CN112733694A - ORB feature-based track laying roller identification method and system - Google Patents
ORB feature-based track laying roller identification method and system Download PDFInfo
- Publication number
- CN112733694A CN112733694A CN202110003484.5A CN202110003484A CN112733694A CN 112733694 A CN112733694 A CN 112733694A CN 202110003484 A CN202110003484 A CN 202110003484A CN 112733694 A CN112733694 A CN 112733694A
- Authority
- CN
- China
- Prior art keywords
- corner
- feature
- roller
- point
- orb
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a method and a system for identifying a track laying roller based on ORB characteristics, which relate to the field of image identification and mainly comprise the following steps: fixing a line laser right above a plane to be scanned and adjusting levelness, enabling laser scanning data to act along the track direction, performing corner detection and BRIEF descriptor description characteristics by using a FAST algorithm, extracting ORB characteristics and matching with a preset roller characteristic model, performing error detection through symmetrical characteristics, screening out a corner characteristic set, and finally calculating a centroid coordinate control manipulator to pick up a roller. After characteristic points are extracted through a FAST algorithm, the angular point information is screened through the judgment of angular point symmetry, so that the identification result is more accurate and quicker, and meanwhile, a BRIEF descriptor is adopted for characteristic description, so that the characteristic matching is quicker and the real-time performance is enhanced; the intelligent identification of the roller is calculated through a series of algorithms, and the placement error of the roller is maintained within an allowable range.
Description
Technical Field
The invention relates to the field of image recognition, in particular to a track laying roller recognition method and system based on ORB characteristics.
Background
With the economic development, the development trend of railways is advanced with the development trend of wide transportation range, high economic benefit, small occupied area and the like. In the past railway development, ballast tracks occupy the leading position of track construction, along with the updating and development of technology, the upper limit of the speed of a track bearing train is gradually increased, and the ballast tracks are more commonly used for low-speed tracks due to the limitation of the ballast tracks on the speed, so that the railway construction is welcomed by the age of ballastless tracks. But the ballastless track has high construction cost, low corresponding maintenance cost, long maintenance period and long service life, and is suitable for passenger lines with high speed and frequent departure. Therefore, ballastless tracks are mostly adopted for domestic high-speed tracks. Railway construction is also constantly in progress.
The ballastless track laying is divided into several steps. Firstly, construction preparation is carried out, materials and machines required by construction are implemented, and the state of construction equipment is checked. The rail car carries out loading and transportation, and auxiliary rail laying workpieces, namely rollers, are placed every few sleepers. The placement of the rollers needs to be aligned with the track and the deviation from the neutral line in the track needs to be within an allowable range. Then the rail car is placed the track on the cylinder, raises the track after accomplishing to place, takes out the cylinder, uses fastener straining track, and follow-up construction requirement detects, accomplishes the shop rail.
In track laying, the placement of the drum requires positioning, within the tolerance limits. This work, together with the subsequent recycling of the drum, is highly repeatable, labor intensive, labor demanding and requires a long time for manual work. Along with the continuous popularization of intellectualization and automation, the track laying of the ballastless track refers to the automation equipment and the realization of automatic track laying is a great tendency. Therefore, the mechanical vision is adopted to carry out the research of rolling positioning, grabbing, placing and recovering, and the practical significance is very strong.
Disclosure of Invention
In order to enable a roller recovery vehicle to quickly and accurately identify a roller, the invention provides an identification method of a track laying roller based on ORB characteristics, which is characterized by comprising the following steps:
s1: fixing the line laser at the bottom of the vehicle, and acquiring distance scanning data of a line laser scanning calibration plane;
s2: judging whether the error of the distance scanning data is within a preset range, if so, entering the next step, otherwise, calibrating the horizontal angle of the line laser and returning to the step S1;
s3: receiving a set of laser scanning data of a line laser scanning target to be detected;
s4: performing corner feature extraction on the set of laser scanning data by using a FAST corner extraction algorithm, and obtaining a corner set of a target to be detected;
s5: representing and describing the detected corner point set through a BRIEF operator to obtain a characteristic descriptor of the roller;
s6: performing feature matching on the feature descriptor and a preset roller feature model through a Hamming distance, and obtaining a matching result;
s7: judging whether the target to be detected is a roller or not according to the matching result, if so, entering the next step, and if not, returning to the step S3;
s8: integrating symmetrical corner points in the corner point set, judging whether the data difference of the symmetrical corner points is within a preset threshold value, if so, entering the next step, and if not, deleting the symmetrical corner point group;
s9: and carrying out weighted average on the angle point set to obtain a centroid coordinate, and grabbing the roller according to the centroid coordinate.
