CN112509138A - Indoor plastering robot high-precision three-dimensional reconstruction system based on LCOS - Google Patents
Indoor plastering robot high-precision three-dimensional reconstruction system based on LCOS Download PDFInfo
- Publication number
- CN112509138A CN112509138A CN202011457762.6A CN202011457762A CN112509138A CN 112509138 A CN112509138 A CN 112509138A CN 202011457762 A CN202011457762 A CN 202011457762A CN 112509138 A CN112509138 A CN 112509138A
- Authority
- CN
- China
- Prior art keywords
- lcos
- image
- matching
- plastering
- dimensional reconstruction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- 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
- G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention belongs to the field of building plastering robots and computer data processing, and particularly relates to an indoor plastering robot high-precision three-dimensional reconstruction system based on LCOS (reflective projection display). The method comprises the steps of projecting a pattern written in advance on a wall surface to be plastered by an LCOS projection technology, capturing the wall surface projection by using a binocular camera, and then performing data processing on the obtained picture by using a PC (personal computer) to complete the reconstruction of three-dimensional information of the wall surface to be plastered; the PC end is connected with the robot main control through the TCP and sends data to the plastering robot, and the plastering robot main control module controls the plastering machine to adjust to a specified operation position for plastering. The invention combines the active three-dimensional reconstruction projection technology with the construction plastering robot, and solves the problems of insufficient precision and poor effect of laser in the construction robot field.
Description
Technical Field
The invention belongs to the field of building plastering robots and computer data processing, and particularly relates to an indoor plastering robot high-precision three-dimensional reconstruction system based on LCOS (reflective projection display).
Background
With the development of computer technology, computer equipment has been deeply involved in various aspects of social life; vision and digital image processing techniques using computers are also becoming mature, and sharp peaks have been revealed in various sophisticated technical fields. The purpose of computer vision is to make the computer intelligent through analysis and research of relevant theories and technologies, so that the computer can achieve the effect similar to that of 'seeing' of human eyes. Specifically, the computer vision technology is to use a camera and a computer to replace human eyes, so that the computer has the functions of segmentation, classification, recognition, tracking, decision-making and the like of human eyes. The computer vision system is a complete artificial intelligence system which can create data of a plane image or a three-dimensional image to acquire required 'information', and can quickly acquire a large amount of data information and quickly process the acquired data information according to an intelligent algorithm, so that the computer vision system is easily integrated with design information and processing control information, and is widely applied to 'intelligence' of various industries.
In the construction industry, the application of computer vision systems is mainly embodied in the application of construction robots. With the popularization of the construction robot, the working efficiency is greatly improved, but the real-time performance, the accuracy and the simplicity of the construction robot operation are insufficient in certain specific scenes, especially in plastering. Under the general condition, traditional robot that plasters adopts laser to measure range, acquires whole wall distance information and needs a large amount of time, and the precision of laser survey distance is not enough moreover, and the wall that makes plaster is unsmooth to be difficult to reach the industry level, needs the manpower to carry out the secondary operation, has increased the cost of company, can not satisfy the rapid development demand of building trade.
Disclosure of Invention
Aiming at the problems in the background art, the invention provides a high-precision three-dimensional reconstruction system of an indoor plastering robot based on LCOS (liquid crystal on silicon), which aims to solve the problems that the obtained distance information of the whole wall surface is poor in real-time performance and low in precision in a mode of carrying out distance measurement by laser in the working process of the traditional building plastering robot, the plastered wall surface is uneven and difficult to achieve the industrial level, and the cost is high.
