CN114693725A - 2.5-dimensional map automatic registration method based on image processing technology - Google Patents
2.5-dimensional map automatic registration method based on image processing technology Download PDFInfo
- Publication number
- CN114693725A CN114693725A CN202011462385.5A CN202011462385A CN114693725A CN 114693725 A CN114693725 A CN 114693725A CN 202011462385 A CN202011462385 A CN 202011462385A CN 114693725 A CN114693725 A CN 114693725A
- Authority
- CN
- China
- Prior art keywords
- registration
- calculation
- image
- points
- dimensional map
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 14
- 238000005516 engineering process Methods 0.000 title claims abstract description 5
- 238000004364 calculation method Methods 0.000 claims description 39
- 238000009877 rendering Methods 0.000 claims description 6
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 abstract description 5
- 239000000284 extract Substances 0.000 abstract 1
- 238000005457 optimization Methods 0.000 abstract 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010923 batch production Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/10—Geometric effects
- G06T15/20—Perspective computation
- G06T15/205—Image-based rendering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/285—Analysis of motion using a sequence of stereo image pairs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Remote Sensing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an automatic 2.5-dimensional map registration method based on an image processing technology, and belongs to the technical field of image processing technology and geographic information. The method adopts a computer graphics mode, automatically extracts the same characteristic points between the remote sensing image and the 2.5-dimensional map through the identification and matching of the image characteristic points, and then automatically generates 2.5-dimensional registration control points meeting the threshold requirement from the characteristic points through algorithm optimization, thereby realizing the automatic registration of the 2.5-dimensional map. The invention optimizes the original mode of relying on manual registration into the mode of automatically identifying the characteristic points and generating the control points by the computer for automatic registration. The method improves the registration precision and the registration efficiency in the 2.5-dimensional production flow and greatly improves the production progress of the 2.5-dimensional map.
Description
Technical Field
The invention relates to the field of computer graphics and geographic information science, in particular to a rapid registration method of a 2.5-dimensional map.
Background
The 2.5-dimensional electronic map inherits the powerful functions of the two-dimensional electronic map in the aspects of data acquisition, analysis, processing and the like, has intuitive and real space expression capability, and makes up the limitation of two-dimensional GIS graphic expression. Compared with a three-dimensional electronic map, the 2.5-dimensional electronic map has smaller data volume and faster background running.
In the current stage of 2.5-dimensional production, the 2.5-dimensional map needs to be registered so as to have correct geographic coordinates. The traditional registration method is to manually select feature points for registration, and the conventional registration method depends on the experience of registration personnel and the accuracy of an original base map, so that the registration accuracy and the registration efficiency are low, and the batch production cannot be met.
Disclosure of Invention
Aiming at the problems in the current 2.5-dimensional map production process, in order to improve the registration accuracy and the production efficiency of the 2.5-dimensional map, the invention adopts a computer graphics mode, and realizes the automatic registration process of large-batch 2.5-dimensional maps by optimizing the original dependence on manual registration into the automatic registration by adopting a computer.
The technical scheme adopted by the invention (see figure 1) is as follows:
s1: and rendering a remote sensing image as a registration reference base map at the same camera visual angle while rendering the 2.5-dimensional map.
S2: and extracting characteristic points between the rendered registration reference image and the original remote sensing image by using an image recognition algorithm.
S3: and (4) importing the rendered 2.5-dimensional image and the registered reference image into ArcGIS, and giving a coordinate system.
S4: and (4) importing the feature points generated in the step (S2) into ArcMap, operating a fitting algorithm to perform on the input feature points, and taking the control points obtained after algorithm calculation as final registration control points.
S4-1: inputting algorithm parameters: grid parameter n, threshold m, threshold step length s and maximum cycle number l.
S4-2: to ensure uniform distribution of control points, the image is divided evenly into a grid of N x N.
S4-3: and randomly selecting a characteristic point in each grid as a control point.
S4-4: fitting calculations are performed using a Least Squares Fit (LSF) algorithm on the selected points.
S4-5: after the fitting calculation is complete, the overall mean square error (Total RMS) is calculated for all control points.
S4-6: after calculation, whether the Total mean square error Total RMS is higher than the set threshold value m is judged.
S4-7: if the Total mean square error Total RMS is higher than the threshold value m, it is determined whether all feature points have already participated in the calculation.
S4-8: if the characteristic points do not participate in the calculation, the control point with the highest RMS value in the control points is removed, then another characteristic point which does not participate in the calculation is randomly selected as the control point, and the removed characteristic points do not participate in the subsequent calculation any more. The process continues to step S4-4, where the calculation is performed.
S4-9: if all feature points have already been involved in the calculation. It is determined whether the number of current cycles exceeds the input maximum number of cycles.
S4-10: and if the maximum cycle number is not exceeded, automatically adjusting the threshold value according to the input threshold value increasing step length s, wherein the specific rule is to add the input step length s to the existing threshold value to obtain a new threshold value m + s. And adds 1 to the cycle number, and then continues to step S4-3 to perform a new round of calculation.
If the combination of control points can make the Total RMS less than the threshold value m or reach the maximum cycle number l in the calculation. The calculation is aborted and the final calculation effect is shown in figure 2.