Further, the step S3 is followed by the step of:
s31: the laser scan data is filtered by mid-pass filtering.
Further, in step S5, the BRIEF operator is a method that randomly selects N pairs of random data points within a preset window with the feature point as a center, performs binary assignment on each pair of random data points according to the height information, and finally forms a binary code, and the formula is:
in the formula, τ (p; x, y) is a binary value of the feature point, p (x) is a height value of (u1, v1) at the random point x, and p (y) is a height value of (u2, v2) at the random point y.
Further, in step S6, the preset drum feature model is obtained by performing corner feature extraction on a standard drum model by using a harris corner detection algorithm.
Further, in step S8, the symmetry points include symmetry points in both horizontal and vertical directions.
The invention also provides an identification system of the track laying roller based on the ORB characteristics, which comprises a fixed platform, a level gauge, a line laser, an upper computer and a manipulator, wherein:
the fixed platform is used for fixing the line laser to the bottom of the vehicle;
the level meter is used for calibrating the installation levelness of the line laser;
the line laser is used for scanning a target to be measured or a calibration plane and respectively obtaining a set of laser scanning data or distance scanning data;
the upper computer is used for judging whether the error is within a preset range according to the distance scanning data, and if not, adjusting the horizontal angle of the line laser;
the upper computer is also used for extracting the corner features of the set of laser scanning data by using a FAST corner extraction algorithm and obtaining a corner set of the target to be detected;
the upper computer is also used for representing and describing the detected corner point set through a BRIEF operator to obtain a characteristic descriptor of the roller, performing characteristic matching on the characteristic descriptor and a preset roller characteristic model through a Hamming distance, and obtaining a matching result; and when the target to be detected is judged to be the roller, carrying out weighted average on the symmetrical angular points of which the data differences accord with the preset threshold value to obtain a centroid coordinate, and controlling the manipulator to grab the roller according to the centroid coordinate.
Furthermore, the upper computer also comprises a preprocessing unit which is used for filtering the laser scanning data by a middle-pass filtering method.
Further, the BRIEF operator is a method that, with a feature point as a center, N pairs of random data points are randomly selected in a preset window, each pair of random data points is subjected to binary assignment according to height information, and finally a binary code is formed, and the formula is as follows:
in the formula, τ (p; x, y) is a binary value of the feature point, p (x) is a height value of (u1, v1) at the random point x, and p (y) is a height value of (u2, v2) at the random point y.
Further, the upper computer further comprises a template generating unit, and the template generating unit is used for extracting the angular point characteristics of the standard roller model by adopting a harris angular point detection algorithm and generating a preset roller characteristic model.
Further, the symmetric corner points include symmetric corner points in both the horizontal direction and the vertical direction.
Compared with the prior art, the invention at least has the following beneficial effects:
(1) according to the track laying roller identification method and system based on ORB characteristics, the horizontal angle of the line laser is calibrated, so that the laser scanning data is high in stability and strong in consistency, and meanwhile, the characteristics can be matched through an ORB algorithm due to the fact that the installation position is fixed;
(2) considering the characteristic of symmetry of each corner point of the roller, after characteristic points (corner points) are extracted through the FAST, and the BRIEF descriptor describes the characteristics, the corner point information is screened through judgment of the symmetry of the corner points, so that the identification result is more accurate and quicker, and meanwhile, the FAST algorithm extraction and the BRIEF descriptor are adopted for description, so that the characteristic point extraction is more stable;
(3) the rollers are intelligently identified through a series of algorithm calculation, the placement error of the rollers is maintained within an allowable range, the working repeatability and high strength of constructors are avoided, and the construction efficiency is improved;
(4) automatic construction is not influenced by environment and weather, and the construction period is shortened.
Drawings
FIG. 1 is a diagram of method steps for an ORB feature based track-laying drum identification algorithm;
FIG. 2 is a system block diagram of an ORB feature based track-laying drum identification system;
FIG. 3 is a schematic view of the installation of a line laser;
FIG. 4 is a schematic view of corner point positioning of a drum;
fig. 5 is a three-dimensional schematic view of the drum.
Description of reference numerals: 1-line laser, 2-fixed platform, 3-laser plane, 4-corner.