The invention adopts the specific scheme that:
an indoor plastering robot high-precision three-dimensional reconstruction system based on LCOS comprises: the system comprises a PC, an LCOS driving module, an equipment module, a plastering machine main control module, a plastering machine movement module and a wall surface;
the PC is used for providing LCOS control signals to the LCOS driving module; receiving image information captured by a binocular camera, performing three-dimensional reconstruction by using the received image information, and providing the reconstructed three-dimensional information to a main control module of the plastering machine;
the LCOS driving module is an LCOS development board written with a pattern program; the LCOS control signal is used for receiving the LCOS control signal provided by the PC and providing the internally written pattern to the equipment module through the control signal;
an equipment module: including binocular cameras and LCOS projection devices; the LCOS projection device receives the pattern provided by the LCOS driving module and projects the pattern to the wall surface to generate an image; the binocular camera is used for capturing an image projected to a wall surface by the LCOS projection equipment and transmitting the captured image to the PC;
the main control module of the plastering machine: and driving a plastering machine motion module to move according to the received reconstructed three-dimensional information to finish wall plastering.
Furthermore, the binocular camera is set to be in an external trigger mode, and is connected with the LCOS driving module through external trigger, so that the projection frame rates of the camera and the LCOS are consistent, and the camera can be rapidly photographed and stored while being projected.
Further, the pc machine further comprises a three-dimensional measurement system module for realizing three-dimensional reconstruction, and the three-dimensional reconstruction is realized by the following processes:
s1, calibrating the binocular camera by a Zhang Yongda calibration method to obtain internal parameters and external parameters; measuring the optical center distance between a left camera and a right camera in the binocular cameras and the focusing parameters of the left camera and the right camera;
s2, carrying out primary processing on the images captured by the binocular camera by adopting bipolar line correction according to the internal reference and the external reference obtained in the step S1, and keeping the left image and the right image captured by the binocular camera on the same line on the same pixel point so as to improve the matching precision of the left image and the right image;
s3, matching the image subjected to the preliminary processing in the step S2 by adopting a binocular cost matching function to obtain a disparity map; the matching process of the binocular cost matching function is as follows:
s3.1, respectively taking a window on the left image and the right image, and taking the windows as a template image T and a search image S;
s3.2, calculating the correlation coefficient of the template image and the search image by using an NCC algorithm, and determining the matching degree of pixel points in the template image T and the search image S; determining the position of the template image in the search image by the position with the maximum correlation coefficient, and obtaining an optimal matching result as a disparity map by calculating the matching success rate;
s4, performing parallax smoothing on the parallax map obtained in the step S3 by hole filling and least square plane fitting to obtain a final parallax;
and S5, calculating the three-dimensional information of the wall surface according to a triangulation principle formula through the internal reference and external reference obtained in S1, the focal length of the camera and the final parallax obtained in S4, and finishing the three-dimensional reconstruction.
Further, since the size of the intercepted window seriously affects the matching accuracy, in order to obtain an optimal matching result, the matching process of the binocular cost matching function is improved, and for convenience of description, the matching process is called as a window peak algorithm, and the calculation process of the window peak algorithm is as follows:
s1, firstly, selecting a window with the size of M × M in the searched image, selecting a window with the size of N × N in the template image, then calculating the correlation coefficient of the template image and the searched image by using an NCC algorithm, and finding out all points successfully matched between the template image and the searched image;
s2, calculating the matching success rate according to a left-right consistency check algorithm (LRC) aiming at all the successfully matched points found in the step S1, and eliminating mismatching points or shielded points; the left-right consistency checking algorithm judges whether the matching position is consistent with the original position according to a mode of matching from left to right and then from right to left, and eliminates mismatching points or shielding points.
S3, template windows with different sizes are selected for multiple times, the size of the selected template window is required to meet the condition that N is more than 1 and less than M, the size of the selected window is calculated by adopting a dichotomy algorithm every time, the steps S1-S2 are repeated for the selected window every time, and the matching degree of the selected window every time is calculated, so that the optimal window of the matching degree is obtained.
Further, since the NCC algorithm involves a large number of pixels and accumulation calculations in the process of calculating the correlation coefficient, in order to improve the usability and the practicability of the algorithm, the NCC algorithm adopts an integral graph algorithm to calculate the pixels and the accumulation in the process of calculating the correlation coefficient.
