CN109358065B - Subway tunnel appearance detection method - Google Patents
Subway tunnel appearance detection method Download PDFInfo
- Publication number
- CN109358065B CN109358065B CN201811228215.3A CN201811228215A CN109358065B CN 109358065 B CN109358065 B CN 109358065B CN 201811228215 A CN201811228215 A CN 201811228215A CN 109358065 B CN109358065 B CN 109358065B
- Authority
- CN
- China
- Prior art keywords
- image
- images
- registration
- subway
- camera
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
- G01N2021/8893—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques providing a video image and a processed signal for helping visual decision
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a subway tunnel appearance detection method, which comprises the steps of carrying out high-definition image acquisition on the full-section surface of a subway structure along the advancing path of a subway at a certain speed by using a subway track image data acquisition platform; transversely placing the high-definition images acquired by each section, walking at a certain speed, and longitudinally placing the images acquired by a certain number of stations, so as to form an image matrix, and longitudinally and transversely splicing the image matrix to form a complete full-section subway high-definition image; then, correcting the spliced image by the subway segment CAD vector diagram to obtain a roaming diagram with position coordinates; and then marking defects such as cracks, penetration and the like on the roaming graph, and finally automatically calculating the size and the position of the defects, and automatically classifying and counting. The invention can obtain all apparent defect information of the surface of the subway tunnel, observe the defect information of sub-millimeter level water seepage, cracks, erosion, leakage, peeling, slab staggering and the like on the surface, and provide scientific and comprehensive data for the electronic management and maintenance of the subway tunnel.
Description
Technical Field
The invention relates to a subway tunnel appearance detection method.
Background
The urban subway tunnel structure is sensitive and complex in environment, is easily influenced by running vibration of subway trains, has construction defects of a connecting part of a lining structure and the lining structure, and is inevitable to generate various diseases such as water leakage, cracks, slab staggering, damage, large deformation and the like, so that the performance and the safety state of the structure are influenced. In order to comprehensively and timely grasp the safety state of the tunnel structure, observation, detection and monitoring are required. At present, the subway tunnel structure diseases are basically detected by adopting a manual inspection method, the problems of large detection labor amount, large subjectivity and the like exist, the detection efficiency is low, the precision is poor, the integrity and the accuracy of a detection result are difficult to ensure, the time evolution rule of the diseases cannot be recorded and analyzed, and the requirement of gradually expanding safety detection of the subway tunnel structure cannot be met.
Therefore, an automatic detection device and method for automatically, scanning, datamation and imaging is needed to comprehensively complete the appearance detection of the full section of the subway tunnel, so that the tunnel safety detection problem can be efficiently and comprehensively solved.
Disclosure of Invention
The invention aims to solve the technical problem that the prior art is insufficient, and provides a subway tunnel appearance detection method.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: a subway tunnel appearance detection method comprises the following steps:
1) according to the detection precision requirement gamma of the subway tunnel, the resolution dpi of the camera in the width direction, the distance d from the central line of the camera lens to the arc surface of the tunnel and the length SS of the photosensitive film of the camera sensorLCalculating the minimum focal length f of the camera;
2) from the focal length f of the lens, the length and width of the camera sensor respectively SSLAnd SSHThe radius r of the tunnel, the vertical distance d from the camera to the tunnel surface, and the angle beta and the width S in the walking direction of the arc surface of the tunnel covered by each camera;
3) determining the number n of cameras according to the image overlapping rate ol, the angle beta covered by a single camera, the width S of the walking direction, the range theta needing to be detected and the shutter interval time t2And the running speed v of the image acquisition platform;
4) n is to be2The images taken by the cameras at the same time are taken as a line of images, and the cameras are differentTaking pictures at intervals to form a series of images, forming a matrix image, i.e. M (n)1,n2)；
5) For n of matrix image1×n2Splicing the images;
6) transforming each pixel in the spliced image to a corresponding coordinate position in a CAD coordinate system through coordinate transformation, so that the spliced image is changed into an image file with geometric coordinates and size information, and image registration is completed;
7) and marking and measuring defects of the appearance of the subway tunnel by using the registered images.
