CN113838033A - Train wheel tread scratch image detection method and image detection system - Google Patents
Train wheel tread scratch image detection method and image detection system Download PDFInfo
- Publication number
- CN113838033A CN113838033A CN202111124462.0A CN202111124462A CN113838033A CN 113838033 A CN113838033 A CN 113838033A CN 202111124462 A CN202111124462 A CN 202111124462A CN 113838033 A CN113838033 A CN 113838033A
- Authority
- CN
- China
- Prior art keywords
- tread
- image
- scratch
- real
- defect
- 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
-
- 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
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/20081—Training; Learning
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses a method and a system for detecting an image of a scratch on a train wheel tread, wherein the method comprises the following steps: acquiring a tread image of a train wheel; finding out a suspected scratch area in the tread image by adopting a defect segmentation network; judging whether real scratches exist in the suspected scratch area by adopting a defect classification network; if there is a real scratch, outputting a tread surface image in which the real scratch exists. According to the method and the system for detecting the image of the scratch on the train wheel tread, provided by the invention, the acquired tread image is subjected to defect segmentation and defect classification by using the AI algorithm network, so that the real scratch can be accurately found out, the detection efficiency and the detection rate are greatly improved, the overhaul work of workers is facilitated, and the market popularization value is higher.
Description
Technical Field
The invention relates to the technical field of tread detection, in particular to a method and a system for detecting an image of a scratch on a train wheel tread.
Background
In recent years, the rail transit industry in China is rapidly developed, the construction of a rail transit network becomes a business card for urban development, rail transit gradually becomes the first choice for citizens to go out due to the characteristics of safety and quickness, and the problem of public traffic jam in each large city is effectively solved.
The wheel is an important part of train operation components, and the operation state of the wheel directly influences the operation speed and safety of the train. Specifically, the wheel includes a wheel rim and a wheel tread, and the wheel tread is a surface of the train contacting with the rail during the running process. Wheel treads of the wheels are often locally scratched due to braking or idle slipping and the like during train running, and the scratched wheels can cause coupling vibration of the whole vehicle track system during the train running process, so that the running safety is endangered. Therefore, it is very important and necessary to detect wheel tread scuff.
At the present stage, most of wheel tread detection still mainly uses manual visual inspection, the vehicle needs to be stopped in an overhaul warehouse, then maintainers are arranged to carry out tread detection on each wheel independently, time and labor are consumed, and almost half of wheels are shielded by a vehicle body, so that all wheels are difficult to detect. In addition, equipment for detecting scratches on the wheel tread by using images also exists in the market, but the algorithm adopted by the equipment has the defects of low detection rate, low repetition precision and the like, and stains, scratches and other interferences and real scratches on the wheel tread cannot be effectively distinguished, so that false alarm is easily caused, and the detection efficiency of workers is influenced.
Therefore, it is one of the problems that the skilled person needs to solve to improve the conventional wheel tread surface detection technology or to provide a new wheel tread surface detection technology.
The above information is given as background information only to aid in understanding the present disclosure, and no determination or admission is made as to whether any of the above is available as prior art against the present disclosure.
Disclosure of Invention
The invention provides an image detection method and an image detection system for a scratch on a train wheel tread, which aim to overcome the defects of the prior art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
in a first aspect, an embodiment of the present invention provides a method for detecting an image of a scratch on a tread of a train wheel, where the method includes:
acquiring a tread image of a train wheel;
finding out a suspected scratch area in the tread image by adopting a defect segmentation network;
judging whether real scratches exist in the suspected scratch area by adopting a defect classification network;
if there is a real scratch, outputting a tread surface image in which the real scratch exists.
Further, in the method for detecting an image of a tread surface scratch of a train wheel, the step of finding out a suspected scratch area in the tread surface image by using a defect segmentation network includes:
preprocessing the tread image and determining a tread detection area;
and finding out a suspected scratch area in the tread detection area by adopting a defect segmentation network.
