CN113405533A - Aerial triangulation operation method for aerial images - Google Patents
Aerial triangulation operation method for aerial images Download PDFInfo
- Publication number
- CN113405533A CN113405533A CN202110583475.8A CN202110583475A CN113405533A CN 113405533 A CN113405533 A CN 113405533A CN 202110583475 A CN202110583475 A CN 202110583475A CN 113405533 A CN113405533 A CN 113405533A
- Authority
- CN
- China
- Prior art keywords
- image
- sub
- aerial
- images
- triangulation
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
- G01C11/30—Interpretation of pictures by triangulation
- G01C11/34—Aerial triangulation
Abstract
The invention discloses an aerial triangulation operation method for aerial images, which relates to the field of aerial image measurement and aims to solve the problems of more equipment and long calculation time required by aerial triangulation and comprises the following steps: dividing a plurality of sub-measuring areas according to the measuring area range and the terrain condition, wherein the number of the divided sub-measuring areas is N; performing image extraction on the sub-measurement areas through aerial photography, performing aerial triangulation processing on the extracted sub-measurement area images, and extracting images of the rest sub-measurement areas while performing aerial triangulation processing until the extraction and processing of the images of all the sub-measurement areas are finished; merging the processed data of the plurality of sub-measurement areas to obtain a complete measurement area; the aerial image aerial triangulation operation method with less computer equipment cost and shorter construction period is realized.
Description
Technical Field
The invention relates to the field of aerial image measurement, in particular to an aerial image aerial triangulation operation method.
Background
The existing aerial image aerial triangulation operation generally adopts the following scheme: firstly, making a flight plan according to a survey area range and a terrain condition; then, according to the flight plan, carrying out aerial photography on the spot by using an unmanned aerial vehicle to obtain an aerial image of the survey area; and finally, after all the areas are subjected to aerial photography, performing data arrangement on the images of the whole measuring area, and then starting to perform aerial triangulation interior work processing. In the technical scheme, aerial photography and interior work processing can be executed in series, namely the interior work processing can be started to be executed after the aerial photography is finished, and the method has the following defects: the completion of a project requires a long construction period, because the range of the unmanned aerial vehicle shooting in one day is limited, and the shooting of the whole measuring area range is generally required to be completed within several days or even one or two months. During this period, the computers responsible for the internal processing can only be in a waiting state, which causes waste of time and resources; more computer equipment is required to complete a project because the project is typically tight, and more computer equipment must be deployed to make delivery on time, increasing the effort and reducing the time of the industry process.
By improving the operation scheme and the technical method, the aerial triangulation project of the aerial image can be completed by fewer computer equipment and shorter construction period.
Disclosure of Invention
The invention discloses an aerial triangulation operation method for aerial images, and aims to solve the problems of more required equipment and long calculation time of aerial triangulation.
In order to solve the problems, the invention adopts the following technical scheme:
an aerial triangulation operation method for aerial images comprises the following steps:
step 1: dividing a plurality of sub-measuring areas according to the measuring area range and the terrain condition, wherein the number of the divided sub-measuring areas is N;
step 2: performing image extraction on the sub-measurement areas through aerial photography, and sequentially performing aerial triangulation processing on the extracted sub-measurement area images;
and step 3: and merging the processed data of the plurality of sub-measurement areas to obtain a complete measurement area.
Preferably, the step 3 comprises the steps of:
step 301: arranging the triangulation results of each sub-measurement area, numbering each camera, each image and each triangulation connection point, wherein the numbering is not repeated;
step 302: obtaining the number M of adjacent images of each image; the method for obtaining the number of the adjacent images of each image comprises the following steps:
using the feature point information of the image to perform image retrieval, and k1 neighboring images before the image retrieval; if the image has GPS information, calculating the Euclidean distance between the image and each image in the sub-measurement area by using the GPS information, and reserving the previous k2 images with the nearest distance as neighbor images of the image, wherein the number of the neighbor images of the image is M, and M is k1+ k 2;
step 303: judging whether the image is a boundary image between the sub-measurement areas;
step 304: performing image feature matching on all boundary images and adjacent images thereof to form an adjacency matrix;
step 305: partitioning all the adjacency relation matrixes obtained in the step 304, independently numbering the partitions, wherein the number of the partitions is P, and performing parallel triangulation processing on the P sub-measurement areas to obtain triangulation results of the P sub-measurement areas;
step 306: and combining the triangulation results of the N sub-measurement areas, the P boundary images and the adjacent image matching areas thereof to form an aerial triangulation result of the whole measurement area.
