CN112590868A - System for detecting abrasion through shooting track by machine vision technology - Google Patents
System for detecting abrasion through shooting track by machine vision technology Download PDFInfo
- Publication number
- CN112590868A CN112590868A CN202011558254.7A CN202011558254A CN112590868A CN 112590868 A CN112590868 A CN 112590868A CN 202011558254 A CN202011558254 A CN 202011558254A CN 112590868 A CN112590868 A CN 112590868A
- Authority
- CN
- China
- Prior art keywords
- module
- image
- processing module
- rail
- detecting
- 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.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
- B61L23/00—Control, warning, or like safety means along the route or between vehicles or vehicle trains
- B61L23/04—Control, warning, or like safety means along the route or between vehicles or vehicle trains for monitoring the mechanical state of the route
- B61L23/042—Track changes detection
- B61L23/045—Rail wear
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/048—Fuzzy inferencing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30136—Metal
Abstract
The invention discloses a system for detecting abrasion of a shooting track through a machine vision technology, which comprises an image acquisition module, a photoelectric coding module, an image processing module, an image detection module, a central processing module and a database server, wherein the image acquisition module is respectively connected with the photoelectric coding module and the image processing module, the image detection module is respectively connected with the image processing module, the photoelectric coding module and the central processing module, and the central processing module is connected with the database module. According to the invention, the multi-angle cameras are arranged, two groups of different image shooting are arranged from the inside and the outside of the track, the cross section lines of the whole rail are formed according to the extracted laser line combination, and compared with the traditional inner layer single-side lines, the combined whole cross section lines have higher referential property, so that the accuracy in subsequent detection and comparison is improved.
Description
Technical Field
The invention relates to the field of mechanical vision detection, in particular to a system for detecting abrasion by shooting a track through a machine vision technology.
Background
The railway line equipment is basic equipment of railway transportation industry, and is exposed in nature throughout the year, and is subjected to the action of wind, rain, freezing and thawing and train load, the geometric dimension of the track is continuously changed, the roadbed and the track bed are continuously deformed, and the steel rail, the connecting parts and the sleeper are continuously worn, so that the technical state of the line equipment is continuously changed, therefore, a work department grasps the change rule of the line equipment, detects the state of the line in time, strengthens the line detection management and becomes important basic work for ensuring the quality of the line and the transportation safety, wherein the static inspection is generally common static inspection, and the static inspection refers to the inspection of the line by using a manual or light measuring trolley when no wheel load acts. The method mainly comprises the inspection of gauge, level, front and back height, direction, empty hanging plates, steel rail joints, anti-climbing equipment, coupling parts, sleepers, crossing equipment and the like, along with the development of computer technology and image processing technology, the railway track detection is gradually developed towards the direction of high efficiency and intellectualization, and the vision measurement technology used at present also appears.
In the prior art, laser irradiation is generally adopted on the inner side of a rail, after image shooting is carried out, sectional linear information of the rail is extracted in an improved image processing mode, and the wear degree is obtained according to comparison between the linear information and the normal rail sectional linear information.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a system for detecting wear by shooting a track through a machine vision technology.
In order to solve the technical problems, the invention provides the following technical scheme:
the invention relates to a system for detecting abrasion of a shooting track through a machine vision technology, which comprises an image acquisition module, a photoelectric coding module, an image processing module, an image detection module, a central processing module and a database server, wherein the image acquisition module, the photoelectric coding module and the image processing module are all integrated equipment;
the image acquisition module comprises a CCD industrial camera, an exposure lamp, a track detection vehicle and a first wireless transmission module;
the photoelectric coding module comprises a photoelectric encoder, a singlechip chip and a laser sensor module;
the image processing module comprises a preprocessor module and a second wireless transmission module;
the image detection module comprises a second processor module, a first input module and a third wireless transmission module;
the central processor module comprises a main processor module, a second input module, a cache module and a signal transmission module;
the database server comprises an information processor module and an information storage module.
As a preferred technical solution of the present invention, the number of the CCD industrial cameras is four, and two of the CCD industrial cameras are grouped, and are disposed at both ends of a single track, and are symmetrically disposed, the laser sensor module includes laser emitters, the number of the laser sensors is the same as that of the CCD industrial cameras, and the laser emitting points are disposed at the center of the shooting range of the CCD industrial cameras.
As a preferable technical scheme of the invention, the exposure lamp is composed of a PCB and an LED array lamp and is mainly used for exposure of a shooting part of a CCD industrial camera.