Number of cameras
Running speed of image acquisition platform
The specific implementation process of the step 5) comprises the following steps:
1) according to the path point sequence of the acquisition platform, the image I to be splicedcThe splicing sequence is pre-ordered, wherein c represents the number of images to be spliced;
2) searching for characteristic points of the images to be spliced by adopting a SURF characteristic detection algorithm;
3) selecting adjacent images I to be splicedi，IjPerforming characteristic point purification by combining the RANSAC algorithm and geometric information between the images to be spliced to obtain an initial matching pair;
4) randomly selecting k feature point matching pairs from the initial matching pairs through multiple iterations, wherein k is more than or equal to 4; carrying out coarse registration on the images to be spliced and utilizing a formula v2＝Hv1Estimating a projective transformation matrix H, where v1＝(x1,y1,z1)，v2＝(x2,y2,z2) Let z be 1 for the corresponding feature point, pass v1，v2The homogeneous linear transformation relation between the elements H is obtainedi,jH is a 3X 3 valueA secondary matrix;
5) deforming the image after coarse registration by adopting a structure-preserving deformation algorithm to obtain Ii'，Ij'；
6) Using Graphcut algorithm, find Ii'，Ij' optimal seam of overlapping area;
7) using formulas
Calculating the distance from the characteristic point in the overlapping region to the pixel point on the stitching seam as the registration error, wherein vfRepresenting randomly selected feature points, m representing vfNumber of (a), vsRepresenting pixel points on the stitch, n representing vsThe number of (2);
8) repeating the steps 3) to 7), calculating all possible image registration relations, reserving the feature point pairs with the registration errors E (v) < delta, merging all the reserved feature point pairs, and calculating the final image registration relation as the image pair Ii，IjThe optimal registration relationship of;
9) repeating the step 8), and calculating the optimal registration relation between all adjacent image pairs;
10) optimizing all image registration relations by adopting a light speed adjustment algorithm;
11) searching the final registered image I in the step 10) by adopting a Graphcut algorithmcThe optimal seam of the overlapping area;
12) image I realization by multiband fusion technologycSeamless fusion splicing.
The implementation process of the step 6) comprises the following steps:
1) carrying out graying, equalization and binarization processing on the spliced image to obtain boundary position information of a structure in the image;
2) the structure boundary coordinates in the image correspond to the structure boundary coordinates in the CAD graph one by one, and the correct coordinate position (x, y) is determined for each pixel in the image through bilinear variation, wherein the bilinear mapping relation is as follows:
3) and solving coefficients in the double-line mapping relation, realizing one-to-one correspondence between the coordinates of each pixel and the actual CAD drawing coordinates, and finishing the registration of the image and the CAD drawing.
Compared with the prior art, the invention has the beneficial effects that: the method can obtain all apparent defect information of the surface of the subway tunnel, observe the defect information of sub-millimeter level water seepage, cracks, erosion, leakage, peeling, slab staggering and the like on the surface, and meet the requirements of actual engineering; and the macroscopic and microscopic evolution trends of the surface defects of the subway tunnel can be obtained, and a subway tunnel health electronic file system is established.
Drawings
Fig. 1 is a flowchart of a subway tunnel appearance detection system provided by the present invention;
fig. 2 and 3 are subway tunnel image acquisition platforms provided by the present invention;
FIG. 4 is a camera pose and LED lights;
101 is an acquisition platform, 102 is a single image acquisition range, and 103 is a subway track; 201 is high definition camera, 202 camera tray, 203 is liftable pole, 204 is the pulley that possesses the meter step function, 206 is LED annular light
Fig. 5 is a camera arrangement and shooting area diagram;
207 is a single camera shooting area, 208 is a transverse overlapping area, and 209 is a longitudinal overlapping area;
FIG. 6 is a subway tunnel image mosaic expansion diagram, namely an image block matrix;
FIG. 7 is a process diagram of a stitching algorithm;
fig. 8 is a diagram of a registration process.