Further, in the method for detecting an image of a scuffing on a tread surface of a train wheel, the method further includes:
and pre-establishing and training the defect segmentation network.
Further, in the method for detecting the image of the scratch on the tread of the train wheel, the step of pre-establishing and training the defect segmentation network comprises the following steps of;
pre-establishing the defect segmentation network;
collecting tread images for training;
marking a suspected scratch area in the training tread image;
and inputting the marked tread image for training to the defect segmentation network so as to train the defect segmentation network.
Further, in the method for detecting an image of a scuffing on a tread surface of a train wheel, the method further includes:
and establishing and training the defect classification network in advance.
Further, in the method for detecting the image of the scuffing on the tread of the train wheel, the step of pre-establishing and training the defect classification network comprises the following steps of;
pre-establishing the defect classification network;
collecting defect images for training;
marking real scratches in the defect images for training;
and inputting the labeled defect image for training to the defect classification network so as to train the defect classification network.
Further, in the method for detecting an image of a tread scratch of a train wheel, after the step of determining whether there is a real scratch in the suspected scratch area by using a defect classification network, the method further includes:
and if the real scratch does not exist, collecting the tread images without the real scratch as training tread images to train the defect segmentation network.
Further, in the method for detecting an image of a tread surface scuffing of a train wheel, after the step of outputting a tread surface image in which a real scuffing exists if the real scuffing exists, the method further includes:
collecting tread surface images with the real scratches as defect images for training to train the defect classification network.
Further, in the method for detecting an image of a tread surface scuffing of a train wheel, the step of outputting the tread surface image with the real scuffing includes:
locating the location of the real scratches and calculating the area of the real scratches;
outputting a tread surface image in which the real abrasion exists, a location of the real abrasion, and an area of the real abrasion.
In a second aspect, an embodiment of the present invention provides a train wheel tread scratch image detection system, including:
the image acquisition module is used for acquiring tread images of train wheels;
the suspected determining module is used for finding out a suspected scratch area in the tread image by adopting a defect segmentation network;
the scratch judging module is used for judging whether real scratches exist in the suspected scratch area by adopting a defect classification network;
and the image output module is used for outputting the tread image with the real scratch if the real scratch exists.
Further, in the train wheel tread scratch image detection system, the suspected determination module is specifically configured to:
preprocessing the tread image and determining a tread detection area;
and finding out a suspected scratch area in the tread detection area by adopting a defect segmentation network.
Further, in the system for detecting an image of a scuffing on a tread surface of a train wheel, the system further comprises a segmentation network establishing module for:
and pre-establishing and training the defect segmentation network.
Further, in the train wheel tread scratch image detection system, the segmentation network establishing module is specifically used for;
pre-establishing the defect segmentation network;
collecting tread images for training;
marking a suspected scratch area in the training tread image;
and inputting the marked tread image for training to the defect segmentation network so as to train the defect segmentation network.
Further, in the system for detecting an image of a scuffing on a tread surface of a train wheel, the system further comprises a classification network establishing module for:
and establishing and training the defect classification network in advance.
Further, in the train wheel tread scratch image detection system, the classification network establishing module is specifically used for;
pre-establishing the defect classification network;
collecting defect images for training;
marking real scratches in the defect images for training;
and inputting the labeled defect image for training to the defect classification network so as to train the defect classification network.
Further, in the system for detecting an image of a scratch on a tread of a train wheel, the system further comprises a segmentation network training module, configured to:
after the step of judging whether real scratches exist in the suspected scratch area by using the defect classification network, if real scratches do not exist, collecting tread images without the real scratches as training tread images to train the defect segmentation network.
Further, in the system for detecting an image of a scuffing on a tread surface of a train wheel, the system further comprises a classification network training module for:
and after the step of outputting the tread surface image with the real scratch if the real scratch exists, collecting the tread surface image with the real scratch as a defect image for training so as to train the defect classification network.
