CN113487485A - Octree map hole completion method based on class gray level image - Google Patents
Octree map hole completion method based on class gray level image Download PDFInfo
- Publication number
- CN113487485A CN113487485A CN202110779087.7A CN202110779087A CN113487485A CN 113487485 A CN113487485 A CN 113487485A CN 202110779087 A CN202110779087 A CN 202110779087A CN 113487485 A CN113487485 A CN 113487485A
- Authority
- CN
- China
- Prior art keywords
- map
- points
- pixel
- point
- coordinate
- 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 33
- 230000002159 abnormal effect Effects 0.000 claims abstract description 31
- 230000008439 repair process Effects 0.000 claims abstract description 8
- 239000011800 void material Substances 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 9
- 230000008569 process Effects 0.000 abstract description 6
- 230000001131 transforming effect Effects 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 239000000523 sample Substances 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- 238000012216 screening Methods 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
- 230000000007 visual effect Effects 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
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- 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
- G06T3/06—Topological mapping of higher dimensional structures onto lower dimensional surfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
本发明公开一种基于类灰度图像的八叉树地图空洞补全方法，先读取八叉树地图，然后提取八叉树地图中每个栅格的中心点三维坐标，并将每个栅格的中心点投影到世界坐标系下的俯视平面，利用双线性插值法补全孤立的地图空洞点的高度信息，再将所有坐标点转换为二维灰度图像；基于Canny算子和梯度算子优化的Criminisi算法，对二维灰度图像像素异常区域进行修复，确定待补全块状空洞区域内所有栅格中心点的空间位置；最后，将地图空洞点和待补全块状空洞区域所有栅格中心点的空间位置存储到八叉树地图，以实现地图补全。本发明的方法较好的适用于于地外星球的探测过程中，用于地外探测车进行环境理解和路径规划，从而提高探测车的探测效率。
The invention discloses a method for filling octree map holes based on quasi-gray image. The octree map is read first, then the three-dimensional coordinates of the center point of each grid in the octree map are extracted, and each grid The center point of the grid is projected to the top-down plane under the world coordinate system, and the height information of the isolated map holes is completed by bilinear interpolation, and then all coordinate points are converted into two-dimensional grayscale images; based on Canny operator and gradient The operator-optimized Criminisi algorithm repairs the abnormal pixel area of the two-dimensional grayscale image, and determines the spatial position of all grid center points in the block-shaped hole area to be filled; The spatial positions of all grid center points in the area are stored in the octree map for map completion. The method of the invention is preferably suitable for the detection process of extraterrestrial planets, and is used for the environment understanding and path planning of the extraterrestrial exploration vehicle, thereby improving the detection efficiency of the detection vehicle.
Description
Technical Field
The invention relates to the technical field of robot environment perception, in particular to an octree map cavity completion method based on a gray-level-like image.
Background
With the rapid development of computer technology, the research of robots is in depth and the demand of people for robots is expanding, and robots capable of autonomous navigation and intelligent movement become the focus and key point of research. The map is used as one of the bases for autonomous positioning, obstacle avoidance and route planning of the robot, and the importance degree of the map is self-evident.
There are many expressions of maps, such as feature point maps, grid maps, topological maps, and so on. At present, most of robots use point cloud maps in mapping systems, and have some obvious defects: the map form is not compact, the way to handle the overlap is not good enough, and it is difficult to use for navigation. Octree maps are often used for collision detection, path planning, positioning and navigation, and other application tasks, and compared with other forms of maps, Octree maps focus more on modeling a three-dimensional space environment, and store maps using a data structure of Octree (Octree), so that maps can be compressed and updated elegantly, and resolution is adjustable.
Because the truth value of the map is difficult to obtain, the qualitative evaluation of the map constructed by the robot is difficult, and the related research facing the map is few. Due to factors such as sight shielding, sensor visual angle blind areas and the like in the moving process of the mobile robot, a plurality of holes exist in the map building process, and great inconvenience is brought to autonomous navigation and environment understanding of the robot.
Disclosure of Invention
The invention aims to solve the problem of cavities in the octree map building process of a robot, and provides a octree map cavity completion method of gray-level-like images, so that a map with complete information is provided for a mobile robot.