Detailed Description
The following are specific embodiments of the present invention and are further described with reference to the drawings, but the present invention is not limited to these embodiments.
Example one
In order to realize the automatic identification of the roller and improve the efficiency of the roller recovery vehicle for recovering the roller, as shown in fig. 1, the invention provides an identification method of a track laying roller based on the ORB characteristic, which comprises the following steps after the system is initialized:
s1: fixing the line laser at the bottom of the vehicle, and acquiring distance scanning data of a line laser scanning calibration plane (meanwhile, in order to acquire distance information, distance information of an encoder is fused to acquire a two-position height matrix);
s2: and judging whether the error of the distance scanning data is within a preset range, if so, entering the next step, otherwise, calibrating the horizontal angle of the line laser and returning to the step S1.
As shown in fig. 3, the line laser 1 is fixed to the bottom of the drum recycling cart by a fixing platform 2, and the levelness of the line laser 1 needs to be calibrated before the line laser is put into use. Because the accuracy of the data in the algorithm calculation can be ensured only when the installation position of the line laser 1 is fixed and the levelness is checked.
S3: receiving a set of laser scanning data of a line laser scanning target to be detected;
s31: the laser scan data is filtered by mid-pass filtering.
In practical application, as shown in fig. 3, when the line laser acquires data by scanning, the laser scanning data needs to be filtered by a median filtering algorithm, so as to eliminate the influence of the huge data caused by the dead zone of the laser triangulation method (the line laser emits a laser plane 3 which is a triangular plane and is also called a knife plane, and a line laser is formed when the laser plane irradiates on an object) and the data lost due to illumination and other reasons.
S4: performing corner feature extraction (Feather Detect) on the set of laser scanning data by adopting a FAST corner extraction algorithm, and obtaining a corner set of a target to be detected; (FAST corner extraction: finding the corners in the image, compared with the original FAST algorithm, the main direction of the feature points is calculated in ORB, and the rotation invariance is added for BRIEF descriptor)
S5: representing and describing the detected corner point set by a BRIEF operator (Feather Descriptor) to obtain a feature Descriptor of the roller; (BRIEF descriptor: describing the pixel area around the key point found in the previous step, since BRIEF is very sensitive to image rotation, ORB improves BRIEF, and the rotational invariance of BRIEF is enhanced by using the direction information calculated in the previous step)
S6: and performing feature matching (FeatherMatch) on the feature descriptors and the preset roller feature model through Hamming distance (Hamming distance), and obtaining a matching result.
In order to realize the simplicity and high efficiency of feature point acquisition, the invention selects a mode of FAST algorithm to extract feature points in a set of laser scanning data, and BRIEF description is used to describe the extracted corner points.
The BRIEF operator is a binary code which is formed by randomly selecting N pairs of random data points in a preset window by taking a characteristic point as a center, carrying out binary assignment on each pair of random data points according to height information and finally forming the binary code, and the formula is as follows:
in the formula, τ (p; x, y) is a binary value of the feature point, p (x) is a height value of (u1, v1) at the random point x, and p (y) is a height value of (u2, v2) at the random point y.
And the preset roller characteristic model adopts a harris angular point detection algorithm to extract the angular point characteristics of the standard roller model. Harris improved the Moravec corner detection algorithm by applying differential operations and autocorrelation matrices. Redefining the formula of the gray scale intensity change by using a differential operator, wherein the gray scale intensity change is expressed as:
in the formula
Is the coefficient of the gaussian window at (u, v). And X and Y are first-order gradients of the pixel points in the X direction and the Y direction, reflect the gray change direction of each pixel point in the image, and are extracted as angular points if the gray of the pixel points (X and Y) in the two directions is changed greatly enough.
S7: judging whether the target to be detected is a roller or not according to the matching result, if so, entering the next step, and if not, returning to the step S3;
s8: and integrating the symmetrical corner points in the corner point set, judging whether the data difference of the symmetrical corner points is within a preset threshold value, if so, entering the next step, and if not, deleting the symmetrical corner point group.
As shown in fig. 4 and 5, the overall structure of the drum is a symmetrical structure as seen from the three-dimensional figures. The characteristic points (the corner points 4, which are circles in fig. 4) are also arranged in a left-right symmetrical manner in the corner point positioning schematic diagram, and include symmetrical corner points 4 in the horizontal direction and the vertical direction. Based on this point, by judging whether the error of the coordinate data of the left and right symmetric corner points 4 is within the preset threshold value, the wrong corner point information can be filtered out, so that the rest corner point information better conforms to the actual corner point characteristics of the roller. By utilizing the judging step, the accuracy of the extraction of the angular point features of the roller is improved, so that the accuracy of roller identification is improved.