Detailed Description
The method comprises a subway track data acquisition platform, image splicing processing, image registration processing and a defect database. Specifically, a subway track image data acquisition platform is used for carrying out high-definition image acquisition on the full-section surface of a subway structure along the travelling path of a subway at a certain speed; transversely placing the high-definition images acquired by each section, walking at a certain speed, and longitudinally placing the images acquired by a certain number of stations, so as to form an image matrix, and longitudinally and transversely splicing the image matrix to form a complete full-section subway high-definition image; then, correcting the spliced image by the subway segment CAD vector diagram to obtain a roaming diagram with position coordinates; and then marking defects such as cracks, penetration and the like on the roaming graph, and finally automatically calculating the size and the position of the defects, and automatically classifying and counting.
(1) The image acquisition platform 101 for the subway tunnel,
the subway tunnel acquisition platform comprises an image acquisition platform 101, a plurality of high-definition cameras 201, a camera tray 202, a lifting rod 203, a pulley 204 with a step counting function, a battery bracket 205 and an LED annular illuminating lamp 206. The implementation process is as follows:
in order to meet the detection precision requirement, the detection precision requirement gamma (gamma is more than 0.1mm and less than 0.2mm), the resolution dpi of the camera in the width direction and the focal length f of the camera must meet the following requirements:
firstly, according to the requirements of gamma and dpi of the resolution of a camera in the width direction of the subway tunnel detection precision, the distance d from the central line of a camera lens to the arc surface of the tunnel, and the length SS of a photosensitive film of a camera sensorLAnd calculating the minimum focal length f of the camera.
Then, the focal length f of the lens and the size SS of the sensorLAnd SSHThe radius r of the tunnel, the vertical distance d from the camera to the tunnel surface, and the angle beta and the width S of the walking direction of the tunnel arc surface covered by each camera
Then determining the number n of cameras according to the image overlapping rate ol (not less than 20 percent and not more than 30 percent), the angle beta covered by a single camera and the width S of the walking direction, the required detection range theta and the shutter interval time t2And the running speed v of the image acquisition platform.
The above steps determine the platform parameters and the operating speed. When platform acquisition is carried out, firstly, n is arranged according to the figure 3 and the figure 42A camera 201, an LED illuminating lamp 206 and a step-counting wheel 204, wherein a battery is arranged on a battery platform 205 and supplies power to all the cameras 201, the LED illuminating lamp and the step-counting wheel 204 on the disc 202; the lifting rod 203 is adjusted to ensure that the center of the disc is the same as the center of the subway tunnel, and the difference is less than 100 mm;
the image acquisition platform can be pushed to operate in an electric or hand-push mode according to the driving speed v, the shooting area of a single camera at each station is shown as 206 in fig. 5, 207 is a transverse overlapping area, and a plurality of cameras cover the section to be detected. The annular LED illuminating lamp 206 is turned on to push the image acquisition platform to operate, all cameras are automatically controlled to shoot simultaneously according to the same time interval, and the shooting times are n within a certain time1N is to be2The images taken simultaneously by the individual cameras at once are taken as one line of images (horizontal images in fig. 6), and the individual cameras are taken at different time intervals to form one column of images (vertical images in fig. 6), thereby forming a matrix image, i.e., M (n) images1,n2)。
(2) Image stitching processing
As shown in fig. 6, the subway tunnel acquisition platform has shot n1×n2The image processing method comprises the steps of opening and closing high definition images, wherein the images are a series of images shot by a platform at different positions in a close distance mode, and different from a camera at a fixed position, the images have motion parallax, so that multi-station image stitching is the core work of the technology.