S5: the generated registration control points are used for image registration of a 2.5-dimensional map, so that the 2.5-dimensional map can be accurately registered on an image, and the image registration is shown in figure 3.
Drawings
FIG. 1: scheme flow chart
FIG. 2: display of extraction result of registration control point
FIG. 3: 2.5D map registration result display
Detailed Description
S1: and rendering a remote sensing image as a registration reference base map at the same camera visual angle while rendering the 2.5-dimensional map. The registration base map will serve as a reference image for image auto-matching.
S2: and extracting characteristic points between the rendered source image and the remote sensing image by using an image recognition tool. The number of feature points generated by the image recognition method is large, and meanwhile, many points which are recognized by mistake need to be removed.
S3: and importing the rendered 2.5-dimensional image and the registration base map into ArcGIS. And in ArcGIS, the rendered 2.5-dimensional map and the registered base map are endowed with the same coordinate system as the original remote sensing image.
S4: and (4) importing the characteristic points generated in the S2 into ArcMap, operating a fitting algorithm to optimize the input characteristic points, wherein the fitting algorithm comprises first-order fitting, second-order fitting and third-order fitting, and selecting a proper fitting algorithm according to actual needs. And extracting the Total Mean Square error (Total Root Mean Square) after the registration, wherein the Total Root Mean Square error is lower than a required threshold value m, or the maximum cycle number l is reached, and finishing the calculation. In order to ensure the registration accuracy of the whole picture, the registration control points must be uniformly distributed in the whole image range. And taking the control point obtained by algorithm calculation as a final registration control point.
S4-1: parameters of the input algorithm: grid parameter n, threshold m, threshold step length s, and maximum cycle number 1.
S4-2: to ensure uniform distribution of control points, the image is divided evenly into a grid of N x N. The larger N is, the higher the registration accuracy is, but the longer the time required for simultaneous calculation is, and in general, N is 4, which can satisfy the minimum required number of control points of various registration methods.
S4-3: and randomly selecting a characteristic point in each grid as a control point.
S4-4: fitting calculations are performed using a Least Squares Fit (LSF) algorithm on the selected points.
S4-5: after the fitting calculation is complete, Total RMS is calculated for all control points. The calculation formula is as follows:
s4-6: after calculation, it is determined whether the Total RMS is higher than the set threshold value m.
S4-7: if the Total RMS is higher than the threshold value m, judging whether all the characteristic points are already existed
The participation is calculated.
S4-8: if the characteristic points do not participate, the control points with the highest RMS value in the control points are removed, then another characteristic point which does not participate in calculation is randomly selected as a control point, and the removed characteristic points do not participate in subsequent calculation. The process continues to step S4-4, where the calculation is performed.
S4-9: if all feature points have already been involved in the calculation. It is determined whether the number of current cycles exceeds the input maximum number of cycles.
S4-10: if the maximum cycle number is not exceeded, the threshold is automatically adjusted according to the input threshold increasing step length s, and the specific rule is that the input step length s is added to the existing threshold to obtain a new threshold m + s. And adds 1 to the cycle number, and then continues to step S4-3 to perform a new round of calculation.
If the combination of control points can make the Total RMS less than the threshold value m or reach the maximum cycle number l in the calculation. The calculation is aborted.
S5: and the generated registration control points are used for image registration of the 2.5-dimensional map, so that the 2.5-dimensional map can be accurately registered on the image.
Claims (1)
1. A2.5-dimensional map automatic registration method based on image processing technology is characterized in that:
s1: rendering a remote sensing image as a registration reference base map at the same camera visual angle while rendering the 2.5-dimensional map;
s2: extracting feature points between the rendered registration reference image and the original remote sensing image by using an image recognition algorithm;
s3: introducing the rendered 2.5-dimensional image and the registration reference image into ArcGIS, and giving a coordinate system;
s4: importing the feature points generated in the step S2 into ArcMap, operating a fitting algorithm to perform on the input feature points, and taking the control points obtained after algorithm calculation as final registration control points;
s4-1: inputting algorithm parameters: grid parameter n, threshold m, threshold step length s and maximum cycle number l;
s4-2: in order to ensure the uniform distribution of the control points, the image is divided into N-by-N grids on average;
s4-3: randomly selecting a characteristic point in each grid as a control point;
s4-4: performing fitting calculation on the selected points by using a Least Square Fitting (LSF) algorithm;
s4-5: after the fitting calculation is completed, calculating the Total mean square error (Total RMS) of all the control points;
s4-6: after calculation, judging whether the Total mean square error (Total RMS) is higher than a set threshold value m;
s4-7: if the Total mean square error Total RMS is higher than the threshold value m, judging whether all the feature points participate in calculation;