S9: and carrying out weighted average on the angle point set to obtain a centroid coordinate, and grabbing the roller according to the centroid coordinate.
Meanwhile, after the drum gripping is completed, if the stop signal is not received, the process returns to step S3.
Example two
In order to better describe the technical content of the present invention and understand the overall composition structure, as shown in fig. 2, the present invention provides an ORB-feature-based track-laying drum identification system, which includes a fixed platform, a level, a line laser, an upper computer and a manipulator, wherein:
the fixed platform is used for fixing the line laser to the bottom of the vehicle;
the level meter is used for calibrating the installation levelness of the line laser;
the line laser is used for scanning a target to be measured or a calibration plane and respectively obtaining a set of laser scanning data or distance scanning data;
the upper computer is used for judging whether the error is within a preset range according to the distance scanning data, and if not, adjusting the horizontal angle of the line laser;
the upper computer is also used for extracting the corner features of the set of laser scanning data by using a FAST corner extraction algorithm and obtaining a corner set of the target to be detected;
the upper computer is also used for representing and describing the detected corner point set through a BRIEF operator to obtain a characteristic descriptor of the roller, performing characteristic matching on the characteristic descriptor and a preset roller characteristic model through a Hamming distance, and obtaining a matching result; and when the target to be detected is judged to be the roller, carrying out weighted average on the symmetrical angular points of which the data differences accord with the preset threshold value to obtain a centroid coordinate, and controlling the manipulator to grab the roller according to the centroid coordinate.
Furthermore, the upper computer also comprises a preprocessing unit which is used for filtering the laser scanning data by a middle-pass filtering method.
Further, the BRIEF operator is a method that, with a feature point as a center, N pairs of random data points are randomly selected in a preset window, each pair of random data points is subjected to binary assignment according to height information, and finally a binary code is formed, and the formula is as follows:
in the formula, τ (p; x, y) is a binary value of the feature point, p (x) is a height value of (u1, v1) at the random point x, and p (y) is a height value of (u2, v2) at the random point y.
Further, the upper computer further comprises a template generating unit, and the template generating unit is used for extracting the angular point characteristics of the standard roller model by adopting a harris angular point detection algorithm and generating a preset roller characteristic model.
In summary, according to the identification method and system of the track-laying roller based on the ORB features, the horizontal angle of the line laser is calibrated, so that the laser scanning data is high in stability and strong in consistency, and meanwhile, the feature matching can be performed through the ORB algorithm due to the fixed installation position; in consideration of the symmetry characteristics of each corner point of the roller, after characteristic points (corner points) are extracted through the FAST, and the BRIEF descriptor describes the characteristics, the corner point information is screened through the judgment of the symmetry of the corner points, so that the identification result is more accurate and quicker, and meanwhile, the FAST algorithm extraction and the BRIEF descriptor are adopted for description, so that the characteristic point extraction is more stable.
The rollers are intelligently identified through a series of algorithm calculation, the placement error of the rollers is maintained within an allowable range, the working repeatability and high strength of constructors are avoided, and the construction efficiency is improved; and automatic construction is adopted, so that the influence of the environment and the weather is avoided, and the construction period is shortened.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.
Claims (10)
1. A method for identifying a track-laying drum based on ORB features, comprising the steps of:
s1: fixing the line laser at the bottom of the vehicle, and acquiring distance scanning data of a line laser scanning calibration plane;
s2: judging whether the error of the distance scanning data is within a preset range, if so, entering the next step, otherwise, calibrating the horizontal angle of the line laser and returning to the step S1;
s3: receiving a set of laser scanning data of a line laser scanning target to be detected;
s4: performing corner feature extraction on the set of laser scanning data by using a FAST corner extraction algorithm, and obtaining a corner set of a target to be detected;
s5: representing and describing the detected corner point set through a BRIEF operator to obtain a characteristic descriptor of the roller;
s6: performing feature matching on the feature descriptor and a preset roller feature model through a Hamming distance, and obtaining a matching result;
s7: judging whether the target to be detected is a roller or not according to the matching result, if so, entering the next step, and if not, returning to the step S3;
s8: integrating symmetrical corner points in the corner point set, judging whether the data difference of the symmetrical corner points is within a preset threshold value, if so, entering the next step, and if not, deleting the symmetrical corner point group;
s9: and carrying out weighted average on the angle point set to obtain a centroid coordinate, and grabbing the roller according to the centroid coordinate.