Preferably, the method for determining whether the image is a boundary image between sub-regions in step 303 includes: for each image, its M neighboring images are examined, and if at least one neighboring image originates from a different sub-area, this image belongs to a boundary image.
Preferably, the independent numbering in step 305 starts with N +1, and each next numbering is incremented by 1.
According to the operation scheme provided by the patent, the measurement areas are divided, so that the field processing can be started without waiting for the completion of aerial photography of the whole measurement area, but the field processing can be started at the same time in the aerial photography stage, so that the time of the aerial photography stage is fully utilized, and the project construction period can be greatly shortened; the number of computers required for image processing and boundary processing is not large; in sum, the aerial triangulation project of aerial image can be completed by fewer computer equipment and shorter construction period.
Drawings
FIG. 1 is a schematic flow chart of example 1.
Detailed Description
Example 1
The invention discloses an aerial triangulation operation method for aerial images, which comprises the following steps:
step 1: dividing a plurality of sub-measuring areas according to the measuring area range and the terrain condition, wherein the number of the divided sub-measuring areas is N;
step 2: performing image extraction on the sub-measurement areas through aerial photography, and sequentially performing aerial triangulation processing on the extracted sub-measurement area images;
in step 2 of this embodiment, the image extraction and triangulation processing work is performed in a relay manner at the same time, for example, after the aerial photography of the first day is finished, the images taken and extracted on that day are submitted as the first sub-measurement area, the aerial triangulation processing is immediately started, and the internal processing of the images is completed by using the time of the evening of the day and the time of the day of the second day; continuing aerial photography in the daytime in the next day, and submitting the images shot in the day to the interior as a second sub-measurement area after the aerial photography is finished, and starting aerial triangulation processing; and finishing the internal processing of the images by using the time of the night of the day and the day of the next day, and repeating the steps until the extraction and measurement operations of all the sub-measurement areas are finished.
And step 3: and merging the processed data of the plurality of sub-measurement areas to obtain a complete measurement area.
Preferably, step 3 comprises the steps of:
step 301: arranging the triangulation results of each sub-measurement area, numbering each camera, each image and each triangulation connection point, wherein the numbering is not repeated;
step 302: obtaining the number M of adjacent images of each image; the method for obtaining the number of the adjacent images of each image comprises the following steps:
using the feature point information of the image to perform image retrieval, and k1 neighboring images before the image retrieval; if the image has GPS information, calculating the Euclidean distance between the image and each image in the sub-measurement area by using the GPS information, and reserving the previous k2 images with the nearest distance as neighbor images of the image, wherein the number of the neighbor images of the image is M, and M is k1+ k 2;
step 303: judging whether the image is a boundary image between the sub-measurement areas; the method for determining whether an image is a boundary image between sub-regions applied in this embodiment is as follows: for each image, its M neighboring images are examined, and if at least one neighboring image originates from a different sub-area, this image belongs to a boundary image.
Step 304: performing image feature matching on all boundary images and adjacent images thereof to form an adjacency matrix;
step 305: partitioning all the adjacency relation matrixes obtained in the step 304, independently numbering the partitions, wherein the number of the partitions is P, and performing parallel triangulation processing on the P sub-measurement areas to obtain triangulation results of the P sub-measurement areas; preferably, in this embodiment, the number is increased by 1 in sequence from N +1, that is, the number is N +1, N +2, and N +3.. so on to N + P;
step 306: and combining the triangulation results of the N sub-measurement areas, the P boundary images and the adjacent image matching areas thereof to form an aerial triangulation result of the whole measurement area.
Example 2
Based on the traditional method and the method of example 1, the effect is verified and comparative analysis is carried out, and the test method is as follows:
taking a survey area of 20 square kilometers as an example, plan to take an aerial photograph of 2 square kilometers per day, and process the survey area with two operation schemes.