The invention provides a high-precision three-dimensional reconstruction system of an indoor plastering robot based on LCOS (liquid Crystal on silicon), which realizes active three-dimensional reconstruction by utilizing the LCOS projection technology. When the LCOS projection technology is utilized, the coded pattern needs to be written into an LCOS development board in advance, and then the development board is connected with a PC through a serial port; when the LCOS development board receives a control command of the pc machine, the LCOS projection equipment is driven to project a pattern on a wall surface.
In the three-dimensional reconstruction process, the binocular matching cost function is adopted to calculate the parallax, the accuracy of parameters adopted in the three-dimensional reconstruction process is improved, and the precision of the finally completed wall three-dimensional reconstruction information is improved. In order to further improve the precision, the calculation process of the binocular matching cost function is improved; specifically, in the matching process, the NCC algorithm is combined with a left-right consistency check (LRC) algorithm, and the optimal matching is found out by intercepting different windows for multiple times.
The adopted NCC algorithm mainly utilizes the correlation degree between normalized targets to be matched, is applied to object detection and identification in the detection and monitoring field of industrial production links, can effectively reduce the influence of illumination on image results, and has the following basic principle:
firstly, two images acquired by a binocular camera are assumed, the size of a template window T in the two images is set to be N x N, and the size of a search window S is set to be M x M, wherein M>N, translating the template T on the image S, and covering the subgraph by the templateIs recorded as Si,j(i, j) is the coordinate of the top left vertex of the subgraph in the search graph S; then, according to the correlation coefficient matrix ρ (i, j) of two sub-blocks with the same size in the two assumed images, that is:
and (3) solving a correlation coefficient according to the matrix, and taking the correlation coefficient as a basis for judging a cross-correlation value of the two images, wherein rho (i, j) is between-1 and 1 as a matched cost function, and the result can be obtained by judging the size of the correlation coefficient, wherein the closer to 1 indicates that the two random variables are more correlated, and the closer to-1 indicates that the two random variables are more uncorrelated.
Due to the adoption of the technical scheme, the invention has the following beneficial effects:
1. the invention combines the active three-dimensional reconstruction projection technology with the building plastering robot, and solves the problems of long time and poor real-time property for obtaining wall information in the field of building robots by laser.
2. The improved binocular matching cost algorithm is applied to the three-dimensional reconstruction of the building plastering robot for the first time, so that the three-dimensional reconstruction precision is improved; through improving the binocular matching cost function, the optimal matching window is selected, and the problem of uneven wall surface is further solved.
3. In general, after the system provided by the invention is adopted, the plastered wall surface does not need to be reworked for many times, so that the cost of the system is lower.
Drawings
FIG. 1 is a flowchart of the operation of an embodiment of the present invention;
FIG. 2 is a basic system configuration according to an embodiment of the present invention;
FIG. 3 is an exemplary diagram illustrating a three-dimensional reconstruction process according to an embodiment of the present invention;
fig. 4 is a graph of calculating the distance of the plastering machine from the wall surface.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings.
Fig. 2 is a schematic diagram of a basic configuration of a system according to an embodiment of the present invention. As shown in fig. 2, the system comprises a PC, an LCOS driving control module, a binocular camera and LCOS projection device, a main control of the plastering machine, and a plastering machine movement control module. The PC is mainly used for sending a control command to the LCOS drive control module and receiving an image captured by the binocular camera; and the received image information is utilized to carry out three-dimensional reconstruction, and then the reconstructed three-dimensional information is provided for the main control module of the plastering machine. The LCOS driving module is an LCOS development board embedded with a pattern program; for receiving the LCOS control signal provided by the PC and providing the embedded pattern to the device module via the control signal. The device module comprises an LCOS projection device and a binocular camera; the LCOS projection device receives the pattern provided by the LCOS driving module and projects the pattern to the wall surface to generate an image; the binocular camera is used for capturing an image projected to a wall surface by the LCOS projection equipment and transmitting the captured image to the PC; in order to ensure that the frame rates of the camera and the LCOS projection are consistent, rapid photographing and storage can be performed while projection is performed, in the embodiment, the binocular camera is set to be in an external trigger mode, and is connected with the LCOS driving module through external trigger. And the main control of the plastering machine drives the plastering machine motion module to move to a specified place according to the received reconstructed three-dimensional information to finish wall plastering. Because a large amount of operations are required in the three-dimensional reconstruction process, a three-dimensional reconstruction system module is arranged in the PC and is used for completing the three-dimensional reconstruction of the wall surface.
