised in that phase It is correlation coefficient process that the measure estimated like property is adopted.
5. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 1, it is characterised in that step In rapid S1, carrying out the parameter that Auto-matching same place is related to includes：Matching process, the threshold value of same place number, minimum correlation Coefficient, moving window size and search box size.
6. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 1, it is characterised in that step Rapid S2, specifically realizes as steps described below：
S21, sets up the transformation model between image after raw video and precision improvement, on the basis of same place, using minimum Two multiply multinomial, use inverse matrix computation model coefficient；
Idiographic flow：Using image after raw video and precision improvement one as left image, another is left and right as right image The same place of two width images, to being expressed as (x, y), (x ', y '), is formula (1) according to least square multinomial, uses matrix Inversion Calculation model coefficient：
x ′ = a 1 + b 1 x + c 1 x 2 y ′ = a 2 + b 2 y + c 2 y 2 - - - ( 1 ) ,
a1, b1, c1, a2, b2, c2It is the deformation coefficient of transformation model；
S22, calculates future position position, i.e., according to the deformation coefficient a of the transformation model for calculating by transformation model1, b1, c1, a2, b2, c2, and formula (2) is combined, left image identical point coordinates (x, y) is converted into right image, as future position position (x0, y0)；
x 0 = a 1 + b 1 x + c 1 x 2 y 0 = a 2 + b 2 y + c 2 y 2 - - - ( 2 ) ;
S23, future position (x is calculated using Euclidean distance formula (3)0,y0) and actual match same place (x ', y ') error distance dij；
d i j = ( x 0 - x ′ ) 2 + ( y 0 - y ′ ) 2 - - - ( 3 ) ;
S24, sets the threshold value of error distance, when certain is more than advance threshold value to same place to the error distance being calculated, goes Except this pair of same place；Remaining point set of the same name is required point set of the same name.
7. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 1, it is characterised in that step Rapid S3, specifically realizes as steps described below：
S31, the pixel planes coordinate transformation of same place is the geodetic coordinates of the longitude and latitude of same place by the conversion of identical point coordinates system System；
S32, multiple dot files of the same name are spliced into a region-wide dot file of the same name, specially：Programming realization is opened multiple same Name dot file, reads file content, and reading of content is written in a newly-built file one by one.
8. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 1, it is characterised in that step Rapid S4, is adopted to vector electronics with the rubber drawing process registration vector data coordinate and adjustment shape in file A free-air corrections Map is corrected, and the Vector Electronic Map after correction is covered and consistent with the image of precision improvement.
9. the Vector Electronic Map increased quality automatic mode of remote sensing is based on according to claim 1, it is characterised in that institute Vector Electronic Map engineer's scale is stated not limit.
CN201611109034.XA 2016-12-06 2016-12-06 Vector Electronic Map increased quality automatic mode based on remote sensing Pending CN106776979A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611109034.XA CN106776979A (en) 2016-12-06 2016-12-06 Vector Electronic Map increased quality automatic mode based on remote sensing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611109034.XA CN106776979A (en) 2016-12-06 2016-12-06 Vector Electronic Map increased quality automatic mode based on remote sensing

Publications (1)

Publication Number Publication Date
CN106776979A true CN106776979A (en) 2017-05-31

Family

ID=58878350

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611109034.XA Pending CN106776979A (en) 2016-12-06 2016-12-06 Vector Electronic Map increased quality automatic mode based on remote sensing

Country Status (1)

Country Link
CN (1) CN106776979A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110675338A (en) * 2019-09-09 2020-01-10 武汉大学 Automatic vector data correction method based on multiple images
CN111462072A (en) * 2020-03-30 2020-07-28 北京百度网讯科技有限公司 Dot cloud picture quality detection method and device and electronic equipment
CN111683221A (en) * 2020-05-21 2020-09-18 武汉大学 Real-time video monitoring method and system for natural resources embedded with vector red line data
CN114581556A (en) * 2022-03-10 2022-06-03 青海省地质调查院 Digital map filling method in regional geological survey
CN115357675A (en) * 2022-08-18 2022-11-18 自然资源部国土卫星遥感应用中心 Method and system for establishing image control point database through image control point standardization processing