Further, in the image detection system for detecting the train wheel tread scratches, the image output module is specifically configured to:
locating the location of the real scratches and calculating the area of the real scratches;
outputting a tread surface image in which the real abrasion exists, a location of the real abrasion, and an area of the real abrasion.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
according to the method and the system for detecting the image of the scratch on the wheel tread of the train, provided by the embodiment of the invention, the acquired tread image is subjected to defect segmentation and defect classification by using an AI algorithm network, so that the real scratch can be accurately found out, the detection efficiency and the detection rate are greatly improved, the method and the system are beneficial to carrying out maintenance work by workers, and the market popularization value is high.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a schematic flow chart of a method for detecting an image of a scratch on a wheel tread of a train according to an embodiment of the present invention;
fig. 2 is a functional schematic diagram of a train wheel tread scratch image detection system according to a second embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it is to be understood that when an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. When a component is referred to as being "disposed on" another component, it can be directly on the other component or intervening components may also be present.
Furthermore, the terms "long", "short", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of describing the present invention, but do not indicate or imply that the referred devices or elements must have the specific orientations, be configured to operate in the specific orientations, and thus are not to be construed as limitations of the present invention.
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
Example one
In view of the defects of the existing wheel tread scratch detection technology, the inventor of the invention actively researches and innovates based on abundant practical experience and professional knowledge in many years of the industry and by matching with the application of theory, so as to create a feasible wheel tread scratch detection technology, and the technology has higher practicability. After continuous research, design and repeated trial and improvement, the invention with practical value is finally created.
Referring to fig. 1, fig. 1 is a schematic flow chart of a method for detecting an image of a scratch on a tread of a train wheel, which is disclosed in an embodiment of the present invention, and is applicable to a scene of abnormal detection on the tread of the train wheel. As shown in fig. 1, the method for detecting an image of a scuffing on a tread of a train wheel may include the steps of:
s101, collecting tread images of train wheels.
It should be noted that, in this embodiment, hardware support that is relied on for acquiring the tread image of the train wheel, for example, the camera is installed beside the rail, when the train passes through, the camera acquires a real-time picture of the train wheel, and then transmits the picture to the host computer, and then the host computer analyzes the tread image of the train wheel through a deployed algorithm.
And S102, finding out a suspected scratch area in the tread image by adopting a defect segmentation network.
In this embodiment, the step S102 may further include:
preprocessing the tread image and determining a tread detection area;
and finding out a suspected scratch area in the tread detection area by adopting a defect segmentation network.
In this embodiment, the defect segmentation network is trained in advance, so the method further includes the content of the defect segmentation network trained in advance, that is, the method further includes the following steps:
and pre-establishing and training the defect segmentation network.
Preferably, the step of pre-establishing and training the defect segmentation network may further include;
pre-establishing the defect segmentation network;
collecting tread images for training;
marking a suspected scratch area in the training tread image;
and inputting the marked tread image for training to the defect segmentation network so as to train the defect segmentation network.
S103, judging whether real scratches exist in the suspected scratch area by adopting a defect classification network; if yes, go to step S104, otherwise go to step S105 directly.
In this embodiment, the defect classification network is also trained in advance, so the method further includes the content of training the defect classification network in advance, that is, the method further includes the following steps:
and establishing and training the defect classification network in advance.
Preferably, the step of pre-establishing and training the defect classification network may further comprise;
pre-establishing the defect classification network;
collecting defect images for training;
marking real scratches in the defect images for training;
and inputting the labeled defect image for training to the defect classification network so as to train the defect classification network.
In this embodiment, a correction path exists for the case where a suspected scratch area can be found but there is no real scratch. Specifically, after step S103 and before step S105, the tread surface images that can find the suspected scratch area but do not have the real scratch are collected as the training tread surface images, and then the training tread surface images are used to train the defect segmentation network, so as to update the defect segmentation network, thereby increasing the accuracy.
And S104, outputting the tread surface image with the real scratch.
In this embodiment, the step S104 may further include:
locating the location of the real scratches and calculating the area of the real scratches;
outputting a tread surface image in which the real abrasion exists, a location of the real abrasion, and an area of the real abrasion.