In this embodiment, the running environment of the PC is a Windows10 operating system, the development language is c + +, the compiler is Visual Studio2017, and the third-party dependent framework used is opencv4.1.0, PCL-1.9.0, and the like. The adopted development board is an embedded MYS-7Z020-C-S (766) development board, and the binocular camera is a Haikang camera. The main control of the plastering machine is replaced by a PC of a high-performance GPU, and the movement module of the plastering machine is replaced by a movement trolley.
FIG. 1 is a flowchart illustrating the operation of an embodiment of the present invention. As shown in fig. 1, the whole work flow of the system provided by the present embodiment is as follows:
and S1, the LCOS development board controls the LCOS development board through a control signal sent by the serial port software of the PC, and the LCOS development board provides the image information embedded in the LCOS development board for the LCOS projection equipment to project to the wall surface.
And S2, capturing wall image information through a preset external trigger mode by the binocular camera, and sending the captured image information to the PC.
And S3, performing matching calculation by a three-dimensional system measuring module arranged in the PC by adopting an improved NCC binocular cost matching algorithm to obtain a disparity map d. Before calculating by adopting a binocular cost matching algorithm of the improved NCC, calibrating a binocular camera by using a Zhang friend calibration method to obtain internal parameters and external parameters; and performing bipolar line correction on the captured pattern according to the obtained internal reference and external reference.
S4, measuring the optical center distance b between the left camera and the right camera in the binocular camera to be 30cm and the focusing f of the left camera and the right camera to be 80cm, combining the parallax map d obtained in the step S3, and obtaining the parallax map by a triangulation principle formula: and Z-f/d, calculating the three-dimensional information of the wall surface, wherein Z represents the three-dimensional information of the wall surface. Note that, in this step, the focus f refers to a distance between the intersection points of the left and right cameras. In addition, the parameters of the camera affect the accuracy of three-dimensional reconstruction, so once the camera is calibrated, the relative position and the focal length of the binocular camera cannot be adjusted, and the influence of the parameters of the camera on the three-dimensional reconstruction is reduced.
S5, sending the three-dimensional wall information obtained in the step S4 to a main control of the plastering machine through TCP;
and S6, controlling the movement module of the plastering machine to move by the main control of the plastering machine according to the received indoor wall surface three-dimensional information. Specifically, as shown in fig. 4, since the main control of the plastering machine and the PC end transmit data through TCP connection, assuming that a distance x between the camera and the motion control module is initially set, and the camera is located at a distance y from the wall surface, y-x is the distance from the motion control module to the wall surface.
In the whole working process, for the disparity map obtained in step S3, in order to improve accuracy, the embodiment further performs hole filling and least square plane fitting on the disparity map obtained in step S3.
It can be seen from the above working flows that the precision of the three-dimensional reconstructed wall surface directly affects the plastering effect of the plastering robot, and fig. 3 is an exemplary diagram of the three-dimensional reconstruction flow according to the embodiment of the present invention. As shown in fig. 3, the improved binocular cost matching function algorithm provided by the invention is adopted in the disparity map in the flow, the plastering effect is improved through the improved binocular cost matching function algorithm, and the problem of uneven plastering in the traditional mode is solved. In order to facilitate the distinction, the improved binocular cost matching function algorithm is named as a window peak algorithm. The calculation flow of the algorithm is as follows:
s1, firstly, selecting a window with the size of M × M in the searched image, selecting a window with the size of N × N in the template image, then calculating the correlation coefficient of the template image and the searched image by using an NCC algorithm, and finding out all points successfully matched between the template image and the searched image;
s2, calculating the matching success rate according to a left-right consistency check algorithm (LRC) aiming at all the successfully matched points found in the step S1, and eliminating mismatching points or shielded points; the left-right consistency checking algorithm judges whether the matching position is consistent with the original position according to a mode of matching from left to right and then from right to left, and eliminates mismatching points or shielding points.