In the traditional operation scheme, aerial photography is carried out for 10 days, the air-to-air three-field processing is started on the 11 th day, 7 days are needed to finish the air-to-air three-field processing, and 17 days are needed in total; the number of required computer nodes is 11.
Adopting the operation scheme of the embodiment 1 and adopting a relay type aerial triangulation operation scheme, carrying out aerial photography for ten days, starting to process the sub-measurement area images of the same day after the aerial photography is finished every day, and finally completing the combination of the sub-measurement areas within 1 day, wherein the total time is 12 days; the number of required computer nodes is 7.
Compared with the prior operation scheme, 4 computers can be reduced, and the construction period of 5 days can be saved.
Claims (4)
1. An aerial triangulation operation method for aerial images is characterized by comprising the following steps:
step 1: dividing a plurality of sub-measuring areas according to the measuring area range and the terrain condition, wherein the number of the divided sub-measuring areas is N;
step 2: performing image extraction on the sub-measurement areas through aerial photography, and sequentially performing aerial triangulation processing on the extracted sub-measurement area images;
and step 3: and merging the processed data of the plurality of sub-measurement areas to obtain a complete measurement area.
2. The aerial triangulation method of aerial image according to claim 1, wherein the step 3 comprises the following steps:
step 301: arranging the triangulation results of each sub-measurement area, and numbering each camera, each image and each triangulation connection point, wherein the numbering is not repeated;
step 302: obtaining the number M of adjacent images of each image; the method for obtaining the number of the adjacent images of each image comprises the following steps:
using the feature point information of the image to perform image retrieval, and k1 neighboring images before the image retrieval; if the image has GPS information, calculating the Euclidean distance between the image and each image in the sub-measurement area by using the GPS information, and reserving the previous k2 images with the nearest distance as neighbor images of the image, wherein the number of the neighbor images of the image is M, and M is k1+ k 2;
step 303: judging whether the image is a boundary image between the sub-measurement areas;
step 304: performing image feature matching on all boundary images and adjacent images thereof to form an adjacency matrix;
step 305: partitioning all the adjacency relation matrixes obtained in the step 304, independently numbering the partitions, wherein the number of the partitions is P, and performing parallel triangulation processing on the P sub-measurement areas to obtain triangulation results of the P sub-measurement areas;
step 306: and combining the triangulation results of the N sub-measurement areas, the P boundary images and the adjacent image matching areas thereof to form an aerial triangulation result of the whole measurement area.
3. The aerial triangulation method of claim 2, wherein the step 303 of determining whether the image is a boundary image between sub-regions comprises: for each image, its M neighboring images are examined, and if at least one neighboring image originates from a different sub-area, this image belongs to a boundary image.
4. The aerial triangulation method of claim 2, wherein the independent numbering in step 305 starts with N +1 and is incremented by 1 for each next number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110583475.8A CN113405533B (en) | 2021-05-27 | 2021-05-27 | Aerial triangulation operation method for aerial images |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110583475.8A CN113405533B (en) | 2021-05-27 | 2021-05-27 | Aerial triangulation operation method for aerial images |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113405533A true CN113405533A (en) | 2021-09-17 |
| CN113405533B CN113405533B (en) | 2022-04-12 |
Family
ID=77674680
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110583475.8A Active CN113405533B (en) | 2021-05-27 | 2021-05-27 | Aerial triangulation operation method for aerial images |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113405533B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114279412A (en) * | 2021-11-26 | 2022-04-05 | 武汉大势智慧科技有限公司 | Multi-block space-three adjustment merging method based on aerial oblique photography image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170336203A1 (en) * | 2014-10-26 | 2017-11-23 | Galileo Group, Inc. | Methods and systems for remote sensing with drones and mounted sensor devices |
