CN114842074A - Unmanned aerial vehicle image positioning method based on model matching - Google Patents
Unmanned aerial vehicle image positioning method based on model matching Download PDFInfo
- Publication number
- CN114842074A CN114842074A CN202210646966.7A CN202210646966A CN114842074A CN 114842074 A CN114842074 A CN 114842074A CN 202210646966 A CN202210646966 A CN 202210646966A CN 114842074 A CN114842074 A CN 114842074A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- aerial vehicle
- unmanned aerial
- dimensional model
- cloud data
- 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 31
- 238000009434 installation Methods 0.000 claims abstract description 7
- 230000009466 transformation Effects 0.000 claims description 48
- 239000011159 matrix material Substances 0.000 claims description 37
- 238000005457 optimization Methods 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 239000000126 substance Substances 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 3
- 230000004807 localization Effects 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 10
- 230000009191 jumping Effects 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 2
- 238000012550 audit Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012360 testing method 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
- G06T7/70—Determining position or orientation of objects or cameras
-
- 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
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/08—Projecting images onto non-planar surfaces, e.g. geodetic screens
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- 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
- G06T2207/10044—Radar image
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Remote Sensing (AREA)
- Computer Graphics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention relates to an unmanned aerial vehicle image positioning method based on model matching, which comprises the following steps: loading a three-dimensional model high-precision map of an unmanned aerial vehicle positioning space, if the loading is successful, acquiring laser radar point cloud data in the high-precision map at the moment k, matching the laser radar point cloud data at the moment k with the high-precision map, and resolving the pose of the unmanned aerial vehicle; if the loading fails, matching the laser radar point cloud data in the high-precision map at the time k with the sub-map of the space three-dimensional model established at the time k-1, and resolving the pose of the unmanned aerial vehicle; based on the installation relation of the laser radar and the vision sensor, the holder angle and the unmanned aerial vehicle pose, the position of the shot image in the three-dimensional model is calculated, and the shot image and the position information of the shot image in the three-dimensional model are output. By the method and the device, unmanned detection can be realized in a satellite rejection environment, images in the three-dimensional model and the positions of the images in the three-dimensional model are obtained, and the detection efficiency and accuracy can be improved.
Description
Technical Field
The invention relates to the technical field of intelligent detection of unmanned aerial vehicle systems, in particular to an unmanned aerial vehicle image positioning method based on model matching.
Background
The enclosed space detection means that a detection mechanism carries out inspection, audit, test and the like on internal structures and equipment of the space. The necessary technical certificate can only be obtained by a corresponding check.
The traditional manual detection method has the defects of low efficiency, high cost and the like, and the unmanned system can perform autonomous positioning in a satellite signal rejection environment based on the laser radar and associate a shot image with a shooting position, so that the detection efficiency and accuracy are effectively improved.
Disclosure of Invention
The invention aims to provide an unmanned aerial vehicle image positioning method based on model matching, which can realize high-precision autonomous positioning based on a laser radar under the condition of existence of a three-dimensional model high-precision map, and can calculate the position of a shot in the three-dimensional model through the installation relation between the laser radar and a vision sensor and the angle of a holder.
In order to achieve the purpose, the invention provides the following scheme:
an unmanned aerial vehicle image positioning method based on model matching comprises the following steps:
loading a three-dimensional model high-precision map of an unmanned aerial vehicle positioning space, if the three-dimensional model high-precision map is successfully loaded, acquiring laser radar point cloud data in the high-precision map at the moment k, matching the laser radar point cloud data at the moment k with the high-precision map, and resolving the pose of the unmanned aerial vehicle;
if the loading fails, matching the laser radar point cloud data in the high-precision map at the time k with a sub-map of a space three-dimensional model established at the time k-1, and resolving the pose of the unmanned aerial vehicle;
based on the installation relation of the laser radar and the vision sensor, the holder angle and the unmanned aerial vehicle pose, calculating the position of the shot image in the three-dimensional model, and outputting the shot image and the position information of the shot image in the three-dimensional model.
Preferably, the high-precision map of the three-dimensional model comprises point cloud data and a fast point feature value histogram (FPFH) of the point cloud.