The purpose of the invention is realized by the following technical scheme:
an octree map hole completion method based on a gray-level-like image comprises the following steps:
the method comprises the following steps: reading an octree map;
step two: extracting a three-dimensional coordinate of a central point of each grid in the octree map to form a coordinate set, projecting the central point of each grid to a top-view plane under a world coordinate system, complementing height information of isolated map void points by using a bilinear interpolation method, adding the complemented map void points into the coordinate set, and converting the updated coordinate set into a two-dimensional gray image;
step three: repairing the two-dimensional gray image pixel abnormal region based on a Criminisi algorithm optimized by a Canny operator and a gradient operator, and determining the spatial positions of the central points of all grids in the block-shaped cavity region to be repaired corresponding to the pixel abnormal region;
step four: and storing the spatial positions of the map hole points and all the grid central points of the block-shaped hole area to be complemented to an octree map so as to realize map complementing.
Further, the second step is realized by the following steps:
(2.1) extracting a three-dimensional coordinate of a central point of each grid in the octree map to form a coordinate set, projecting the central point to a top view plane (xoy plane) in a world coordinate system, setting the height of an isolated map cavity point in the xoy plane as an abnormal value, and extracting the maximum height value of each coordinate point of the projected plane;
(2.2) complementing the height information of the isolated map cavity points in the xoy plane by using a bilinear interpolation method, adding the complemented isolated cavity points into a coordinate set, and updating the coordinate set;
(2.3) zooming and translating the x value and the y value of the coordinate point in the updated coordinate set according to the boundary size of the map and the resolution of the octree map, namely respectively transforming the minimum value of the x value and the y value to the pixel coordinate origin, transforming the coordinate points with the resolution of the octree map as an interval into continuous pixel coordinate points, and transforming the height z value corresponding to the coordinate points into the pixel value of the corresponding pixel point, thereby transforming the coordinate set into a two-dimensional gray image.
Further, the third step is realized by the following steps:
(3.1) determining a corresponding block-shaped cavity area to be compensated and a boundary thereof according to the pixel abnormal area in the two-dimensional gray level image corresponding to the abnormal value;
(3.2) optimizing a priority function by combining a Canny operator and a gradient operator, balancing the influence of confidence and data items on the priority, calculating the priority of the edge pixel, finding a target block with the maximum priority, searching a known pixel block which is most matched with the target block by using a Criminisi algorithm, repairing the target block, and finishing the updating of the abnormal pixel value in the gray level image;
(3.4) screening the updated abnormal pixel points of the two-dimensional gray image, re-determining the boundaries of the abnormal pixel points, and repeating the steps (3.2) and (3.3) until the updating of the pixel values of all the abnormal pixel points is completed;
and (3.5) determining the pixel values of all abnormal pixel points according to the updated values, converting the continuous pixel coordinate points into x and y coordinate values through inverse transformation, and converting the pixel values of the corresponding pixel points into height values z, thereby determining the three-dimensional space positions of all map completion points.
Further, the calculation formula of the height information of the map hole point is as follows:
wherein HaIs an isolated map hole point (x)a，ya) Height of (H)11、H21、H12And H22Respectively are coordinate points (x) next to the four positive directions of the isolated hole point1，y1)、(x2，y1)、(x1，y2) And (x)2，y2) Corresponding heightInformation; selecting coordinate points which are close to four positive directions of the isolated hole point, namely selecting points which are parallel to the x direction and the y direction and are separated from the hole point of the map by one octree map resolution ratio, namely x1＝xa-0.5，x2＝xa+0.5，y1＝ya-0.5，y2＝ya+0.5, thereby ensuring that the screened map hole point is an isolated single point.
Further, the Canny operator and the gradient operator are combined to optimize a priority function of
P(p)＝αC(p)+βD(p)+γLG(p)
Wherein, the coefficients α, β, y are weight factors adjusted according to the texture features of different images, and c (p) is a confidence term in the Criminisi algorithm, which is the ratio of the sum of known information in the pixel block to be repaired to the area of the pixel block to be repaired; d (p) is a data item in the Criminisi algorithm and represents the structural information in the pixel block to be repaired; LG (x, y) is a structural term,
LG(x，y)＝|dx(i，j)|+|dy(i，j)|
dx(i，j)＝Canny(i+1，j)-Canny(i，j)；
dy(i，j)＝Canny(i，j+1)-Canny(i，j)；
dx (i, j) and dy (i, j) represent the amount of change in Canny value in the x direction and the y direction, respectively, of the point (i, j).
The invention has the following beneficial effects:
(1) the method of the invention utilizes a bilinear interpolation method to fill isolated void points, and utilizes a gray-like image method to fill block-shaped void areas. The method can complement the map by combining the characteristics of the linear structure, the two-dimensional texture and the like of the environment, so that a more complete map is obtained, the decision of the robot is facilitated, and the robot is guided to carry out path planning better.