s4-8: if the characteristic points do not participate in the calculation, the control point with the highest RMS value in the control points is removed, then another characteristic point which does not participate in the calculation is randomly selected as the control point, and the removed characteristic points do not participate in the subsequent calculation any more. Continuing to step S4-4, calculating;
s4-9: if all feature points have already been involved in the calculation. Judging whether the current cycle times exceed the input maximum cycle times;
s4-10: and if the maximum cycle number is not exceeded, automatically adjusting the threshold value according to the input threshold value increasing step length s, wherein the specific rule is to add the input step length s to the existing threshold value to obtain a new threshold value m + s. Adding 1 to the cycle number, and continuing to the step S4-3 to calculate a new round;
if the combination of control points can make the Total RMS less than the threshold value m or reach the maximum cycle number l in the calculation. Stopping the calculation, and finally calculating the effect as shown in figure 2;
s5: the generated registration control points are used for image registration of the 2.5-dimensional map, so that the 2.5-dimensional map can be accurately registered on the image, and the image is shown in figure 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011462385.5A CN114693725A (en) | 2020-12-15 | 2020-12-15 | 2.5-dimensional map automatic registration method based on image processing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011462385.5A CN114693725A (en) | 2020-12-15 | 2020-12-15 | 2.5-dimensional map automatic registration method based on image processing technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114693725A true CN114693725A (en) | 2022-07-01 |
Family
ID=82130026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011462385.5A Pending CN114693725A (en) | 2020-12-15 | 2020-12-15 | 2.5-dimensional map automatic registration method based on image processing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114693725A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065311A (en) * | 2012-12-26 | 2013-04-24 | 中国土地勘测规划院 | Satellite image automatic registration method based on standard image database |
| CN104599277A (en) * | 2015-01-27 | 2015-05-06 | 中国科学院空间科学与应用研究中心 | Image registration method for area-preserving affine transformation |
| CN111161413A (en) * | 2019-12-20 | 2020-05-15 | 东南大学 | Construction method of three-dimensional virtual airport platform based on GIS |
-
2020
- 2020-12-15 CN CN202011462385.5A patent/CN114693725A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065311A (en) * | 2012-12-26 | 2013-04-24 | 中国土地勘测规划院 | Satellite image automatic registration method based on standard image database |
| CN104599277A (en) * | 2015-01-27 | 2015-05-06 | 中国科学院空间科学与应用研究中心 | Image registration method for area-preserving affine transformation |
| CN111161413A (en) * | 2019-12-20 | 2020-05-15 | 东南大学 | Construction method of three-dimensional virtual airport platform based on GIS |
Non-Patent Citations (4)
| Title |
|---|
| 张继贤, 李国胜, 曾钰: "多源遥感影像高精度自动配准的方法研究", 遥感学报, no. 01, 10 February 2005 (2005-02-10) * |
| 张谦;贾永红;胡忠文;: "多源遥感影像配准中的SIFT特征匹配改进", 武汉大学学报(信息科学版), no. 04, 5 April 2013 (2013-04-05) * |
| 朱贵方;张西军;: "利用ENVI与ArcGIS一体化快速生产高精度数字正射影像技术", 测绘与空间地理信息, no. 12, 25 December 2019 (2019-12-25), pages 139 - 142 * |
| 黄志坚;张金芳;: "基于图像配准的高分辨率遥感图像道路追踪", 电子技术, no. 02, 25 February 2010 (2010-02-25) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111612880B (en) | Three-dimensional model construction method based on two-dimensional drawing, electronic equipment and storage medium | |
| CN109241580B (en) | Land block design method and device, computer equipment and storage medium | |
| CN111177814B (en) | Transformer substation engineering three-dimensional modeling method and system based on DEM data | |
| CN105068918A (en) | Page test method and device | |
| CN111275821A (en) | Power line fitting method, system and terminal | |
| CN111754431B (en) | Image area replacement method, device, equipment and storage medium | |
| CN117745955B (en) | Method and device for generating urban building scene based on building base vector data | |
| CN115098717A (en) | Three-dimensional model retrieval method and device, electronic equipment and storage medium | |
| CN114693725A (en) | 2.5-dimensional map automatic registration method based on image processing technology | |
| CN116452604B (en) | Complex substation scene segmentation method, device and storage medium | |
| CN115409951B (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN111898525B (en) | Construction method of smoke identification model, and method and device for detecting smoke | |
| CN111639691B (en) | Image data sampling method based on feature matching and greedy search | |
| CN111192363B (en) | User power distribution room design generation method based on cloud computing | |
| CN109509234B (en) | Urban building group high-definition grid map vectorization-based method | |
| CN112487124A (en) | Method for converting dot-shaped elements in CorelDraw map into SuperMap by using VBA | |
| CN114120011B (en) | Method for improving image matching precision and speed by using group intelligent algorithm | |
| CN114972891B (en) | Automatic identification method for CAD (computer aided design) component and BIM (building information modeling) method | |
| CN112989541B (en) | Three-dimensional grid model generation method and device, electronic equipment and readable storage medium | |
| CN110795491B (en) | Structure parameter information visualization method | |
| CN115146349B (en) | Method and device for locally updating design | |
| CN113627437A (en) | Method for rapidly identifying graphite electrode label | |
| CN111062921B (en) | Ceramic tile inspection method for plane design drawing | |
| CN114266953B (en) | Target detection flow self-optimization method based on deep learning | |
| Yang et al. | Intelligent Human-machine Game System Based on Machine Vision |
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 |