Preferably, matching the laser radar point cloud data at the moment k with the high-precision map, and resolving the pose of the unmanned aerial vehicle, wherein the method comprises the following steps:
calculating a fast point characteristic value histogram (FPFH) of each point cloud in the laser radar point cloud data at the moment k; searching corresponding points of a plurality of sampling points based on the FPFH, and calculating a transformation matrix; and performing projection transformation on the sampling points, calculating distance errors between the sampling points and the corresponding points after projection transformation through the transformation matrix, and calculating the pose of the unmanned aerial vehicle according to the distance errors and the transformation matrix.
Preferably, the method for calculating the fast point feature value histogram FPFH is as follows:
wherein p is a point to be calculated, p ki Neighborhood points for point p, ki the number of neighborhood points, SPFH is a point reduced feature histogram, ω ki Denotes p and p ki The distance between them.
Preferably, calculating the distance error between the sampling point and the corresponding point after projective transformation includes:
wherein m is l To a preset value,/ i The distance difference between the ith sample point and the corresponding point after projection transformation is obtained.
Preferably, when the distance error and the score are smaller than a set threshold or reach a set calculation number, selecting the transformation matrix when the score is the minimum value as a point cloud registration result; wherein the calculation method of the score is as follows:
l i the distance difference between the ith sample point and the corresponding point after projection transformation is obtained.
Preferably, based on the transformation matrix when score is the minimum value, projection transformation is performed on the laser radar point cloud data at the time k, the transformed point cloud is put into a space grid with the nearest distance, and the transformation matrix is optimized; the constructed optimization function is as follows:
T * in order to be the optimal transformation matrix,
is a probability density function, T is a pose transformation matrix, x k And (4) for inputting the point cloud, iterating the formula by a Newton optimization method to obtain an optimal solution.
Preferably, obtaining the spatial grid comprises:
carrying out gridding operation on the space occupied by the point cloud data in the high-precision map, and calculating the probability density function of the point cloud data in each grid:
wherein the content of the first and second substances,
as a function of probability density, x k For inputting a point cloud, d 1 、d 2 Normalizing constants for probability density functions, (. The)' denotes transpose operation, mu, sigma k The calculation is made by the following formula:
where m is the number of point clouds in the grid, y j Is the jth point cloud in the grid.
Preferably, matching the laser radar point cloud data in the high-precision map at the time k with the three-dimensional model sub-map constructed at the time k-1 includes:
carrying out gridding operation on the space occupied by the point cloud data at the moment k-1, and calculating the probability density function of the point cloud data in each grid:
wherein the content of the first and second substances,
is a probability density function;
performing projection transformation on the laser radar point cloud data in the high-precision map at the time k, putting the transformed point cloud into a space grid with the nearest distance, and optimizing a transformation matrix, wherein the constructed optimization function is as follows:
T * for the optimal transformation matrix, iteration is carried out through a Newton optimization method to obtain an optimal solution; based on the optimal transformation matrix T * And updating the three-dimensional model sub-map by the laser radar point cloud data at the moment k to obtain an updated three-dimensional model sub-map MS (k).
Preferably, calculating the position of the shot image in the three-dimensional model based on the pose of the unmanned aerial vehicle comprises:
based on the installation relation of the laser radar and the vision sensor, the holder angle and the unmanned aerial vehicle pose, calculating the position of the shot image in the three-dimensional model:
wherein
Is a rotation matrix and a translation vector between the laser radar and the vision sensor,
rotation matrix for the change in pan-tilt angle at time k, P c The coordinates of the captured image in the coordinate system of the vision sensor,
is a rotation matrix of the unmanned aerial vehicle under a navigation coordinate system,
for the translation vector of the unmanned plane in the navigation coordinate system, P n The position of the shot image in the three-dimensional model.
The invention has the beneficial effects that:
by the method, the problems of low unmanned detection accuracy caused by poor positioning accuracy of the unmanned aerial vehicle in a satellite signal rejection environment and low manual detection efficiency and high cost can be solved, images in the three-dimensional model and corresponding positions of the images in the three-dimensional model can be obtained by the method, and the detection efficiency and accuracy can be improved effectively.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of a method according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
The flow of the method of the invention is shown in figure 1, and the specific steps are as follows:
the method comprises the following steps: the high-precision map M loaded with the three-dimensional model includes both Point cloud data and a Fast Point Feature Histogram (FPFH) of the Point cloud. And jumping to the step two if the loading is successful, and jumping to the step three if the loading is failed.