As a preferred technical solution of the present invention, the photoelectric encoder is disposed on a wheel axle of the rail detection vehicle, and generates a pulse signal through rotation, so as to output the pulse signal to the image detection module, and assist the image detection module in calculating the relative distance and the vehicle speed of the rail detection vehicle.
As a preferred technical solution of the present invention, the preprocessor module of the image processing module is loaded with an image preprocessing module, and according to the picture taken by the image acquisition module, the preprocessor module performs denoising and filtering on the image, calculates an image histogram, and highlights stripe characteristics of laser; the image detection module is one of an all-in-one mobile computer, a tablet personal computer or a notebook computer, is mainly provided with an image main processing module, connects pixel discontinuous points according to an expansion area, then binarizes the image, extracts main characteristic points according to a Hessian matrix, performs characteristic point matching by adopting a minimum Euclidean distance criterion, and eliminates mismatching points by adopting an RANSAC algorithm.
As a preferred technical solution of the present invention, the central processing unit is equipped with an image association module for associating an original image with a processed image, forming an entry, and storing the entry in the database server, wherein the entry includes an image forming time, a place, a rail wear status obtained by the image forming time, the place, and a manual remark item; the method comprises the steps that an artificial intelligence algorithm is carried by the database server, a Bayes classifier is combined with fuzzy rule reasoning to classify flaws of processed images, a flaw segmentation method based on the variation degree and the local entropy is adopted, the variation degree and the local entropy of rail images are firstly calculated, worn area information and class edges are extracted, then a flaw area in an original rail image is segmented by using a Pulse Coupled Neural Network (PCNN) and an area growing method, and finally a comparison test is carried out on the flaw area and the flaw images in the database.
Compared with the prior art, the invention has the following beneficial effects:
1: according to the invention, the multi-angle cameras are arranged, two groups of different image shooting are arranged from the inside and the outside of the track, the cross section lines of the whole rail are formed according to the extracted laser line combination, and compared with the traditional inner layer single-side lines, the combined whole cross section lines have higher referential property, so that the accuracy in subsequent detection and comparison is improved.
2: the invention is provided with the database server, carries out artificial intelligence comparison according to the information stored by the database server, compares the stored rail picture of the abrasion condition with the current shot picture, carries out omnibearing comparison on the past rail abrasion working condition and the current shot picture according to the environmental condition information, finds out the most similar information for auxiliary judgment, increases the information source of the judgment and leads the judged result to be more accurate.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic view of the integral module connection structure of the present invention;
FIG. 2 is a schematic view of a module assembly of the present invention;
FIG. 3 is a schematic flow diagram of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Example 1
As shown in fig. 1-3, the present invention provides a system for detecting wear of a track photographed by a machine vision technique, comprising an image acquisition module, a photoelectric coding module, an image processing module, an image detection module, a central processing module and a database server, wherein the image acquisition module, the photoelectric coding module and the image processing module are all integrated devices, the image detection module is a movable device, the central processing module and the database server are fixed devices, the image acquisition module is respectively connected with the photoelectric coding module and the image processing module, the image detection module is respectively connected with the image processing module, the photoelectric coding module and the central processing module, and the central processing module is connected with the database module;
the image acquisition module comprises a CCD industrial camera, an exposure lamp, a track detection vehicle and a first wireless transmission module;
the photoelectric coding module comprises a photoelectric coder, a singlechip chip and a laser sensor module;
the image processing module comprises a preprocessor module and a second wireless transmission module;
the image detection module comprises a second processor module, a first input module and a third wireless transmission module;
the central processor module comprises a main processor module, a second input module, a cache module and a signal transmission module;
the database server comprises an information processor module and an information storage module.
Further, the number of CCD industry camera is four, and two liang of groups all set up in single orbital both ends, and for the symmetry setting, and the laser sensor module includes laser emitter, and its laser sensor is the same with the number of CCD industry camera, and the laser emission point all sets up in the shooting scope center department of CCD industry camera.
The exposure lamp is composed of a PCB board and an LED array lamp and is mainly used for exposure of a CCD industrial camera shooting part.
The photoelectric encoder is arranged on a wheel shaft of the track detection vehicle, generates a pulse signal through rotation, is used for outputting the signal to the image detection module, and assists the image detection module to calculate the relative distance and the vehicle speed of the track detection vehicle.