2. The method for identifying a track-laying drum based on ORB features as claimed in claim 1, wherein step S3 is followed by the steps of:
s31: the laser scan data is filtered by mid-pass filtering.
3. The method as claimed in claim 1, wherein in step S5, the BRIEF operator is a method that randomly selects N pairs of random data points within a preset window by taking the feature point as a center, performs binary assignment on each pair of random data points according to the height information, and finally forms a binary code, and the formula is:
in the formula, τ (p; x, y) is a binary value of the feature point, p (x) is a height value of (u1, v1) at the random point x, and p (y) is a height value of (u2, v2) at the random point y.
4. The method for identifying an ORB-feature-based track-laying drum as claimed in claim 1, wherein the preset drum feature model is a standard drum model with harris corner detection algorithm for corner feature extraction in step S6.
5. The method for identifying a track-laying drum based on ORB features as claimed in claim 1, wherein the symmetry points comprise symmetry points in both horizontal and vertical directions in step S8.
6. The utility model provides an identification system of cylinder is laid to track based on ORB characteristic which characterized in that, includes fixed platform, spirit level, line laser ware, host computer and manipulator, wherein:
the fixed platform is used for fixing the line laser to the bottom of the vehicle;
the level meter is used for calibrating the installation levelness of the line laser;
the line laser is used for scanning a target to be measured or a calibration plane and respectively obtaining a set of laser scanning data or distance scanning data;
the upper computer is used for judging whether the error is within a preset range according to the distance scanning data, and if not, adjusting the horizontal angle of the line laser;
the upper computer is also used for extracting the corner features of the set of laser scanning data by using a FAST corner extraction algorithm and obtaining a corner set of the target to be detected;
the upper computer is also used for representing and describing the detected corner point set through a BRIEF operator to obtain a characteristic descriptor of the roller, performing characteristic matching on the characteristic descriptor and a preset roller characteristic model through a Hamming distance, and obtaining a matching result; and when the target to be detected is judged to be the roller, carrying out weighted average on the symmetrical angular points of which the data differences accord with the preset threshold value to obtain a centroid coordinate, and controlling the manipulator to grab the roller according to the centroid coordinate.
7. The ORB-feature-based track-laying drum identification system as claimed in claim 6, wherein the host computer further comprises a preprocessing unit for filtering the laser scan data by mid-pass filtering.
8. The ORB-feature-based track-laying drum identification system according to claim 6, wherein the BRIEF operator is a binary code obtained by randomly selecting N pairs of random pixel points within a preset window by centering on a feature point, and performing binary assignment on each pair of random pixel points according to pixel size, and finally:
in the formula, τ (p; x, y) is a binary value of the feature point, p (x) is a height value of (u1, v1) at the random point x, and p (y) is a height value of (u2, v2) at the random point y.
9. The ORB-feature-based track-laying drum identification system according to claim 6, wherein the upper computer further comprises a template generation unit for performing corner feature extraction on the standard drum model by using a harris corner detection algorithm and generating a preset drum feature model.