Step two: and collecting laser radar point cloud data S (k) at the moment k, matching the point cloud data S (k) with the point cloud data M, and resolving the unmanned aerial vehicle pose at the moment k.
1) Calculating the FPFH of each point cloud in S (k):
wherein p is a point to be calculated, p ki Neighborhood points for point p, ki the number of neighborhood points, SPFH is a point reduced feature histogram, ω ki Denotes p and p ki The distance between them. In the step S (k), n sampling points are randomly selected, the distance between every two sampling points is larger than a set threshold value d, and the selected sampling points are ensured to have different FPFH (field programmable gate hydrofs) as much as possible.
2) Finding out the corresponding points of n sampling points in the point cloud data of M according to FPFH, and calculating the corresponding transformation matrix under the matching condition according to n pairs of sampling-corresponding points
N samples are collectedSampling point according to
Performing projection transformation, calculating the distance error between the point after projection transformation and the corresponding point, and judging the point cloud registration performance according to the distance error:
wherein, H (l) i ) Is specifically represented as follows:
wherein m is l To a preset value,/ i The distance difference between the ith sample point and the corresponding point after projection transformation is obtained.
3) Repeating 2) until the distance error and score are less than the set threshold or the set calculation times are reached, and selecting score to obtain the minimum value
As a result of the point cloud registration.
4) Gridding the space occupied by the point cloud data in the M, and calculating the probability density function of the point cloud data in each grid:
wherein the content of the first and second substances,
as a function of probability density, x k For inputting a point cloud, d 1 、d 2 Normalizing constants for probability density functions, mu, sigma k The calculation is made by the following formula:
where m is the number of point clouds in the grid, y j Is the jth point cloud in the grid.
5) According to calculation in 3)
And (k) performing projection transformation on the S (k), putting the transformed point cloud into a space grid with the closest distance, and optimizing a transformation matrix to maximize the possibility that the input point cloud is positioned on the target point cloud. The constructed optimization function is as follows:
T * the optimal transformation matrix is obtained, and the above formula is iterated by a Newton optimization method to obtain the optimal solution.
And 6, jumping to the step four.
Step three: and collecting laser radar point cloud data S (k) at the moment k, matching the S (k) with the three-dimensional model sub-map MS (k-1) constructed at the moment k-1, calculating the pose of the unmanned aerial vehicle at the moment k, and updating the three-dimensional model sub-map.
Gridding the space occupied by point cloud data in MS (k-1), and calculating the probability density function of the point cloud data in each grid:
wherein the content of the first and second substances,
is a probability density function.
And (k) performing projection transformation on the S (k), putting the transformed point cloud into a space grid with the closest distance, and optimizing a transformation matrix to maximize the possibility that the input point cloud is positioned on the target point cloud. The constructed optimization function is as follows:
T * the optimal transformation matrix is obtained, and the above formula is iterated by a Newton optimization method to obtain the optimal solution. According to the optimal transformation matrix T * And S (k) updating the three-dimensional model sub-map to obtain MS (k).
And 6, jumping to the step four.
Step four: the method comprises the steps that a visual sensor carried by the unmanned aerial vehicle is used for shooting a three-dimensional model image, and the position of the image in the three-dimensional model is calculated based on the installation relation of a laser radar and the visual sensor, the angle of a holder and the pose of the unmanned aerial vehicle at the moment k.
K time transformation matrix T of unmanned aerial vehicle * Including rotation matrix of unmanned aerial vehicle under navigation coordinate system
And translation vector
And (3) setting a laser radar coordinate system to be consistent with an unmanned aerial vehicle body coordinate system, and calculating the position of the image in the three-dimensional model according to the following formula:
wherein
Is a rotation matrix and a translation vector between the laser radar and the vision sensor,
rotation matrix for the change in pan-tilt angle at time k, P c Is the coordinate of the photographed image in the vision sensor coordinate system, P n The position of the shot image in the three-dimensional model is obtained.
Step five: and outputting the shot image and the position information of the shot image in the three-dimensional model, and jumping to the second step under the condition that the M is successfully loaded, or jumping to the third step.
The above-described embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements of the technical solutions of the present invention can be made by those skilled in the art without departing from the spirit of the present invention, and the technical solutions of the present invention are within the scope of the present invention defined by the claims.