(2) Due to the fact that the texture characteristics and the structural characteristics of the earth surface environment of the extraterrestrial celestial sphere are rich, the extraterrestrial celestial sphere detection method can be well used for the extraterrestrial planet detection process and used for the extraterrestrial planet detection vehicle to conduct environment understanding and path planning, and therefore detection efficiency of the detection vehicle is improved.
Drawings
Fig. 1 is a flowchart of an octree map void completion method based on a class gray image according to this embodiment;
fig. 2 is a structure diagram of the Criminisi algorithm optimized based on the Canny operator and the gradient operator in this embodiment.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments, and the objects and effects of the present invention will become more apparent, it being understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.
The octree map void completion method based on the class gray level image, disclosed by the invention, has a flow chart shown in figure 1, and comprises the following steps:
step one, reading an octree map.
An octree map is made up of many small squares, each square representing the probability that the cell is occupied. When the resolution is high, the square is small; at lower resolutions, the square is large. Specifically, the resolution of the octree map is set to 0.5 in this embodiment, and a suitable resolution can be set for a specific probe vehicle and a specific probe task in the specific task process;
extracting three-dimensional coordinates of the central point of each grid in the octree map to form a coordinate set, projecting the central point of each grid to a top-view plane under a world coordinate system, complementing the height information of the isolated void points by using a bilinear interpolation method, adding the complemented isolated void points into the coordinate set, and converting the updated coordinate set into a two-dimensional gray image;
(2.1) inquiring the center coordinate of each leaf node in the octree map by using a leaf _ iterator batch inquiry iterator and a getCoordinate () function, thereby extracting the three-dimensional coordinate of each grid center point in the octree map, and projecting each grid center point to a xoy plane, wherein the maximum height value of the grid center point corresponding to each projection point is extracted as the height information of the projection point because a plurality of grid center points in the octree map are likely to correspond to the same projection point after being projected to the xoy plane; and defining points without height values in the xoy plane as map hole points, and setting the height of the map hole points as abnormal high values. In specific implementation, the height of the map void point can be set as an abnormal value, so that the map void point can be screened from the octree map. In this embodiment, the abnormal value is set to 100.
(2.2) complementing the height information of the isolated map cavity points in the xoy plane by using a bilinear interpolation method, adding the complemented isolated cavity points into a coordinate set, and updating the coordinate set;
when the coordinates in four positive directions next to the hole point (with the resolution of the octree map as an interval) correspond to the height value, the hole point is defined as an isolated map hole point. And calculating the height information of the isolated hole points of the map by using the height information corresponding to the coordinates in the four positive directions adjacent to the hole points through a bilinear interpolation method:
wherein HaIs an isolated map hole point (x)a，ya) Height of (H)11、H21、H12And H22Respectively are coordinate points (x) next to the four positive directions of the isolated hole point1，y1)、(x2，y1)、(x1，y2) And (x)2，y2) Corresponding height information. Selecting coordinate points which are close to four positive directions of the isolated hole point, namely selecting points which are parallel to the x direction and the y direction and are separated from the hole point of the map by one octree map resolution ratio, namely x1＝xa-0.5，x2＝xa+0.5，y1＝ya-0.5，y2＝ya+0.5, thereby ensuring that the screened map hole point is an isolated single point.
(2.3) zooming and translating the x value and the y value of the coordinate point in the updated coordinate set according to the boundary size of the map and the resolution of the octree map, namely respectively transforming the minimum value of the x value and the y value to the pixel coordinate origin, transforming the coordinate points with the resolution of the octree map as an interval into continuous pixel coordinate points, and transforming the height z value corresponding to the coordinate points into the pixel value of the corresponding pixel point, thereby transforming the coordinate set into a two-dimensional gray image.
And thirdly, repairing the pixel abnormal region of the two-dimensional gray image based on a Criminisi algorithm optimized by a Canny operator and a gradient operator, and determining the spatial positions of all grid center points of the block-shaped cavity region to be repaired corresponding to the pixel abnormal region.
The Criminisi algorithm is an image restoration method based on samples, and the main work is to continuously select the best matching block from a known region and fill the best matching block into a region to be restored from outside to inside. The method has the advantages that places with more complete areas and stronger structures around the boundary of the cavity are repaired preferentially, filling accuracy is guaranteed to the maximum extent through priority calculation, the accuracy is transmitted orderly, and the fuzzy phenomenon generated when large-size damaged images are repaired can be avoided.