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102693542A (en) * 2012-05-18 2012-09-26 中国人民解放军信息工程大学 Image characteristic matching method
CN104021556A (en) * 2014-06-13 2014-09-03 西南交通大学 Heterological remote-sensing image registration method based on geometric structure similarity

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102693542A (en) * 2012-05-18 2012-09-26 中国人民解放军信息工程大学 Image characteristic matching method
CN104021556A (en) * 2014-06-13 2014-09-03 西南交通大学 Heterological remote-sensing image registration method based on geometric structure similarity

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
冯莞舒: "基于特征点的影像匹配", 《电子测试》 *
李月华: "基于遥感的矢量电子地图质量提升自动化方法研究", 《北京测绘》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110675338A (en) * 2019-09-09 2020-01-10 武汉大学 Automatic vector data correction method based on multiple images
CN111462072A (en) * 2020-03-30 2020-07-28 北京百度网讯科技有限公司 Dot cloud picture quality detection method and device and electronic equipment
CN111462072B (en) * 2020-03-30 2023-08-29 北京百度网讯科技有限公司 Point cloud picture quality detection method and device and electronic equipment
CN111683221A (en) * 2020-05-21 2020-09-18 武汉大学 Real-time video monitoring method and system for natural resources embedded with vector red line data
CN114581556A (en) * 2022-03-10 2022-06-03 青海省地质调查院 Digital map filling method in regional geological survey
CN115357675A (en) * 2022-08-18 2022-11-18 自然资源部国土卫星遥感应用中心 Method and system for establishing image control point database through image control point standardization processing
CN115357675B (en) * 2022-08-18 2023-04-14 自然资源部国土卫星遥感应用中心 Method and system for establishing image control point database through standardized processing of image control points

Similar Documents

Publication Publication Date Title
CN106776979A (en) Vector Electronic Map increased quality automatic mode based on remote sensing
CN104484648B (en) Robot variable visual angle obstacle detection method based on outline identification
CN108960135B (en) Dense ship target accurate detection method based on high-resolution remote sensing image
CN112084869B (en) Compact quadrilateral representation-based building target detection method
CN105389817B (en) A kind of two phase remote sensing image variation detection methods
CN104134220A (en) Low-altitude remote sensing image high-precision matching method with consistent image space
CN101901343A (en) Remote sensing image road extracting method based on stereo constraint
CN105160686B (en) A kind of low latitude various visual angles Remote Sensing Images Matching Method based on improvement SIFT operators
CN102034101A (en) Method for quickly positioning circular mark in PCB visual detection
CN112819871B (en) Table image registration method based on straight line segmentation
CN104732531B (en) A kind of high-resolution remote sensing image signal to noise ratio curve self-adapting acquisition methods
CN113344956B (en) Ground feature contour extraction and classification method based on unmanned aerial vehicle aerial photography three-dimensional modeling
CN103295239A (en) Laser-point cloud data automatic registration method based on plane base images
CN103646395B (en) A kind of High-precision image method for registering based on grid method
CN104166989B (en) Rapid ICP method for two-dimensional laser radar point cloud matching
CN109345513B (en) Cigarette package defect detection method with cigarette package posture calculation function
CN106920235A (en) Star-loaded optical remote sensing image automatic correction method based on the matching of vector base map
CN106295503A (en) The high-resolution remote sensing image Ship Target extracting method of region convolutional neural networks
CN105139355A (en) Method for enhancing depth images
CN107958443A (en) A kind of fingerprint image joining method based on crestal line feature and TPS deformation models
CN104899892A (en) Method for quickly extracting star points from star images
CN115147418B (en) Compression training method and device for defect detection model
CN105354845B (en) A kind of semi-supervised change detecting method of remote sensing image
CN103854271B (en) A kind of planar pickup machine scaling method
CN107328371A (en) Sub-pix contours extract based on Gaussian and the optimization using Softmax recurrence in the case where metal plate detects scene

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20170531

RJ01 Rejection of invention patent application after publication