CN112581368A - Multi-robot grid map splicing method based on optimal map matching - Google Patents
Multi-robot grid map splicing method based on optimal map matching Download PDFInfo
- Publication number
- CN112581368A CN112581368A CN202011445782.1A CN202011445782A CN112581368A CN 112581368 A CN112581368 A CN 112581368A CN 202011445782 A CN202011445782 A CN 202011445782A CN 112581368 A CN112581368 A CN 112581368A
- Authority
- CN
- China
- Prior art keywords
- optimal
- grid map
- matching
- matrix
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 239000011159 matrix material Substances 0.000 claims abstract description 66
- 230000005540 biological transmission Effects 0.000 claims abstract description 45
- 230000004927 fusion Effects 0.000 claims abstract description 13
- 230000008859 change Effects 0.000 claims abstract description 7
- 230000001131 transforming effect Effects 0.000 claims abstract description 4
- 238000013519 translation Methods 0.000 claims description 12
- 230000003416 augmentation Effects 0.000 claims description 8
- 230000003190 augmentative effect Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 239000000523 sample Substances 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 238000010276 construction Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- HCHKCACWOHOZIP-UHFFFAOYSA-N Zinc Chemical compound [Zn] HCHKCACWOHOZIP-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000011701 zinc Substances 0.000 description 1
- 229910052725 zinc Inorganic materials 0.000 description 1
Images
Classifications
-
- 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
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention provides a multi-robot grid map splicing method based on optimal graph matching, which comprises the following steps: extracting key points and feature descriptors of the grid map to be spliced; searching the key points with the most similar characteristic descriptors for each key point to be matched as a rough matching point pair; constructing a characteristic dissimilarity matrix and calculating a residual error matrix; combining and normalizing the residual error matrix and the characteristic dissimilarity matrix to generate a transmission cost matrix; introducing a transmission cost matrix to construct an optimal transmission objective function, carrying out negative entropy regularization on the optimal transmission objective function, and solving optimal matching through a Sinkhorn-Knopp algorithm; and solving rigid body change parameters between the optimal matching points by a least square method, and transforming the whole grid map to obtain a fusion map. The multi-robot grid map splicing method effectively improves the splicing speed and precision of the grid map, the number of correct matching point pairs in the result is large, and the grid map to be spliced can be spliced without a large overlapping area.
Description
Technical Field
The invention relates to the technical field of multi-robot map construction, in particular to a multi-robot grid map splicing method based on optimal graph matching.
Background
The map construction for the unknown environment is a basic challenge in the mobile robot technology, and the map construction generally needs to accurately estimate the pose of the robot, so that the mobile robot mainly constructs the environment map by the Simultaneous Localization and Mapping (SLAM) technology at present, various mature single-robot SLAM algorithms exist at present, the map construction accuracy, efficiency and robustness of a single robot in the large-scale unknown environment are limited, the cooperation of multiple robots becomes a hotspot of the current research in the past decade, and the introduction of multiple robots is helpful for breaking through the limitation of the single-robot SLAM algorithm. In the robot SLAM algorithm, the map types can be divided into a feature map, a topological map and a grid map, wherein the grid map can visually express the structure of a real environment, is convenient to store and has more advantages in the aspects of accurate positioning and navigation.
In the multi-robot SLAM, how to splice local maps and determine the relative pose between robots after each robot constructs a respective local map is a problem of multi-robot mapping technology. The existing grid map generation operation needs manual participation in map splicing, and the splicing method is low in efficiency and is not suitable for multi-robot map real-time construction scenes. Grid map stitching differs from ordinary image stitching in that non-rigid deformation of the grid map occurs due to the existence of accumulated errors and observation noise.
Therefore, the existing grid map splicing method [1] congratulatory, Zhouyi, Wangchun, Han Wen zinc, Marangong based on the image registration [ J ] automatic chemical report, 2015,41(02):285-294 ], which adopts a SIFT feature and ICP registration method to realize grid map splicing, but on one hand, the SIFT feature has large calculation amount and is difficult to meet the real-time calculation requirement of multi-robot cooperative task, on the other hand, the ICP registration method has local optimal solution, and if the initial value is improperly selected, the image registration process is easy to fall into the local optimal solution, which causes splicing failure. Grid map splicing method and device based on Voronoi diagram and readable storage medium [2] Cheng. CN110910313A,2020-03-24, the method is based on establishing a voronoi diagram for a grid map, and realizing grid map splicing by using a constraint relation between the voronoi diagram and the grid map, and the method requires that the overlapping area of the voronoi diagram and the voronoi diagram is larger, and for a map with a smaller overlapping area, the constraint between the voronoi diagram and the grid map is weaker.