A preprocessor module of the image processing module is provided with an image preprocessing module, and is used for denoising and filtering the image, calculating an image histogram and highlighting the stripe characteristics of the laser according to the picture shot by the image acquisition module;
the image detection module is one of an all-in-one mobile computer, a tablet personal computer or a notebook computer, is mainly provided with an image main processing module, connects pixel discontinuous points according to an expansion area, then binarizes the picture, extracts main characteristic points according to a Hessian matrix, performs characteristic point matching by adopting a minimum Euclidean distance criterion, and eliminates mismatching points by adopting an RANSAC algorithm.
The central processing unit module is provided with an image association module for associating the original image with the processed image and forming an item, wherein the item comprises image forming time, image forming place, the obtained rail wear state and a manual remark item, and is finally stored in the database server;
the method comprises the steps that an artificial intelligence algorithm is carried by a database server, a Bayes classifier is combined with fuzzy rule reasoning to classify flaws of processed images, a flaw segmentation method based on the variation degree and the local entropy is adopted, the variation degree and the local entropy of rail images are firstly calculated, worn area information and class edges are extracted, then a flaw area in an original rail image is segmented by using a Pulse Coupled Neural Network (PCNN) and a region growing method, and finally a comparison test is carried out on the flaw area and the flaw images in the database.
Specifically, the rails are generally arranged on two sides, so that when the rail detection vehicle of the image acquisition module detects one of the rails, a CCD industrial camera and a laser emitter with the same angle are carried, wherein the two sides are oppositely arranged, and the abrasion value of the rail cannot be changed greatly within a short distance, so that data can be sampled at regular intervals, each shooting of the CCD industrial camera is synchronous action, so that the laser ray emitted by the laser emitter forms a specific linear profile on the rail and is captured into a picture by the CCD industrial camera, the light source is supplemented by an exposure lamp in the shooting process, and the lens aperture of the CCD industrial camera is adjusted to be in a gray-black state with the background color;
the captured image and information of a photoelectric encoder are transmitted to an image processing module through a first wireless transmission module and a second wireless transmission module, a preprocessor module of the image processing module firstly processes the information of the photoelectric encoder, detects the travel information of a vehicle according to a calculated track, then preliminarily processes the image shot by a CCD industrial camera, records pixels of an over-exposure area, filters the image, confirms the characteristic points of red laser pixels needing to be extracted through a Hessian matrix, forms a vector normal space by using a Haar wavelet response function, calculates the sub-pixel coordinates of the center of a light strip to form a preprocessed shot image, the preprocessed shot image is shot images on two sides, a user can shoot the image according to the preprocessing on the inner side at the moment, operates an image processing module device to directly run an image extraction algorithm on site, and directly adopts a light strip center point sub-pixel coordinate extraction mode, based on a Hessian matrix binary function, the image is subjected to convolution operation in an expansion mode, after extraction is finished, a worn part can be found out on site according to standard steel rail image comparison, the wear degree can be checked on site, and on-site identification is facilitated;
transmitting the two pre-processed shot images and the two original shot images to a central processor module through a third wireless transmission module, storing related information according to the established items by the central processor module according to shot time, input positioning information, working condition information and manual remark items, then performing median filtering, interpolation and extraction processing according to the steps according to a picture splicing algorithm, splicing red light pixel strips of the two pre-processed shot images shot on site to form a steel rail with an integral section, so that the collected images form a complete line image, inputting a final conclusion into the items after manual judgment and final conclusion obtaining, and storing the final conclusion into a database server;
the database server is loaded with an artificial intelligence algorithm, adopts a multithreading technology, a python language and a Matlab algorithm library, classifies the wear information of different steel rails according to severity after receiving various samples, extracts characteristic values in the samples according to a line drawing, and trains an ANN model by using the characteristic values; a flaw segmentation method based on the degree of variation and the local entropy includes calculating the degree of variation and the local entropy of an original rail image, comparing the original rail image with a red light pixel line graph, extracting flaw area information and similar edges, segmenting flaw areas in the rail image by using a neural network (PCNN) based on pulse coupling and a region growing method, finally performing comparison test with flaw images in