These image files are first organized into a matrix image sequence, as shown in fig. 6. The specific splicing algorithm process is shown in fig. 7, and the specific implementation process of the algorithm is as follows:
1) according to the path point sequence of the acquisition platform, the image I to be splicedcThe splicing sequence is pre-ordered, wherein c represents the number of images to be spliced;
2) searching image characteristic points by adopting an SURF characteristic detection algorithm;
3) selectingTaking adjacent images I to be registeredi，IjExtracting feature points by combining the RANSAC algorithm with geometric information between images to obtain an initial matching pair;
4) randomly selecting k feature point matching pairs from the initial matching pairs through multiple iterations, wherein k is more than or equal to 4; coarse registration of the images and use of formulas
Estimating a projective transformation matrix H, where v1＝(x1,y1,z1)，v2＝(x2,y2,z2) Let z be 1 for the corresponding feature point, pass v1，v2The homogeneous linear transformation relation between the elements H is obtainedi,jH is a 3 x 3 homogeneous matrix;
5) deforming the image after coarse registration by adopting a structure-preserving deformation algorithm to obtain Ii'，Ij'；
6) Using Graphcut algorithm, find Ii'，Ij' optimal seam of overlapping area;
7) using formulas
Calculating the distance from the characteristic point in the image overlapping region to the pixel point on the stitching seam as a registration error, wherein vfRepresenting randomly selected feature points, m representing vfNumber of (a), vsRepresenting pixel points on the stitch, n representing vsThe number of (2);
8) repeating the steps 3) to 7), calculating all possible image registration relations, reserving the feature point pairs with the registration errors E (v) < delta, generally taking delta as 100, merging all reserved feature point pairs, and calculating the final image registration relation as the image pair Ii，IjThe optimal registration relationship of;
9) repeating the step 8), and calculating the optimal registration relation between all adjacent image pairs;
10) optimizing all image registration relations by adopting a light speed adjustment algorithm;
11) searching the final registered image I in the step 10) by adopting a Graphcut algorithmcThe optimal seam of the overlapping area;
12) image I realization by multiband fusion technologycSeamless fusion splicing.
(3) Graphic registration process
The high-definition images of the tunnel are spliced in the last step, and the high-definition panoramic image of the tunnel is established. The following is the registration work of the high definition images, which aims to: and cutting and calibrating the high-definition panoramic image to enable the size of each segment on the high-definition panoramic image to be consistent with the coordinate and the size of the corresponding CAD vector diagram, namely registering the structural object image into the CAD graph.
The registration method is to transform each pixel in the image to a corresponding coordinate position in the CAD coordinate system through coordinate change, so that the image becomes an image file with geometric coordinates and size information similar to a map. The method comprises the following specific steps:
a) and carrying out graying, equalization and binarization processing on the high-definition image graph to obtain boundary position information of a structure in the image.
b) The structure in the panoramic image deviates from its actual shape and size, and as shown in fig. 8, the purpose of the graph registration is to perform geometric transformation on the image data to achieve one-to-one correspondence with the CAD graph. The structure boundary coordinates in the image are in one-to-one correspondence with the structure boundary coordinates in the CAD drawing, and the correct coordinate position (x, y) is determined for each pixel in the image through bilinear variation, and the two-line mapping relationship of the two drawings in fig. 8 is as follows:
c) the 8 coefficients of a-h can be obtained by bringing the corresponding 4 points A and A ', B and B' into the formula (9).
e) After the transformation, the one-to-one correspondence between the coordinates of each pixel and the actual CAD drawing coordinates is realized, and the registration of the image and the CAD drawing is completed.
(4) Database system
After the graphs are registered, the relation between the image pixels and the actual size of the structure is determined by the vector CAD file. And determining the length, width and area of the defect according to the number of pixels. The registered images can mark and measure the defects of the appearance of the structure in a map-like manner. And the defect data are as follows: the position, length, width and area are recorded in the database to form a defect information database.