10. An ORB-feature-based track-laying drum identification system as claimed in claim 6, wherein the symmetry points comprise symmetry points in both horizontal and vertical directions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110003484.5A CN112733694B (en) | 2021-01-04 | 2021-01-04 | ORB feature-based track laying roller identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110003484.5A CN112733694B (en) | 2021-01-04 | 2021-01-04 | ORB feature-based track laying roller identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112733694A true CN112733694A (en) | 2021-04-30 |
| CN112733694B CN112733694B (en) | 2022-04-29 |
Family
ID=75589597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110003484.5A Active CN112733694B (en) | 2021-01-04 | 2021-01-04 | ORB feature-based track laying roller identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112733694B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930253A (en) * | 2012-10-31 | 2013-02-13 | 中国矿业大学(北京) | Coal and rock identification method based on image discrete multi-wavelet transform |
| CN107377398A (en) * | 2017-07-21 | 2017-11-24 | 北京巴龙机电科技有限公司 | A kind of transmission identification sorting mechanism for Bottle ＆ Can recovery |
| CN110032983A (en) * | 2019-04-22 | 2019-07-19 | 扬州哈工科创机器人研究院有限公司 | A kind of track recognizing method based on ORB feature extraction and FLANN Rapid matching |
| CN110363254A (en) * | 2019-08-12 | 2019-10-22 | 河北工业大学 | Part category method for quickly identifying and identification device for crankshaft flexible production line |
| CN110415221A (en) * | 2019-07-12 | 2019-11-05 | 中南大学 | A kind of container truck based on Image Feature Point Matching is anti-to sling automatic testing method |
| EP3683722A1 (en) * | 2019-01-18 | 2020-07-22 | Alitheon, Inc. | Active identification and dimensioning |
| CN112012057A (en) * | 2020-08-12 | 2020-12-01 | 湖南长院悦诚装备有限公司 | Intelligent control method for ballastless track steel rail laying process |
-
2021
- 2021-01-04 CN CN202110003484.5A patent/CN112733694B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930253A (en) * | 2012-10-31 | 2013-02-13 | 中国矿业大学(北京) | Coal and rock identification method based on image discrete multi-wavelet transform |
| CN107377398A (en) * | 2017-07-21 | 2017-11-24 | 北京巴龙机电科技有限公司 | A kind of transmission identification sorting mechanism for Bottle ＆ Can recovery |
| EP3683722A1 (en) * | 2019-01-18 | 2020-07-22 | Alitheon, Inc. | Active identification and dimensioning |
| CN110032983A (en) * | 2019-04-22 | 2019-07-19 | 扬州哈工科创机器人研究院有限公司 | A kind of track recognizing method based on ORB feature extraction and FLANN Rapid matching |
| CN110415221A (en) * | 2019-07-12 | 2019-11-05 | 中南大学 | A kind of container truck based on Image Feature Point Matching is anti-to sling automatic testing method |
| CN110363254A (en) * | 2019-08-12 | 2019-10-22 | 河北工业大学 | Part category method for quickly identifying and identification device for crankshaft flexible production line |
| CN112012057A (en) * | 2020-08-12 | 2020-12-01 | 湖南长院悦诚装备有限公司 | Intelligent control method for ballastless track steel rail laying process |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112733694B (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106871805B (en) | vehicle-mounted rail gauge measuring system and measuring method | |
| CN111476767B (en) | High-speed rail fastener defect identification method based on heterogeneous image fusion | |
| CN109754368B (en) | Crack splicing method in bridge quality detection | |
| WO2023045299A1 (en) | Road surface technical condition detection method and device based on three-dimensional contour | |
| CN111666947B (en) | Pantograph head offset measuring method and system based on 3D imaging | |
| CN110378957B (en) | Torpedo tank car visual identification and positioning method and system for metallurgical operation | |
| CN105891220B (en) | A kind of pavement marker line defect detection device and its detection method | |
| CN110197157B (en) | Pavement crack growth detection method based on historical crack data | |
| CN109544607B (en) | Point cloud data registration method based on road sign line | |
| CN112461122B (en) | Tunnel surface feature detection device and method | |
| CN112082770A (en) | Rail vehicle body detection system and method based on artificial intelligence technology | |
| CN111768417A (en) | Railway wagon overrun detection method based on monocular vision 3D reconstruction technology | |
| CN113295094B (en) | Pantograph dynamic envelope intrusion detection method, device and system | |
| CN110490342B (en) | Contact net static geometrical parameter detection method based on Faster R-CNN | |
| CN105303564A (en) | Tower type crane load stereo pendulum angle vision detection method | |
| CN112733694B (en) | ORB feature-based track laying roller identification method and system | |
| CN113269729B (en) | Assembly body multi-view detection method and system based on depth image contrast | |
| CN104850869A (en) | Automatic identification system and method for high-speed railway CPIII marker | |
| CN111640155B (en) | Pantograph head inclination angle measurement method and system based on 3D imaging | |
| CN116681912A (en) | Rail gauge detection method and device for railway turnout | |
| CN114119355B (en) | Method and system for early warning of blocking dropping risk of shield tunnel | |
| CN112857252B (en) | Tunnel image boundary line detection method based on reflectivity intensity | |
| CN114511620A (en) | Mask R-CNN-based structure displacement monitoring method | |
| CN107316302A (en) | Driving automatic positioning equipment and its corresponding localization method based on binocular vision | |
| CN106845493A (en) | The identification at railroad track close-range image rail edge and matching process |
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 |