a database, after training is completed, after a central processing unit module receives a new rail shooting image, a database server firstly searches similar red light pixel line graphs in stored entries and after the original shooting image is subjected to sheet-shaped flaw wear area interception, comparing original shooting graphs of the database bound with the red light pixel line graphs of the database, extracting all entry information of the original shooting graphs of the most similar database, and performing image segmentation according to positioning information, working condition information and the like, And the manual remark items and the conclusions, the database red light pixel line graph and the database original shot graph are fed back to the central processor module for assisting manual judgment, and if the central processor module receives a new steel rail shot graph which is completely the same as the data graph stored in the database server, the conclusion can be directly output, so that the abrasion information conclusion of the steel rail is output more accurately and rapidly, and the repeated work in the work content is reduced.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. A system for detecting abrasion of a shooting track through a machine vision technology comprises an image acquisition module, a photoelectric coding module, an image processing module, an image detection module, a central processing module and a database server, and is characterized in that the image acquisition module, the photoelectric coding module and the image processing module are all integrated equipment, the image detection module is movable equipment, the central processing module and the database server are fixed equipment, the image acquisition module is respectively connected with the photoelectric coding module and the image processing module, the image detection module is respectively connected with the image processing module, the photoelectric coding module and the central processing module, and the central processing module is connected with the database module;
the image acquisition module comprises a CCD industrial camera, an exposure lamp, a track detection vehicle and a first wireless transmission module;
the photoelectric coding module comprises a photoelectric encoder, a singlechip chip and a laser sensor module;
the image processing module comprises a preprocessor module and a second wireless transmission module;
the image detection module comprises a second processor module, a first input module and a third wireless transmission module;
the central processor module comprises a main processor module, a second input module, a cache module and a signal transmission module;
the database server comprises an information processor module and an information storage module.
2. The system for detecting the abrasion of the shooting track through the machine vision technology as claimed in claim 1, wherein the number of the CCD industrial cameras is four, two of the CCD industrial cameras are grouped into two groups, the two groups are arranged at two ends of a single track and symmetrically arranged, the laser sensor module comprises laser emitters, the number of the laser sensors of the laser emitters is the same as that of the CCD industrial cameras, and laser emitting points are arranged at the center of the shooting range of the CCD industrial cameras.
3. A system for detecting abrasion of a photographing rail by a machine vision technology as claimed in claim 1, wherein the exposure lamp is composed of a PCB board and an LED array lamp, and is mainly used for exposure of a photographing part of a CCD industrial camera.
4. The system of claim 1, wherein the photoelectric encoder is disposed on a wheel shaft of the rail detecting vehicle, and generates a pulse signal through rotation for outputting the signal to the image detecting module to assist the image detecting module in calculating the relative distance and speed of the rail detecting vehicle.
5. The system for detecting the abrasion of the shooting track through the machine vision technology as claimed in claim 1, wherein the preprocessor module of the image processing module is loaded with an image preprocessing module, and is used for denoising and filtering the image, calculating an image histogram and highlighting the stripe feature of the laser according to the picture shot by the image acquisition module;
the image detection module is one of an all-in-one mobile computer, a tablet personal computer or a notebook computer, is provided with an image main processing module, connects pixel discontinuous points according to an expansion area, binarizes the picture, extracts main characteristic points according to a Hessian matrix, performs characteristic point matching by adopting a minimum Euclidean distance criterion, and eliminates mismatching points by adopting an RANSAC algorithm.
6. A system for detecting wear of a track by machine vision technology photographing as claimed in claim 1, wherein the central processor module is equipped with an image association module for associating the original image with the processed image and forming an entry, the entry comprising the image forming time, the image forming location, the obtained rail wear status and a manual remark item, and finally storing the entry in the database server;