(3.1) determining a corresponding block-shaped cavity area to be compensated and a boundary thereof according to the pixel abnormal area in the two-dimensional gray level image corresponding to the abnormal value set in the step two;
(3.2) combining a Canny operator and a gradient operator to optimize a priority function, calculating the priority of the edge pixels, finding out a target block with the maximum priority, searching a known pixel block which is most matched with the target block by using a Criminisi algorithm, repairing the target block, and finishing the updating of the abnormal pixel value in the gray level image;
the priority of point p at the boundary of the void region in the Criminisi algorithm is:
P(p)＝C(p)×D(p)
wherein, C (p) is a confidence term and is the ratio of the sum of the known information in the pixel block to be repaired to the area of the pixel block to be repaired, and D (p) is a data term and represents the structural information in the pixel block to be repaired. The priority function is performed in a multiplicative fashion, and as repair work progresses, any decrease in the confidence term or data term causes the priority value to decrease rapidly. In the subsequent repair process, the size of the priority value cannot correctly reflect the repair sequence of the block to be repaired.
The gradient calculation formula of a general image is as follows:
where I is the entire image, G (x, y) is the gradient at point (x, y), and dx and dy are the gradient values in the x and y directions, respectively. In most cases, the repair order does not depend only on the boundary information of the image, but also the texture information plays a very important role. The Canny operator is simple in calculation, can identify actual edges in the image, and is sensitive to the edges of the image, but the edges in the image can be identified only once, and possible image noise can be identified as the edges, but the extraction efficiency of the complex image is not high, so that the accurate repair sequence cannot be completely obtained by using the Canny characteristic or the gradient characteristic alone. Aiming at the problems, the Canny operator is applied to gradient operation, a structural item based on the combination of the Canny operator and the gradient operator is provided, and is merged into the priority in an adding mode, and the specific definition is as follows:
dx (i, j) and dy (i, j) represent Canny value variations of the point (i, j) in the x direction and the y direction respectively, and can well describe texture information around the sample block and reflect information of the block edge. The priority function in the original Criminisi algorithm is in a product form, so that the problem that the repair sequence is unreliable due to zero priority value easily occurs, and the priority function in the Criminisi algorithm is defined in a weighted sum form, so that the problems can be avoided, and the influence of confidence and data items on the priority can be balanced. Finally, the improved priority formula is as follows:
P(p)＝αC(p)+βD(p)+γLG(p)
wherein the coefficients α, β, γ need to be adjusted according to the texture features of different images.
When the gradient at the central pixel point p is large and the surrounding texture is rich, the larger the constraint term lg (p), the larger the sample block priority centered on p, and vice versa.
(3.4) screening the updated abnormal pixel points of the two-dimensional gray image, re-determining the boundaries of the abnormal pixel points, and repeating the steps (3.2) and (3.3) until the updating of the pixel values of all the abnormal pixel points is completed;
and (3.5) determining the pixel values of all abnormal pixel points according to the updated values, converting the continuous pixel coordinate points into x and y coordinate values through inverse transformation, and converting the pixel values of the corresponding pixel points into height values z, thereby determining the three-dimensional space positions of all map completion points.
And fourthly, storing the spatial positions of the map hole points and all the grid central points of the block-shaped hole area to be supplemented to the octree map to complete the map supplementation.
The invention is used for complementing the outline of the missing area in the octree map, can display whether barriers exist at a certain height in the space map, and does not influence the functionality of the navigation map. In practical application, a certain threshold value can be set between the ground and the outer contour for completion according to the requirements of a practical scene, so that a complete three-dimensional map is obtained.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and although the invention has been described in detail with reference to the foregoing examples, it will be apparent to those skilled in the art that various changes in the form and details of the embodiments may be made and equivalents may be substituted for elements thereof. All modifications, equivalents and the like which come within the spirit and principle of the invention are intended to be included within the scope of the invention.