And classifying and numbering different types of defects in the tunnel in the registered image file, and measuring the positions, the sizes and other geometric dimensions of the defects. Historical detection results can be compared and analyzed, and the aging rule and trend of the subway tunnel can be researched and judged by combining a big data technology. Scientific and comprehensive data are provided for the electronic management and maintenance of the subway tunnel.
Claims (4)
1. The subway tunnel appearance detection method is characterized by comprising the following steps of:
1) according to the detection precision requirement gamma of the subway tunnel, the resolution dpi of the camera in the width direction, the distance d from the central line of the camera lens to the arc surface of the tunnel and the length SS of the photosensitive film of the camera sensorLCalculating the minimum focal length f of the camera;
2) from the focal length f of the lens, the length and width of the camera sensor respectively SSLAnd SSHThe radius r of the tunnel, the vertical distance d from the camera to the tunnel surface, and the angle beta and the width S in the walking direction of the arc surface of the tunnel covered by each camera;
3) determining the number n of cameras according to the image overlapping rate ol, the angle beta covered by a single camera, the width S of the walking direction, the range theta needing to be detected and the shutter interval time t2And the running speed v of the image acquisition platform;
4) n is to be2The images taken by the cameras at one time are taken as a line of images, and the single camera takes images at different time intervals to form a column of images, forming a matrix image, namely M (n)1,n2)；
5) For n of matrix image1×n2Splicing the images; the specific implementation process comprises the following steps:
a) according to the walking direction of the acquisition platform and the arrangement sequence of the cameras, images I to be splicedcThe splicing sequence is pre-ordered, wherein c represents the number of images to be spliced;
b) searching image characteristic points by adopting an SURF characteristic point detection algorithm;
c) selecting adjacent images I to be registeredi，IjExtracting feature points by combining the RANSAC algorithm with geometric information between images to obtain an initial matching pair;
d) randomly selecting k feature point matching pairs from the initial matching pairs through multiple iterations, wherein k is more than or equal to 4; coarse registration of the images and use of formulas
Estimating a projective transformation matrix H, where v1＝(x1,y1,z1)，v2＝(x2,y2,z2) Is Ii，IjLet z be 1, pass v1，v2The homogeneous linear transformation relation between the elements H is obtainedi,jH is a 3 x 3 homogeneous matrix;
e) deforming the image after coarse registration by adopting a structure-preserving deformation algorithm to obtain Ii'，Ij'；
f) Using Graphcut algorithm, find Ii'，Ij' optimal seam of overlapping area;
g) using formulas
Calculating the distance from the characteristic point in the image overlapping region to the pixel point on the stitching seam as a registration error, wherein vfRepresenting randomly selected feature points, m representing vfNumber of (a), vsRepresenting pixel points on the stitch, n representing vsThe number of (2);
h) repeating step c)Step g), calculating all possible image registration relations, reserving the characteristic point pairs with the registration errors E (v) < delta, merging all the reserved characteristic point pairs, and calculating the final image registration relation as the image pair Ii，IjThe optimal registration relationship of;
i) repeating the step h), and calculating the optimal registration relation between all adjacent image pairs;
j) optimizing all image registration relations by adopting a light beam adjustment algorithm;
k) searching the final registered image I in the step j) by adopting a Graphcut algorithmcThe optimal seam of the overlapping area;
l) implementing image I by multiband fusion techniquecSeamless fusion splicing;
6) transforming each pixel in the spliced image to a corresponding coordinate position in a CAD coordinate system through coordinate transformation, so that the spliced image is changed into an image file with geometric coordinates and size information, and image registration is completed;
7) and marking and measuring defects of the appearance of the subway tunnel by using the registered images.