| CN107907111A (en) * | 2017-11-14 | 2018-04-13 | 泰瑞数创科技(北京)有限公司 | A kind of automatic distributed aerial triangulation calculation method |
| CN108332721A (en) * | 2018-03-01 | 2018-07-27 | 北京中测智绘科技有限公司 | The parallel sky three of aviation image and recursion fusion method |
| CN112113544A (en) * | 2019-06-20 | 2020-12-22 | 四川轻化工大学 | Remote sensing data processing method and system based on unmanned aerial vehicle image |
| CN112802177A (en) * | 2020-12-31 | 2021-05-14 | 广州极飞科技股份有限公司 | Processing method and device of aerial survey data, electronic equipment and storage medium |
-
2021
- 2021-05-27 CN CN202110583475.8A patent/CN113405533B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170336203A1 (en) * | 2014-10-26 | 2017-11-23 | Galileo Group, Inc. | Methods and systems for remote sensing with drones and mounted sensor devices |
| CN107907111A (en) * | 2017-11-14 | 2018-04-13 | 泰瑞数创科技(北京)有限公司 | A kind of automatic distributed aerial triangulation calculation method |
| CN108332721A (en) * | 2018-03-01 | 2018-07-27 | 北京中测智绘科技有限公司 | The parallel sky three of aviation image and recursion fusion method |
| CN112113544A (en) * | 2019-06-20 | 2020-12-22 | 四川轻化工大学 | Remote sensing data processing method and system based on unmanned aerial vehicle image |
| CN112802177A (en) * | 2020-12-31 | 2021-05-14 | 广州极飞科技股份有限公司 | Processing method and device of aerial survey data, electronic equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114279412A (en) * | 2021-11-26 | 2022-04-05 | 武汉大势智慧科技有限公司 | Multi-block space-three adjustment merging method based on aerial oblique photography image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113405533B (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pereira et al. | A deep learning-based approach for road pothole detection in timor leste | |
| CN111178206B (en) | Building embedded part detection method and system based on improved YOLO | |
| CN110163213B (en) | Remote sensing image segmentation method based on disparity map and multi-scale depth network model | |
| CN107480727A (en) | The unmanned plane image fast matching method that a kind of SIFT and ORB are combined | |
| CN113408423B (en) | Aquatic product target real-time detection method suitable for TX2 embedded platform | |
| CN113405533B (en) | Aerial triangulation operation method for aerial images | |
| CN110751099A (en) | Unmanned aerial vehicle aerial video track high-precision extraction method based on deep learning | |
| CN110910360B (en) | Positioning method of power grid image and training method of image positioning model | |
| CN110032654B (en) | Supermarket commodity entry method and system based on artificial intelligence | |
| CN106056625A (en) | Airborne infrared moving target detection method based on geographical homologous point registration | |
| CN114972759A (en) | Remote sensing image semantic segmentation method based on hierarchical contour cost function | |
| CN111353396A (en) | Concrete crack segmentation method based on SCSEOCUnet | |
| CN113658117A (en) | Method for identifying and dividing aggregate boundaries in asphalt mixture based on deep learning | |
| CN114267025A (en) | Traffic sign detection method based on high-resolution network and light-weight attention mechanism | |
| CN115310598A (en) | Welding defect real-time detection method and system integrating deep learning and machine learning models | |
| CN111444916A (en) | License plate positioning and identifying method and system under unconstrained condition | |
| CN110728269A (en) | High-speed rail contact net support pole number plate identification method | |
| CN112488043B (en) | Unmanned aerial vehicle target detection method based on edge intelligence | |
| CN113221839A (en) | Automatic truck image identification method and system | |
| CN115761613B (en) | Automatic tunnel crack detection method based on convolutional network | |
| CN116778357A (en) | Power line unmanned aerial vehicle inspection method and system utilizing visible light defect identification | |
| CN115830592A (en) | Overlapping cervical cell segmentation method and system | |
| CN116343095A (en) | Vehicle track extraction method based on video stitching and related equipment | |
| CN114120154B (en) | Automatic detection method for breakage of glass curtain wall of high-rise building | |
| CN114627027A (en) | Semi-automatic machine learning denoising method for moving laser scanning point cloud of mountain tunnel |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhang Yong Inventor after: Nie Dan Inventor before: Zhang Yong Inventor before: Nie Dan Inventor before: Li Xuechuan Inventor before: Huang Haipeng Inventor before: Zhang Pei |
|
| CB03 | Change of inventor or designer information |