Claims (5)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110779087.7A CN113487485B (en) | 2021-07-09 | 2021-07-09 | Octree map hole complement method based on gray-like image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110779087.7A CN113487485B (en) | 2021-07-09 | 2021-07-09 | Octree map hole complement method based on gray-like image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113487485A true CN113487485A (en) | 2021-10-08 |
| CN113487485B CN113487485B (en) | 2024-09-17 |
Family
ID=77938338
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110779087.7A Active CN113487485B (en) | 2021-07-09 | 2021-07-09 | Octree map hole complement method based on gray-like image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113487485B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114147970A (en) * | 2021-11-22 | 2022-03-08 | 浙江闪铸三维科技有限公司 | Method for automatic compensation platform of 3D printer |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090284528A1 (en) * | 2008-05-15 | 2009-11-19 | Tyson Wayne Jensen | Software processing apparatus and method for creating three-dimensional topologically complete surface boundary representations from arbitrary polygon models |
| CN108648221A (en) * | 2018-05-10 | 2018-10-12 | 重庆大学 | A kind of depth map cavity restorative procedure based on mixed filtering |
| CN111179344A (en) * | 2019-12-26 | 2020-05-19 | 广东工业大学 | Efficient mobile robot SLAM system for repairing semantic information |
| CN113012063A (en) * | 2021-03-05 | 2021-06-22 | 北京未感科技有限公司 | Dynamic point cloud repairing method and device and computer equipment |
-
2021
- 2021-07-09 CN CN202110779087.7A patent/CN113487485B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090284528A1 (en) * | 2008-05-15 | 2009-11-19 | Tyson Wayne Jensen | Software processing apparatus and method for creating three-dimensional topologically complete surface boundary representations from arbitrary polygon models |
| CN108648221A (en) * | 2018-05-10 | 2018-10-12 | 重庆大学 | A kind of depth map cavity restorative procedure based on mixed filtering |
| CN111179344A (en) * | 2019-12-26 | 2020-05-19 | 广东工业大学 | Efficient mobile robot SLAM system for repairing semantic information |
| CN113012063A (en) * | 2021-03-05 | 2021-06-22 | 北京未感科技有限公司 | Dynamic point cloud repairing method and device and computer equipment |
Non-Patent Citations (2)
| Title |
|---|
| 王晶晶, 何兴恒, 龙蒙蒙: "基于八叉树分解法的空间散乱点数据场可视化", 微机发展, no. 11, 10 November 2005 (2005-11-10) * |
| 程效军;何桂珍;: "适用于多值曲面修复的空洞边界提取方法及应用", 测绘学报, no. 06, 15 December 2012 (2012-12-15) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114147970A (en) * | 2021-11-22 | 2022-03-08 | 浙江闪铸三维科技有限公司 | Method for automatic compensation platform of 3D printer |
| CN114147970B (en) * | 2021-11-22 | 2023-10-13 | 浙江闪铸集团有限公司 | Method for automatically compensating platform of 3D printer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113487485B (en) | 2024-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111429574B (en) | Mobile robot positioning method and system based on three-dimensional point cloud and vision fusion | |
| CN111486855B (en) | Indoor two-dimensional semantic grid map construction method with object navigation points | |
| CN110458939B (en) | Indoor scene modeling method based on visual angle generation | |
| Wang et al. | Mvdepthnet: Real-time multiview depth estimation neural network | |
| US10803613B2 (en) | Computer vision systems and methods for ground surface condition detection and extraction from digital images | |
| CN108898676B (en) | Method and system for detecting collision and shielding between virtual and real objects | |
| JP5778237B2 (en) | Backfill points in point cloud | |
| Budroni et al. | Automated 3D reconstruction of interiors from point clouds | |
| CN113409410A (en) | Multi-feature fusion IGV positioning and mapping method based on 3D laser radar | |
| CN105096386A (en) | Method for automatically generating geographic maps for large-range complex urban environment | |
| CN110223351B (en) | Depth camera positioning method based on convolutional neural network | |
| CN108230247A (en) | Generation method, device, equipment and the application program of three-dimensional map based on high in the clouds | |
| CN113191213B (en) | High-resolution remote sensing image newly-added building detection method | |
| US20210304411A1 (en) | Map construction method, apparatus, storage medium and electronic device | |
| Tian et al. | Lidar super-resolution based on segmentation and geometric analysis | |
| CN113487485A (en) | Octree map hole completion method based on class gray level image | |
| CN115018934A (en) | Three-dimensional image depth detection method combining cross skeleton window and image pyramid | |
| Yoon et al. | Practical implementation of semi-automated as-built BIM creation for complex indoor environments | |
| CN111815684B (en) | Space multivariate feature registration optimization method and device based on unified residual error model | |
| CN111986320B (en) | Smart city application-oriented DEM and oblique photography model space fitting optimization method | |
| CN114419155B (en) | Visual image construction method based on laser radar assistance | |
| CN111198563A (en) | Terrain recognition method and system for dynamic motion of foot type robot | |
| CN115187736A (en) | Target map generation method and device, and AR map generation method and device | |
| Bai et al. | Application of unmanned aerial vehicle multi-vision image 3D modeling in geological disasters | |
| Hesami et al. | Range segmentation of large building exteriors: A hierarchical robust approach |
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 |