2. The subway tunnel appearance detection method according to claim 1, wherein the number of cameras
3. The subway tunnel appearance detection method according to claim 1, wherein the running speed of the image acquisition platform
4. The subway tunnel appearance detection method according to claim 1, wherein the implementation process of step 6) comprises:
1) carrying out graying, equalization and binarization processing on the spliced image to obtain boundary position information of a structure in the image;
2) the structure boundary coordinates in the image correspond to the structure boundary coordinates in the CAD graph one by one, and the correct coordinate position (x, y) is determined for each pixel in the image through bilinear variation, wherein the bilinear mapping relation is as follows:
3) and solving coefficients in the double-line mapping relation, realizing one-to-one correspondence between the coordinates of each pixel and the actual CAD drawing coordinates, and finishing the registration of the image and the CAD drawing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811228215.3A CN109358065B (en) | 2018-10-22 | 2018-10-22 | Subway tunnel appearance detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811228215.3A CN109358065B (en) | 2018-10-22 | 2018-10-22 | Subway tunnel appearance detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109358065A CN109358065A (en) | 2019-02-19 |
| CN109358065B true CN109358065B (en) | 2021-08-24 |
Family
ID=65346115
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811228215.3A Active CN109358065B (en) | 2018-10-22 | 2018-10-22 | Subway tunnel appearance detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109358065B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109992012B (en) * | 2019-04-23 | 2022-03-25 | 云南交投集团投资有限公司 | Camera cluster attitude control method |
| CN110276258A (en) * | 2019-05-21 | 2019-09-24 | 智慧互通科技有限公司 | A kind of method and system identifying vehicle appearance |
| CN111062939B (en) * | 2019-12-31 | 2023-08-15 | 上海大学 | Method for rapidly screening quality of strip steel surface and automatically extracting defect characteristics |
| CN111310748B (en) * | 2020-03-17 | 2023-04-07 | 郑州航空工业管理学院 | Chinese image stone sports image acquisition and comparison device |
| CN111610193A (en) * | 2020-05-29 | 2020-09-01 | 武汉至科检测技术有限公司 | System and method for inspecting structural defects of subway tunnel segment by adopting multi-lens shooting |
| CN111665263B (en) * | 2020-06-11 | 2022-04-29 | 长江空间信息技术工程有限公司(武汉) | Shield diversion tunnel structure seam detection method |
| CN112435170B (en) * | 2020-12-04 | 2023-11-03 | 安徽圭目机器人有限公司 | Tunnel vault image splicing method |
| CN112862790B (en) * | 2021-02-18 | 2023-08-22 | 中国矿业大学(北京) | Subway tunnel crack positioning device and method based on linear array camera |
| CN113446953B (en) * | 2021-06-22 | 2023-03-07 | 深圳市市政设计研究院有限公司 | Subway tunnel deformation monitoring system based on digital photogrammetry technology |
| CN114062372B (en) * | 2021-11-15 | 2024-02-23 | 北京环安工程检测有限责任公司 | Intelligent analysis system for subway tunnel defects |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1691777A (en) * | 2004-02-19 | 2005-11-02 | 罗伯特·博世有限公司 | Image stabilization system and method for a video camera |
| CN1913590A (en) * | 2005-08-09 | 2007-02-14 | 罗伯特·博世有限公司 | Image stabilization system and method using video camera |
| CN104537898A (en) * | 2015-01-08 | 2015-04-22 | 西北工业大学 | Air-ground coordination unmanned aerial vehicle sensing and avoiding system and method |
| CN105352975A (en) * | 2015-12-02 | 2016-02-24 | 湖南拓达结构监测技术有限公司 | Bridge cable appearance detecting method |
| CN105354855A (en) * | 2015-12-02 | 2016-02-24 | 湖南拓达结构监测技术有限公司 | High-rise structure appearance detection device and method |
| CN205158453U (en) * | 2015-12-02 | 2016-04-13 | 湖南拓达结构监测技术有限公司 | Structure outward appearance detection device stands tall and erects |