Disclosure of Invention
The invention aims to overcome the technical defects that the conventional grid map splicing method has large calculation amount and high calculation cost, and large-area overlapping of images is required for splicing, and provides a multi-robot grid map splicing method based on optimal map matching.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a multi-robot grid map splicing method based on optimal map matching comprises the following steps:
s1: carrying out ORB feature point extraction on the grid map to be spliced to obtain key points and feature descriptors;
s2: searching the most similar key points of the feature descriptors for each key point to be matched through a multi-probe locality sensitive hashing algorithm to serve as rough matching point pairs;
s3: combining Hamming distances among the feature descriptors to construct a feature dissimilarity matrix; respectively constructing a median K neighbor graph for the key points, and calculating a residual matrix; combining and normalizing the residual error matrix and the characteristic dissimilarity matrix to generate a transmission cost matrix;
s4: introducing a transmission cost matrix to construct an optimal transmission target function, constructing an augmentation node to remove an outlier, carrying out negative entropy regularization on the optimal transmission target function, and solving optimal matching through a Sinkhorn-Knopp algorithm;
s5: and solving rigid body change parameters between the optimal matching points by a least square method, and transforming the whole grid map to obtain a fusion map.
In the scheme, the ORB feature points of the grid map are extracted aiming at the problems of non-rigid deformation and observation noise interference of the grid map, the ORB feature point matching problem is converted into the image matching problem by constructing a median K neighbor map between key points, and simultaneously, the augmentation nodes are introduced to remove outliers, construct an optimal transmission objective function, further solve the optimal matching between the key points and fuse the grid map.
Wherein the step S1 includes the steps of:
s11: carrying out Gaussian blur filtering on the edge noise of the grid map to be spliced and enabling the edge of the binary image to generate a continuous and smooth gradient;
s12: extracting multiscale FAST key points and extracting BRIEF feature descriptors.
In the scheme, the ORB feature points are used as key points to be matched, so that the speed and the precision of grid map splicing are improved.
Wherein, the step S3 specifically includes the following steps:
s31: constructing a feature dissimilarity matrix D according to the Hamming distance between every two feature points extracted from two grid maps to be spliced, and respectively calculating the median of Euclidean distances between the concentration points and the points of each key point;
s32: for each key point, K neighbor key points closest to the key point are found, and the distances between all the neighbor key points and the central key point are ensured to be smaller than a median value;
s33: constructing a non-directional edge for all the neighbor key points and the central key point which meet the constraint of the step S32, thereby constructing a median K neighbor graph;
s34: calculating the difference of adjacent matrixes of the median K neighbor graph as a residual matrix R;
s35: and summing the characteristic dissimilarity matrix D and the residual error matrix R, and generating a transmission cost matrix through index normalization processing.
In the scheme, the characteristic point matching problem is converted into the graph matching problem by constructing the local geometric relationship of the median K neighbor aggregation key points of the OBR key points, so that the splicing robustness and the non-rigid deformation interference resistance of the grid map are effectively improved.
In step S35, the transmission cost matrix is specifically:
wherein n ispAnd nqRepresenting the number of key points; c is the transmission cost matrix.
In step S4, the optimal transfer objective function is specifically expressed as:
wherein P and q are key point sets, P is a soft distribution matrix, mu and upsilon are transmission quality of the key points, and a is an entropy regular term coefficient and is set to be 1; solving the optimal transmission objective function by adopting a Sinkhorn-Knopp algorithm to obtain an optimal matching matrix P*。
In step S4, two augmented nodes g are introduced because the coarse matching result has an outlier that cannot be matchedaAnd gbAnd introducing an optimal transmission constraint, specifically:
where α is the fixed augmented node mass.