Claims (10)
1. The utility model provides an unmanned aerial vehicle image positioning method based on model matching which characterized in that includes:
loading a three-dimensional model high-precision map of an unmanned aerial vehicle positioning space, if the loading is successful, acquiring laser radar point cloud data in the high-precision map at the time k, matching the laser radar point cloud data at the time k with the high-precision map, and resolving the pose of the unmanned aerial vehicle;
if the loading fails, matching the laser radar point cloud data in the high-precision map at the time k with a sub-map of a space three-dimensional model established at the time k-1, and resolving the pose of the unmanned aerial vehicle;
based on the installation relation of the laser radar and the vision sensor, the holder angle and the unmanned aerial vehicle pose, calculating the position of the shot image in the three-dimensional model, and outputting the shot image and the position information of the shot image in the three-dimensional model.
2. The model matching-based unmanned aerial vehicle image localization method of claim 1, wherein the high-precision map of the three-dimensional model comprises point cloud data and a fast point feature value histogram (FPFH) of the point cloud.
3. The model matching-based unmanned aerial vehicle image positioning method of claim 1, wherein the matching of the lidar point cloud data at the time k with the high-precision map and the resolving of the unmanned aerial vehicle pose comprise:
calculating a fast point characteristic value histogram (FPFH) of each point cloud in the laser radar point cloud data at the moment k; searching corresponding points of a plurality of sampling points based on the FPFH, and calculating a transformation matrix; and performing projection transformation on the sampling points, calculating distance errors between the sampling points and the corresponding points after projection transformation through the transformation matrix, and calculating the pose of the unmanned aerial vehicle according to the distance errors and the transformation matrix.
4. The model matching-based unmanned aerial vehicle image positioning method of claim 3, wherein the fast point feature value histogram (FPFH) is calculated by:
wherein p is a point to be calculated, p ki Neighborhood points for point p, ki the number of neighborhood points, SPFH is a point reduced feature histogram, ω ki Denotes p and p ki The distance between them.
5. The unmanned aerial vehicle image positioning method based on model matching as claimed in claim 3, wherein calculating the distance error between the sampling point and the corresponding point after the projective transformation comprises:
wherein m is l To a preset value,/ i The distance difference between the ith sample point and the corresponding point after projection transformation is obtained.
6. The model matching-based unmanned aerial vehicle image positioning method according to claim 5, wherein when the distance error and score are smaller than a set threshold or reach a set number of calculations, the transformation matrix with the score at the minimum value is selected as a point cloud registration result; wherein the calculation method of the score is as follows:
l i the distance difference between the ith sample point and the corresponding point after projection transformation is obtained.
7. The unmanned aerial vehicle image positioning method based on model matching as claimed in claim 6, wherein based on the transformation matrix when score is the minimum value, projection transformation is performed on the lidar point cloud data at the k time, the transformed point cloud is put into the space grid closest to the k time, and the transformation matrix is optimized; the constructed optimization function is as follows:
T * in order to be the optimal transformation matrix,
is a probability density function, T is a pose transformation matrix, x k And (4) for inputting the point cloud, iterating the formula by a Newton optimization method to obtain an optimal solution.
8. The model matching-based unmanned aerial vehicle image positioning method of claim 7, wherein obtaining the spatial grid comprises:
carrying out gridding operation on the space occupied by the point cloud data in the high-precision map, and calculating the probability density function of the point cloud data in each grid:
wherein the content of the first and second substances,
as a function of probability density, x k For the input point cloud, d 1 、d 2 Normalizing constants for probability density functions, (. The)' denotes transpose operation, mu, sigma k The calculation is made by the following formula:
where m is the number of point clouds in the grid, y j Is the jth point cloud in the grid.
9. The unmanned aerial vehicle image positioning method based on model matching as claimed in claim 1, wherein matching with the laser radar point cloud data in the high-precision map at time k and the three-dimensional model sub-map constructed at time k-1 comprises:
carrying out gridding operation on the space occupied by the point cloud data at the moment k-1, and calculating the probability density function of the point cloud data in each grid:
wherein the content of the first and second substances,
is a probability density function;
performing projection transformation on the laser radar point cloud data in the high-precision map at the time k, putting the transformed point cloud into a space grid with the nearest distance, and optimizing a transformation matrix, wherein the constructed optimization function is as follows:
T * for the optimal transformation matrix, iteration is carried out through a Newton optimization method to obtain an optimal solution; based on the optimal transformation matrixT * And updating the three-dimensional model sub-map by the laser radar point cloud data at the moment k to obtain an updated three-dimensional model sub-map MS (k).