| CN106657789A (en) * | 2016-12-29 | 2017-05-10 | 核动力运行研究所 | Thread panoramic image synthesis method |
-
2018
- 2018-10-22 CN CN201811228215.3A patent/CN109358065B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1691777A (en) * | 2004-02-19 | 2005-11-02 | 罗伯特·博世有限公司 | Image stabilization system and method for a video camera |
| CN1913590A (en) * | 2005-08-09 | 2007-02-14 | 罗伯特·博世有限公司 | Image stabilization system and method using video camera |
| CN104537898A (en) * | 2015-01-08 | 2015-04-22 | 西北工业大学 | Air-ground coordination unmanned aerial vehicle sensing and avoiding system and method |
| CN105352975A (en) * | 2015-12-02 | 2016-02-24 | 湖南拓达结构监测技术有限公司 | Bridge cable appearance detecting method |
| CN105354855A (en) * | 2015-12-02 | 2016-02-24 | 湖南拓达结构监测技术有限公司 | High-rise structure appearance detection device and method |
| CN205158453U (en) * | 2015-12-02 | 2016-04-13 | 湖南拓达结构监测技术有限公司 | Structure outward appearance detection device stands tall and erects |
| CN106657789A (en) * | 2016-12-29 | 2017-05-10 | 核动力运行研究所 | Thread panoramic image synthesis method |
Non-Patent Citations (4)
| Title |
|---|
| 地铁隧道渗漏水病害快速检测及图像处理技术;姚旭朋等;《城市轨道交通研究》;20171231;46-49 * |
| 基于多摄像头的实时视频拼接技术的研究与实现;陶荷梦;《中国优秀硕士学位论文全文数据库 信息科技辑》;20170315;I138-5370 * |
| 基于特征的航拍图像拼接技术研究，;庞兆峰;《中国优秀硕士学位论文全文数据库 信息科技辑》;20160815;I138-1270 * |
| 桥塔外观检测新技术及健康信息系统研究;谢海波;《中国公路学会养护与管理分会第八届学术年会论文集》;20180118;118-124 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109358065A (en) | 2019-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109358065B (en) | Subway tunnel appearance detection method | |
| CN109544679B (en) | Three-dimensional reconstruction method for inner wall of pipeline | |
| CN105550995B (en) | tunnel image splicing method and system | |
| CN111260615B (en) | Laser and machine vision fusion-based method for detecting apparent diseases of unmanned aerial vehicle bridge | |
| CN104484648B (en) | Robot variable visual angle obstacle detection method based on outline identification | |
| CN103940369A (en) | Quick morphology vision measuring method in multi-laser synergic scanning mode | |
| CN105511462B (en) | A kind of AGV air navigation aids of view-based access control model | |
| CN110826549A (en) | Inspection robot instrument image identification method and system based on computer vision | |
| CN103593663B (en) | A kind of image position method of money forme | |
| CN105352975A (en) | Bridge cable appearance detecting method | |
| CN106996748A (en) | A kind of wheel footpath measuring method based on binocular vision | |
| JP5303405B2 (en) | Vehicle inspection device | |
| CN113435420B (en) | Pavement defect size detection method and device and storage medium | |
| CN106023193B (en) | A kind of array camera observation procedure detected for body structure surface in turbid media | |
| CN102446355B (en) | Method for detecting target protruding from plane based on double viewing fields without calibration | |
| CN105741296A (en) | Auxiliary calibration method of 360-degre all-visual-angle aerial view panorama travelling crane | |
| CN109472778B (en) | Appearance detection method for towering structure based on unmanned aerial vehicle | |
| CN116164704A (en) | Tunnel defect real-time detection early warning system | |
| CN116625258A (en) | Chain spacing measuring system and chain spacing measuring method | |
| Deng et al. | Detection of loosening angle for mark bolted joints with computer vision and geometric imaging | |
| CN114113116A (en) | Accurate detection process method for micro-defects on surface of large-diameter element | |
| CN105354855A (en) | High-rise structure appearance detection device and method | |
| JP2009052907A (en) | Foreign matter detecting system | |
| CN109373901B (en) | Method for calculating center position of hole on plane | |
| CN205158453U (en) | Structure outward appearance detection device stands tall and erects |
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 |