In step S5, the solving of the rigid body variation parameter between the optimal matching points by the least square method specifically includes:
after the optimal matching key points are obtained, solving the following formula by adopting a least square method:
wherein p ismAnd q isc(m)The optimal matching key point is obtained; n is a radical ofmThe number of the optimal matching key points; r and t areA rotation matrix and a translation vector; and then creating a blank image with a proper size, directly copying the grid map A to be spliced to a new image, then carrying out image transformation on the whole grid map B according to the solved rigid body change parameters, and splicing the map according to a fusion function.
In step S5, the rotation matrix R is specifically represented as:
where θ represents the angle of rotation.
In step S5, the translation vector t is specifically represented as:
wherein x represents the abscissa of the translation; y denotes the ordinate of the translation.
In step S5, the fusion function is specifically expressed as:
wherein h (-) represents a fusion function, x and y are the transformed two grid map pixel values, and a is a non-occupied pixel value.
In the scheme, the structural difference between residual matrix measurement graphs of the median K neighbor graph adjacency matrix and the characteristic dissimilarity matrix combination formed by the Hamming distance of the ORB characteristic descriptor are calculated and normalized, the transmission cost matrix is constructed, the augmented node is introduced to remove the outlier, the graph matching problem is converted into the optimal transmission problem, the optimal transmission objective function is constructed, the optimization of the ORB characteristic matching relation and the removal of the outlier are simultaneously realized, after the negative entropy regularization, the optimal transmission is solved through the Sinkhorn-Knopp algorithm, the number of correct matching point pairs in the result is large, the grid map to be spliced is allowed to be spliced without a large overlapping area, and the splicing robustness and the application range of the grid map splicing are improved.
Compared with the prior art, the technical scheme of the invention has the beneficial effects that:
the multi-robot grid map splicing method based on optimal map matching effectively improves the splicing speed and precision of the grid map, is small in calculated amount and high in splicing cost, has a large number of correct matching point pairs in the result, and can complete splicing of the grid map to be spliced without a large overlapping area.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
fig. 2 is a schematic diagram of a transmission cost matrix construction process;
FIG. 3 is a diagram illustrating outlier removal by an augmented node;
fig. 4 is a flowchart of grid map transformation parameter solving and stitching.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
for the purpose of better illustrating the embodiments, certain features of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product;
it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
Example 1
As shown in fig. 1, a multi-robot grid map stitching method based on optimal map matching includes the following steps:
s1: carrying out ORB feature point extraction on the grid map to be spliced to obtain key points and feature descriptors;
s2: searching the most similar key points of the feature descriptors for each key point to be matched through a multi-probe locality sensitive hashing algorithm to serve as rough matching point pairs;
s3: combining Hamming distances among the feature descriptors to construct a feature dissimilarity matrix; respectively constructing a median K neighbor graph for the key points, and calculating a residual matrix; combining and normalizing the residual error matrix and the characteristic dissimilarity matrix to generate a transmission cost matrix;
s4: introducing a transmission cost matrix to construct an optimal transmission target function, constructing an augmentation node to remove an outlier, carrying out negative entropy regularization on the optimal transmission target function, and solving optimal matching through a Sinkhorn-Knopp algorithm;
s5: and solving rigid body change parameters between the optimal matching points by a least square method, and transforming the whole grid map to obtain a fusion map.
In the specific implementation process, aiming at the problems of non-rigid deformation and observation noise interference existing in the grid map, ORB feature points of the grid map are extracted, the ORB feature point matching problem is converted into an image matching problem by constructing a median K neighbor map between key points, and simultaneously, an augmentation node is introduced to remove an outlier, construct an optimal transmission objective function, further solve the optimal matching between the key points, and fuse the grid map.
Example 2
More specifically, each step is further illustrated on the basis of example 1.
More specifically, the step S1 includes the following steps:
s11: carrying out Gaussian blur filtering on the edge noise of the grid map to be spliced and enabling the edge of the binary image to generate a continuous and smooth gradient;
s12: extracting multiscale FAST key points and extracting BRIEF feature descriptors.
In the specific implementation process, the ORB feature points are used as key points to be matched, so that the speed and the precision of grid map splicing are improved.