S3, template windows with different sizes are selected for multiple times, the size of the selected template window is required to meet the condition that N is more than 1 and less than M, the size of the selected window is calculated by adopting a binary algorithm every time, the steps S1-S2 are repeated for the selected window every time, the matching degree of the selected window every time is calculated, and therefore the window with the optimal matching degree is obtained, and the optimal disparity map is obtained.
It should be noted that, because of the LCOS-based active three-dimensional reconstruction system used in the system, the required indoor environment should be ensured to be in a weak light or dark environment to eliminate the influence of strong light on the projection picture, which causes a problem that the subsequent three-dimensional reconstruction cannot be performed due to the failure of binocular matching; if the operation is performed in sunny days, the light can be shielded by cloth, so that the plastering efficiency is avoided, and the plastering performance is improved. The binocular camera lens needs to be cleaned regularly, so that the camera is prevented from being covered by a large amount of dust, and subsequent reconstruction is prevented from being influenced. The plastering machine motion control module is located in an indoor flat space without obstacles, no obstacles are arranged around the plastering machine motion control module to block the movement of the trolley, the plastering machine main control module can receive indoor wall surface three-dimensional information sent by a PC (personal computer) end through TCP (transmission control protocol) connection in real time, and the plastering machine main control module controls the motion control module to move to a specified position to carry out indoor operation.
Claims (5)
1. An indoor plastering robot high-precision three-dimensional reconstruction system based on LCOS comprises: PC, LCOS drive module, equipment module, mechanical float master control module, mechanical float motion module and wall, its characterized in that:
the PC is used for providing LCOS control signals to the LCOS driving module; receiving image information captured by a binocular camera, performing three-dimensional reconstruction by using the received image information, and providing the reconstructed three-dimensional information to a main control module of the plastering machine;
the LCOS driving module is an LCOS development board embedded with a pattern program; receiving LCOS control signal provided by PC, and providing embedded pattern to device module via the control signal;
the equipment module: including binocular cameras and LCOS projection devices; the LCOS projection device receives the pattern provided by the LCOS driving module and projects the pattern to the wall surface to generate an image; the binocular camera is used for capturing an image projected to a wall surface by the LCOS projection equipment and transmitting the captured image to the PC;
and the plastering machine main control module drives the plastering machine motion module to move according to the received reconstructed three-dimensional information so as to finish wall plastering.
2. The LCOS-based indoor plastering robot high precision three-dimensional reconstruction system as claimed in claim 1, wherein: the binocular camera is set to be in an external trigger mode and is connected with the LCOS driving module through external trigger so that the projection frame rates of the camera and the LCOS are consistent.
3. The LCOS-based indoor plastering robot high precision three-dimensional reconstruction system as claimed in claim 1, wherein: the pc machine further comprises a three-dimensional measurement system module for realizing three-dimensional reconstruction, and the three-dimensional reconstruction is realized by the module through the following processes:
s1, calibrating the binocular camera by a Zhang Yongda calibration method to obtain internal parameters and external parameters; measuring the optical center distance between a left camera and a right camera in the binocular cameras and the focusing parameters of the left camera and the right camera;
s2, carrying out primary processing on the images captured by the binocular camera by adopting bipolar line correction according to the internal reference and the external reference obtained in the step S1, and keeping the left image and the right image captured by the binocular camera on the same line on the same pixel point so as to improve the matching precision of the left image and the right image;
s3, matching the image subjected to the preliminary processing in the step S2 by adopting a binocular cost matching function to obtain a disparity map; the matching process of the binocular cost matching function is as follows:
s3.1, respectively taking a window on the left image and the right image, and taking the windows as a template image T and a search image S;
s3.2, calculating the correlation coefficient of the template image and the search image by using an NCC algorithm, and determining the matching degree of pixel points in the template image T and the search image S; determining the position of the template image in the search image by the position with the maximum correlation coefficient, and obtaining an optimal matching result as a disparity map by calculating the matching success rate;
s4, performing parallax smoothing on the parallax map obtained in the step S3 by hole filling and least square plane fitting to obtain a final parallax;
and S5, calculating the three-dimensional information of the wall surface according to a triangulation principle formula through the internal reference and external reference obtained in S1, the focal length of the camera and the final parallax obtained in S4, and finishing the three-dimensional reconstruction.