10. The method of claim 1, wherein calculating the position of the captured image in the three-dimensional model based on the pose of the drone comprises:
based on the installation relation of the laser radar and the vision sensor, the holder angle and the unmanned aerial vehicle pose, calculating the position of the shot image in the three-dimensional model:
wherein
Is a rotation matrix and a translation vector between the laser radar and the vision sensor,
rotation matrix for the change in pan-tilt angle at time k, P c The coordinates of the captured image in the coordinate system of the vision sensor,
is a rotation matrix of the unmanned aerial vehicle under a navigation coordinate system,
for the translation vector of the unmanned plane in the navigation coordinate system, P n The position of the shot image in the three-dimensional model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210646966.7A CN114842074A (en) | 2022-06-09 | 2022-06-09 | Unmanned aerial vehicle image positioning method based on model matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210646966.7A CN114842074A (en) | 2022-06-09 | 2022-06-09 | Unmanned aerial vehicle image positioning method based on model matching |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114842074A true CN114842074A (en) | 2022-08-02 |
Family
ID=82573696
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210646966.7A Pending CN114842074A (en) | 2022-06-09 | 2022-06-09 | Unmanned aerial vehicle image positioning method based on model matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114842074A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116203554A (en) * | 2023-05-06 | 2023-06-02 | 武汉煜炜光学科技有限公司 | Environment point cloud data scanning method and system |
-
2022
- 2022-06-09 CN CN202210646966.7A patent/CN114842074A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116203554A (en) * | 2023-05-06 | 2023-06-02 | 武汉煜炜光学科技有限公司 | Environment point cloud data scanning method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108871353B (en) | Road network map generation method, system, equipment and storage medium | |
| CN109270545B (en) | Positioning true value verification method, device, equipment and storage medium | |
| CN110889324A (en) | Thermal infrared image target identification method based on YOLO V3 terminal-oriented guidance | |
| CN110930495A (en) | Multi-unmanned aerial vehicle cooperation-based ICP point cloud map fusion method, system, device and storage medium | |
| CN112346104A (en) | Unmanned aerial vehicle information fusion positioning method | |
| CN112052807B (en) | Vehicle position detection method, device, electronic equipment and storage medium | |
| CN114820749A (en) | Unmanned vehicle underground positioning method, system, equipment and medium | |
| CN116206223A (en) | Fire detection method and system based on unmanned aerial vehicle edge calculation | |
| CN114842074A (en) | Unmanned aerial vehicle image positioning method based on model matching | |
| CN115343744A (en) | Optical single-double-star combined on-satellite positioning method and system for aerial moving target | |
| CN113012215A (en) | Method, system and equipment for space positioning | |
| CN112148817B (en) | SLAM optimization method, device and system based on panorama | |
| CN111735447A (en) | Satellite-sensitive-simulation type indoor relative pose measurement system and working method thereof | |
| CN113673288A (en) | Idle parking space detection method and device, computer equipment and storage medium | |
| CN117029817A (en) | Two-dimensional grid map fusion method and system | |
| CN111104965A (en) | Vehicle target identification method and device | |
| CN110927765A (en) | Laser radar and satellite navigation fused target online positioning method | |
| CN113850173A (en) | Method, device and medium for eliminating repeated positioning of faulty photovoltaic module | |
| CN111024063A (en) | Star map recognition algorithm under large-mobility condition based on star point re-extraction | |
| Tang et al. | Automatic geo‐localization framework without GNSS data | |
| CN114390270B (en) | Real-time intelligent site panorama exploration method and device and electronic equipment | |
| CN117115414B (en) | GPS-free unmanned aerial vehicle positioning method and device based on deep learning | |
| CN116704037B (en) | Satellite lock-losing repositioning method and system based on image processing technology | |
| CN116380023B (en) | Land mapping system based on remote sensing technology | |
| CN113920180B (en) | Point cloud registration optimization method based on normal distribution transformation hypothesis verification |
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 |