More specifically, as shown in fig. 2, the step S3 specifically includes the following steps:
s31: constructing a feature dissimilarity matrix D according to the Hamming distance between every two feature points extracted from two grid maps to be spliced, and respectively calculating the median of Euclidean distances between the concentration points and the points of each key point;
s32: for each key point, K neighbor key points closest to the key point are found, and the distances between all the neighbor key points and the central key point are ensured to be smaller than a median value;
s33: constructing a non-directional edge for all the neighbor key points and the central key point which meet the constraint of the step S32, thereby constructing a median K neighbor graph;
s34: calculating the difference of adjacent matrixes of the median K neighbor graph as a residual matrix R;
s35: and summing the characteristic dissimilarity matrix D and the residual error matrix R, and generating a transmission cost matrix through index normalization processing.
In the specific implementation process, the characteristic point matching problem is converted into the graph matching problem by constructing the local geometric relationship of the median K neighbor aggregation key points of the OBR key points, so that the splicing robustness and the non-rigid deformation interference resistance of the grid map are effectively improved.
More specifically, in step S35, the transmission cost matrix specifically includes:
wherein n ispAnd nqRepresenting the number of key points; c is the transmission cost matrix.
More specifically, in step S4, the optimal transfer objective function is specifically expressed as:
wherein P and q are key point sets, P is a soft distribution matrix, mu and upsilon are transmission quality of the key points, and a is an entropy regular term coefficient and is set to be 1; solving the optimal transmission objective function by adopting a Sinkhorn-Knopp algorithm to obtain an optimal matching matrix P*。
More specifically, as shown in FIG. 3, in the step S4, the method comprisesIntroducing two augmentation nodes g when the rough matching result has unmatched outliersaAnd gbAnd introducing an optimal transmission constraint, specifically:
where α is the fixed augmented node mass.
More specifically, as shown in fig. 4, in step S5, the solving of the rigid body variation parameter between the optimal matching points by the least square method specifically includes:
after the optimal matching key points are obtained, solving the following formula by adopting a least square method:
wherein p ismAnd q isc(m)The optimal matching key point is obtained; n is a radical ofmThe number of the optimal matching key points; r and t are a rotation matrix and a translation vector; and then creating a blank image with a proper size, directly copying the grid map A to be spliced to a new image, then carrying out image transformation on the whole grid map B according to the solved rigid body change parameters, and splicing the map according to a fusion function.
More specifically, in step S5, the rotation matrix R is specifically represented as:
where θ represents the angle of rotation.
More specifically, in step S5, the translation vector t is specifically represented as:
wherein x represents the abscissa of the translation; y denotes the ordinate of the translation.
More specifically, in step S5, the fusion function is specifically expressed as:
wherein h (-) represents a fusion function, x and y are the transformed two grid map pixel values, and a is a non-occupied pixel value.
In the specific implementation process, the structural difference between residual matrix measurement graphs of a median K neighbor graph adjacency matrix and a characteristic dissimilarity matrix combination formed by the Hamming distance of an ORB characteristic descriptor are calculated and normalized, a transmission cost matrix is constructed, an augmentation node is introduced to remove an outlier, the graph matching problem is converted into an optimal transmission problem, an optimal transmission objective function is constructed, optimization of an ORB characteristic matching relation and removal of the outlier are achieved, after negative entropy regularization, the optimal transmission is solved through a Sinkhorn-Knopp algorithm, the number of correct matching point pairs in the result is large, the grid map to be spliced is allowed to be spliced without a large overlapping area, and the robustness and the application range of grid map splicing are improved.
It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. A multi-robot grid map splicing method based on optimal map matching is characterized by comprising the following steps:
s1: carrying out ORB feature point extraction on the grid map to be spliced to obtain key points and feature descriptors;
s2: searching the most similar key points of the feature descriptors for each key point to be matched through a multi-probe locality sensitive hashing algorithm to serve as rough matching point pairs;
s3: combining Hamming distances among the feature descriptors to construct a feature dissimilarity matrix; respectively constructing a median K neighbor graph for the key points, and calculating a residual matrix; combining and normalizing the residual error matrix and the characteristic dissimilarity matrix to generate a transmission cost matrix;
s4: introducing a transmission cost matrix to construct an optimal transmission target function, constructing an augmentation node to remove an outlier, carrying out negative entropy regularization on the optimal transmission target function, and solving optimal matching through a Sinkhorn-Knopp algorithm;
s5: and solving rigid body change parameters between the optimal matching points by a least square method, and transforming the whole grid map to obtain a fusion map.