4. The LCOS-based indoor plastering robot high precision three-dimensional reconstruction system according to claim 3, wherein: the matching of the binocular cost matching function in step S3 is an improved binocular cost matching function, which is hereinafter referred to as a window peak algorithm for convenience of description, and the calculation process of the window peak algorithm is as follows:
s1, firstly, selecting a window with the size of M × M in the searched image, selecting a window with the size of N × N in the template image, then calculating the correlation coefficient of the template image and the searched image by using an NCC algorithm, and finding out all points successfully matched between the template image and the searched image;
s2, calculating the matching success rate according to a left-right consistency check algorithm (LRC) aiming at all the successfully matched points found in the step S1, and eliminating mismatching points or shielded points; the left-right consistency checking algorithm judges whether the matching position is consistent with the original position according to a mode of matching from left to right and then from right to left, and eliminates mismatching points or shielding points;
s3, template windows with different sizes are selected for multiple times, the size of the selected template window is required to meet the condition that N is more than 1 and less than M, the size of the selected window is calculated by adopting a dichotomy algorithm every time, the steps S1-S2 are repeated for the selected window every time, and the matching degree of the selected window every time is calculated, so that the optimal window of the matching degree is obtained.
5. A high-precision three-dimensional reconstruction system of an indoor plastering robot based on LCOS according to claim 3 or 4, wherein: the NCC algorithm employs an integral graph algorithm in the process of calculating the correlation coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011457762.6A CN112509138B (en) | 2020-12-11 | 2020-12-11 | LCOS-based high-precision three-dimensional reconstruction system for indoor plastering robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011457762.6A CN112509138B (en) | 2020-12-11 | 2020-12-11 | LCOS-based high-precision three-dimensional reconstruction system for indoor plastering robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112509138A true CN112509138A (en) | 2021-03-16 |
| CN112509138B CN112509138B (en) | 2023-07-18 |
Family
ID=74973686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011457762.6A Active CN112509138B (en) | 2020-12-11 | 2020-12-11 | LCOS-based high-precision three-dimensional reconstruction system for indoor plastering robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112509138B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113077504A (en) * | 2021-04-12 | 2021-07-06 | 中国电子科技集团公司第二十八研究所 | Large scene depth map generation method based on multi-granularity feature matching |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799318A (en) * | 2012-08-13 | 2012-11-28 | 深圳先进技术研究院 | Human-machine interaction method and system based on binocular stereoscopic vision |
| CN104677308A (en) * | 2015-01-30 | 2015-06-03 | 宋展 | Three-dimensional scanning method for high-frequency two-value strip |
| US20170184852A1 (en) * | 2014-08-18 | 2017-06-29 | Sony Corporation | Image display device and display apparatus |
| CN107170043A (en) * | 2017-06-19 | 2017-09-15 | 电子科技大学 | A kind of three-dimensional rebuilding method |
| CN109341589A (en) * | 2018-10-17 | 2019-02-15 | 深圳市华汉伟业科技有限公司 | A kind of raster image projecting method, three-dimensional rebuilding method and three-dimensional reconstruction system |
| CN109887071A (en) * | 2019-01-12 | 2019-06-14 | 天津大学 | A kind of 3D video image dendoscope system and three-dimensional rebuilding method |
| CN111145342A (en) * | 2019-12-27 | 2020-05-12 | 山东中科先进技术研究院有限公司 | Binocular speckle structured light three-dimensional reconstruction method and system |
-
2020
- 2020-12-11 CN CN202011457762.6A patent/CN112509138B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799318A (en) * | 2012-08-13 | 2012-11-28 | 深圳先进技术研究院 | Human-machine interaction method and system based on binocular stereoscopic vision |