the method comprises the steps that an artificial intelligence algorithm is carried by the database server, a Bayes classifier is combined with fuzzy rule reasoning to classify flaws of processed images, a flaw segmentation method based on the variation degree and the local entropy is adopted, the variation degree and the local entropy of rail images are firstly calculated, worn area information and class edges are extracted, then a flaw area in an original rail image is segmented by using a Pulse Coupled Neural Network (PCNN) and an area growing method, and finally a comparison test is carried out on the flaw area and the flaw images in the database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011558254.7A CN112590868A (en) | 2020-12-24 | 2020-12-24 | System for detecting abrasion through shooting track by machine vision technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011558254.7A CN112590868A (en) | 2020-12-24 | 2020-12-24 | System for detecting abrasion through shooting track by machine vision technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112590868A true CN112590868A (en) | 2021-04-02 |
Family
ID=75202526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011558254.7A Pending CN112590868A (en) | 2020-12-24 | 2020-12-24 | System for detecting abrasion through shooting track by machine vision technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112590868A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114670899A (en) * | 2022-04-20 | 2022-06-28 | 北京运达华开科技有限公司 | Image acquisition device for track detection system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101833484B1 (en) * | 2016-11-28 | 2018-02-28 | (주)익스트리플 | A device for inspecting rail for railway |
| CN108009484A (en) * | 2017-11-28 | 2018-05-08 | 西南交通大学 | Intelligent striation acquisition system and its method based on machine vision technique |
| CN108248634A (en) * | 2018-01-24 | 2018-07-06 | 西南交通大学 | Track switch and track full-sized abrasion vision measurer and its measuring method |
| CN108986082A (en) * | 2018-06-28 | 2018-12-11 | 武汉理工大学 | A kind of profile of steel rail detection method and system based on EPNP |
| CN110047070A (en) * | 2019-04-22 | 2019-07-23 | 山东师范大学 | A kind of recognition methods and system of path wear degree |
-
2020
- 2020-12-24 CN CN202011558254.7A patent/CN112590868A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101833484B1 (en) * | 2016-11-28 | 2018-02-28 | (주)익스트리플 | A device for inspecting rail for railway |
| CN108009484A (en) * | 2017-11-28 | 2018-05-08 | 西南交通大学 | Intelligent striation acquisition system and its method based on machine vision technique |
| CN108248634A (en) * | 2018-01-24 | 2018-07-06 | 西南交通大学 | Track switch and track full-sized abrasion vision measurer and its measuring method |
| CN108986082A (en) * | 2018-06-28 | 2018-12-11 | 武汉理工大学 | A kind of profile of steel rail detection method and system based on EPNP |
| CN110047070A (en) * | 2019-04-22 | 2019-07-23 | 山东师范大学 | A kind of recognition methods and system of path wear degree |
Non-Patent Citations (1)
| Title |
|---|
| 金侠挺等: "基于贝叶斯CNN和注意力网络的钢轨表面缺陷检测系统", 《自动化学报》, vol. 45, no. 12, 31 December 2019 (2019-12-31) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114670899A (en) * | 2022-04-20 | 2022-06-28 | 北京运达华开科技有限公司 | Image acquisition device for track detection system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11551341B2 (en) | Method and device for automatically drawing structural cracks and precisely measuring widths thereof | |
| CN111661590B (en) | Method for detecting tearing damage of conveying belt of mining belt conveyor | |
| CN111899288A (en) | Tunnel leakage water area detection and identification method based on infrared and visible light image fusion | |
| CN110490842B (en) | Strip steel surface defect detection method based on deep learning | |
| CN109489724A (en) | A kind of tunnel safe train operation environment comprehensive detection device and detection method | |
| CN109506628A (en) | Object distance measuring method under a kind of truck environment based on deep learning | |
| CN109559324A (en) | A kind of objective contour detection method in linear array images | |
| CN111768417B (en) | Railway wagon overrun detection method based on monocular vision 3D reconstruction technology | |
| CN113436157A (en) | Vehicle-mounted image identification method for pantograph fault | |
| Tsintotas et al. | DOSeqSLAM: Dynamic on-line sequence based loop closure detection algorithm for SLAM | |
| CN111942434A (en) | Intelligent fault image detection device for key parts of railway wagon | |
| Zhao et al. | Image-based comprehensive maintenance and inspection method for bridges using deep learning | |
| CN113112501A (en) | Vehicle-mounted track inspection device and method based on deep learning | |
| CN116309277A (en) | Steel detection method and system based on deep learning | |
| CN112590868A (en) | System for detecting abrasion through shooting track by machine vision technology | |
| AlNaimi et al. | IoT based on-the-fly visual defect detection in railway tracks | |
| CN111127381B (en) | Non-parallel detection method for pantograph slide plate | |
| WO2024037408A1 (en) | Underground coal mine pedestrian detection method based on image fusion and feature enhancement | |
| CN113469974A (en) | Method and system for monitoring grate plate state of pellet chain grate | |
| CN113076959A (en) | Concrete structure surface microcrack feature extraction method based on convolutional neural network | |
| CN109993741B (en) | Steel rail welding seam contour automatic positioning method based on K-means clustering | |
| CN111091553A (en) | Method for detecting loss of blocking key | |
| Gao et al. | Intelligent crack damage detection system in shield tunnel using combination of retinanet and optimal adaptive selection | |
| Saini et al. | Feature-based template matching for joggled fishplate detection in railroad track with drone images | |
| CN116539723A (en) | Railway track damage detection method and system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210402 |
|
| RJ01 | Rejection of invention patent application after publication |