2. The multi-robot grid map stitching method based on optimal map matching as claimed in claim 1, wherein the step S1 comprises the steps of:
s11: carrying out Gaussian blur filtering on the edge noise of the grid map to be spliced and enabling the edge of the binary image to generate a continuous and smooth gradient;
s12: extracting multiscale FAST key points and extracting BRIEF feature descriptors.
3. The multi-robot grid map stitching method based on optimal map matching as claimed in claim 1, wherein the step S3 specifically comprises the steps of:
s31: constructing a feature dissimilarity matrix D according to the Hamming distance between every two feature points extracted from two grid maps to be spliced, and respectively calculating the median of Euclidean distances between the concentration points and the points of each key point;
s32: for each key point, K neighbor key points closest to the key point are found, and the distances between all the neighbor key points and the central key point are ensured to be smaller than a median value;
s33: constructing a non-directional edge for all the neighbor key points and the central key point which meet the constraint of the step S32, thereby constructing a median K neighbor graph;
s34: calculating the difference of adjacent matrixes of the median K neighbor graph as a residual matrix R;
s35: and summing the characteristic dissimilarity matrix D and the residual error matrix R, and generating a transmission cost matrix through index normalization processing.
5. The multi-robot grid map stitching method based on optimal map matching as claimed in claim 4, wherein in the step S4, the optimal transfer objective function is specifically expressed as:
wherein P and q are key point sets, P is a soft distribution matrix, mu and upsilon are transmission quality of the key points, and a is an entropy regular term coefficient and is set to be 1; solving the optimal transmission objective function by adopting a Sinkhorn-Knopp algorithm to obtain an optimal matching matrix P*。
6. The method for multi-robot grid map stitching based on OPMP according to claim 5, wherein in step S4, two augmentations are introduced due to the existence of unmatched outliers as a result of rough matchingNode gaAnd gbAnd introducing an optimal transmission constraint, specifically:
where α is the fixed augmented node mass.
7. The multi-robot grid map stitching method based on optimal map matching as claimed in claim 5, wherein in the step S5, the solving of the rigid body variation parameters between the optimal matching points by the least square method specifically comprises:
after the optimal matching key points are obtained, solving the following formula by adopting a least square method:
wherein p ismAnd q isc(m)The optimal matching key point is obtained; n is a radical ofmThe number of the optimal matching key points; r and t are a rotation matrix and a translation vector; and then creating a blank image with a proper size, directly copying the grid map A to be spliced to a new image, then carrying out image transformation on the whole grid map B according to the solved rigid body change parameters, and splicing the map according to a fusion function.
10. The method for multi-robot grid map stitching based on optimal map matching as claimed in claim 7, wherein in the step S5, the fusion function is specifically expressed as:
wherein h (-) represents a fusion function, x and y are the transformed two grid map pixel values, and a is a non-occupied pixel value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011445782.1A CN112581368B (en) | 2020-12-11 | 2020-12-11 | Multi-robot grid map splicing method based on optimal map matching |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011445782.1A CN112581368B (en) | 2020-12-11 | 2020-12-11 | Multi-robot grid map splicing method based on optimal map matching |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112581368A true CN112581368A (en) | 2021-03-30 |
CN112581368B CN112581368B (en) | 2024-01-09 |
Family
ID=75131048
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011445782.1A Active CN112581368B (en) | 2020-12-11 | 2020-12-11 | Multi-robot grid map splicing method based on optimal map matching |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112581368B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115290098A (en) * | 2022-09-30 | 2022-11-04 | 成都朴为科技有限公司 | Robot positioning method and system based on variable step length |
CN115861666A (en) * | 2022-12-22 | 2023-03-28 | 河北工程大学 | 3D image point cloud matching method, system, equipment and medium |
CN117612145A (en) * | 2023-12-15 | 2024-02-27 | 上海青翼工业软件有限公司 | Automatic part machining method and device, computer equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105806344A (en) * | 2016-05-17 | 2016-07-27 | 杭州申昊科技股份有限公司 | Raster map building method based on local map splicing |