| US20170184852A1 (en) * | 2014-08-18 | 2017-06-29 | Sony Corporation | Image display device and display apparatus |
| CN104677308A (en) * | 2015-01-30 | 2015-06-03 | 宋展 | Three-dimensional scanning method for high-frequency two-value strip |
| CN107170043A (en) * | 2017-06-19 | 2017-09-15 | 电子科技大学 | A kind of three-dimensional rebuilding method |
| CN109341589A (en) * | 2018-10-17 | 2019-02-15 | 深圳市华汉伟业科技有限公司 | A kind of raster image projecting method, three-dimensional rebuilding method and three-dimensional reconstruction system |
| CN109887071A (en) * | 2019-01-12 | 2019-06-14 | 天津大学 | A kind of 3D video image dendoscope system and three-dimensional rebuilding method |
| CN111145342A (en) * | 2019-12-27 | 2020-05-12 | 山东中科先进技术研究院有限公司 | Binocular speckle structured light three-dimensional reconstruction method and system |
Non-Patent Citations (3)
| Title |
|---|
| M. BOCK等: "Adaptive spiral phase elements for the generation of few-cycle vortex pulses", pages 1 * |
| 夏军等: "新型高分辨率三维显示器件与系统的基础研究年度报告", no. 12, pages 164 - 165 * |
| 李维等: "LCOS微型显示器的应用", no. 04, pages 6 - 8 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113077504A (en) * | 2021-04-12 | 2021-07-06 | 中国电子科技集团公司第二十八研究所 | Large scene depth map generation method based on multi-granularity feature matching |
| CN113077504B (en) * | 2021-04-12 | 2021-11-12 | 中国电子科技集团公司第二十八研究所 | Large scene depth map generation method based on multi-granularity feature matching |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112509138B (en) | 2023-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106767399B (en) | The non-contact measurement method of logistics goods volume based on binocular stereo vision and dot laser ranging | |
| CN108177143B (en) | Robot positioning and grabbing method and system based on laser vision guidance | |
| CN104626206B (en) | The posture information measuring method of robot manipulating task under a kind of non-structure environment | |
| CN106650701B (en) | Binocular vision-based obstacle detection method and device in indoor shadow environment | |
| GB2580691A (en) | Depth estimation | |
| US11132804B2 (en) | Hybrid depth estimation system | |
| WO2020063058A1 (en) | Calibration method for multi-degree-of-freedom movable vision system | |
| CN111780715A (en) | Visual ranging method | |
| CN108161930A (en) | A kind of robot positioning system of view-based access control model and method | |
| CN110851978B (en) | Camera position optimization method based on visibility | |
| CN112509138B (en) | LCOS-based high-precision three-dimensional reconstruction system for indoor plastering robot | |
| CN113334380A (en) | Robot vision calibration method, control system and device based on binocular vision | |
| CN113781576A (en) | Binocular vision detection system, method and device for multi-degree-of-freedom pose real-time adjustment | |
| CN111429571B (en) | Rapid stereo matching method based on spatio-temporal image information joint correlation | |
| CN111862170A (en) | Optical motion capture system and method | |
| US20230152887A1 (en) | Systems and methods for calibrating an eye tracking system | |
| CN111002349A (en) | Robot following steering method and robot system adopting same | |
| CN113781573B (en) | Visual odometer method based on binocular refraction and reflection panoramic camera | |
| CN212256370U (en) | Optical motion capture system | |
| CN115546829A (en) | Pedestrian spatial information sensing method and device based on ZED (zero-energy-dimension) stereo camera | |
| CN114972539A (en) | Machine room camera plane online calibration method, system, computer equipment and medium | |
| CN109903325B (en) | Ground accurate description method based on stereoscopic vision depth information | |
| CN211028657U (en) | Intelligent welding robot system | |
| Grigorescu et al. | Controlling Depth Estimation for Robust Robotic Perception | |
| CN115019167B (en) | Fusion positioning method, system, equipment and storage medium based on mobile terminal |
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 |