US20160286137A1 (en) * | 2013-09-16 | 2016-09-29 | Duke University | Method for combining multiple image fields |
CN109978767A (en) * | 2019-03-27 | 2019-07-05 | 集美大学 | The ground laser SLAM drawing method based on multirobot collaboration |
US20190333187A1 (en) * | 2018-04-30 | 2019-10-31 | Tata Consultancy Services Limited | Method and system for frame stitching based image construction in an indoor environment |
-
2020
- 2020-12-11 CN CN202011445782.1A patent/CN112581368B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160286137A1 (en) * | 2013-09-16 | 2016-09-29 | Duke University | Method for combining multiple image fields |
CN105806344A (en) * | 2016-05-17 | 2016-07-27 | 杭州申昊科技股份有限公司 | Raster map building method based on local map splicing |
US20190333187A1 (en) * | 2018-04-30 | 2019-10-31 | Tata Consultancy Services Limited | Method and system for frame stitching based image construction in an indoor environment |
CN109978767A (en) * | 2019-03-27 | 2019-07-05 | 集美大学 | The ground laser SLAM drawing method based on multirobot collaboration |
Non-Patent Citations (2)
Title |
---|
R. ELVIRA ET AL.: "ORBSLAM-Atlas: a robust and accurate multi-map system", 2019 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), pages 6253 - 6259 * |
李雅梅 等: "一种基于局部特征的栅格地图拼接方法", 计算机应用与软件, vol. 37, no. 01, pages 110 - 115 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115290098A (en) * | 2022-09-30 | 2022-11-04 | 成都朴为科技有限公司 | Robot positioning method and system based on variable step length |
CN115290098B (en) * | 2022-09-30 | 2022-12-23 | 成都朴为科技有限公司 | Robot positioning method and system based on variable step length |
CN115861666A (en) * | 2022-12-22 | 2023-03-28 | 河北工程大学 | 3D image point cloud matching method, system, equipment and medium |
CN115861666B (en) * | 2022-12-22 | 2023-06-27 | 河北工程大学 | 3D image point cloud matching method, system, equipment and medium |
CN117612145A (en) * | 2023-12-15 | 2024-02-27 | 上海青翼工业软件有限公司 | Automatic part machining method and device, computer equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN112581368B (en) | 2024-01-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111028277B (en) | SAR and optical remote sensing image registration method based on pseudo-twin convolution neural network | |
CN108107444B (en) | Transformer substation foreign matter identification method based on laser data | |
CN106204574B (en) | Camera pose self-calibrating method based on objective plane motion feature | |
CN107818598B (en) | Three-dimensional point cloud map fusion method based on visual correction | |
CN111462210B (en) | Monocular line feature map construction method based on epipolar constraint | |
CN111047620A (en) | Unmanned aerial vehicle visual odometer method based on depth point-line characteristics | |
CN111998862B (en) | BNN-based dense binocular SLAM method | |
CN112581368A (en) | Multi-robot grid map splicing method based on optimal map matching | |
CN112132752B (en) | Fine splicing method for multi-view scanning point cloud of large complex curved surface | |
WO2022237225A1 (en) | Online real-time registration method for incomplete three-dimensional scanning point cloud having plane reference | |
CN113034593B (en) | 6D pose labeling method, system and storage medium | |
CN113808180B (en) | Heterologous image registration method, system and device | |
Wei et al. | Fgr: Frustum-aware geometric reasoning for weakly supervised 3d vehicle detection | |
CN110119768B (en) | Visual information fusion system and method for vehicle positioning | |
CN112507056A (en) | Map construction method based on visual semantic information | |
CN107123138B (en) | Based on vanilla-R point to the point cloud registration method for rejecting strategy | |
CN112633088A (en) | Power station capacity estimation method based on photovoltaic component identification in aerial image | |
CN109766903A (en) | A kind of point cloud model SURFACES MATCHING method based on curved surface features | |
CN109829459B (en) | Visual positioning method based on improved RANSAC | |
CN114119987A (en) | Feature extraction and descriptor generation method and system based on convolutional neural network | |
CN108520502A (en) | A kind of quick antidote of scanning electron microscope image drift | |
CN111881934A (en) | Method for discovering spatial relationship between attributes and categories of electronic components | |
CN104156952B (en) | A kind of image matching method for resisting deformation | |
CN112927304B (en) | Fish-eye lens calibration method based on convolutional neural network | |
CN104091308B (en) | Polar line correction method for reducing image distortion |
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 |