It should be noted that, in this embodiment, after the tread surface image with the real scratch is detected, not only the tread surface image itself is output, but also the position of the real scratch on the image is located, and information such as the area and the length is calculated, so as to facilitate the maintenance of the worker.
In this embodiment, an enhanced path exists for the case where a suspected scratch area can be found and a real scratch exists. Specifically, after step S104 and before step S105, the tread surface images with suspected scratch areas and the real scratches are collected as training defect images, and the training defect images are used to train the defect classification network, so as to update the defect classification network, thereby increasing the accuracy.
And S105, ending.
According to the method for detecting the image of the scratch on the train wheel tread, provided by the embodiment of the invention, the acquired tread image is subjected to defect segmentation and defect classification by using the AI algorithm network, so that the real scratch can be accurately found out, the detection efficiency and the detection rate are greatly improved, the method is beneficial to carrying out maintenance work by workers, and the method has higher market popularization value.
Example two
Referring to fig. 2, fig. 2 is a functional module schematic diagram of a system for detecting an image of a scratch on a train wheel tread according to a second embodiment of the present invention, where the system is suitable for executing the method for detecting an image of a scratch on a train wheel tread according to the second embodiment of the present invention. The system specifically comprises the following modules:
the image acquisition module 201 is used for acquiring tread images of train wheels;
a suspected determining module 202, configured to find out a suspected scratch area in the tread surface image by using a defect segmentation network;
the scratch judging module 203 is configured to judge whether there is a real scratch in the suspected scratch area by using a defect classification network;
an image output module 204, configured to output a tread surface image with a real scratch if the real scratch exists.
Preferably, the suspected determination module 202 is specifically configured to:
preprocessing the tread image and determining a tread detection area;
and finding out a suspected scratch area in the tread detection area by adopting a defect segmentation network.
Preferably, the system further comprises a split network establishment module configured to:
and pre-establishing and training the defect segmentation network.
Preferably, the split network establishing module is specifically configured to establish a split network;
pre-establishing the defect segmentation network;
collecting tread images for training;
marking a suspected scratch area in the training tread image;
and inputting the marked tread image for training to the defect segmentation network so as to train the defect segmentation network.
Preferably, the system further comprises a classification network establishing module, configured to:
and establishing and training the defect classification network in advance.
Preferably, the classification network establishing module is specifically configured to establish a classification network;
pre-establishing the defect classification network;
collecting defect images for training;
marking real scratches in the defect images for training;
and inputting the labeled defect image for training to the defect classification network so as to train the defect classification network.
Preferably, the system further comprises a segmentation network training module, configured to:
after the step of judging whether real scratches exist in the suspected scratch area by using the defect classification network, if real scratches do not exist, collecting tread images without the real scratches as training tread images to train the defect segmentation network.
Preferably, the system further comprises a classification network training module, configured to:
and after the step of outputting the tread surface image with the real scratch if the real scratch exists, collecting the tread surface image with the real scratch as a defect image for training so as to train the defect classification network.
Preferably, the image output module 204 is specifically configured to:
locating the location of the real scratches and calculating the area of the real scratches;
outputting a tread surface image in which the real abrasion exists, a location of the real abrasion, and an area of the real abrasion.
According to the train wheel tread scratch image detection system provided by the embodiment of the invention, the acquired tread image is subjected to defect segmentation and defect classification by using the AI algorithm network, so that the real scratch can be accurately found out, the detection efficiency and the detection rate are greatly improved, the maintenance work of workers is facilitated, and the market popularization value is higher.
The system can execute the method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
The foregoing description of the embodiments has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the disclosure. Individual elements or features of a particular embodiment are generally not limited to that particular embodiment, but, where applicable, are interchangeable and can be used in a selected embodiment, even if not specifically shown or described. The same elements or features may also vary in many respects. Such variations are not to be regarded as a departure from the disclosure, and all such modifications are intended to be included within the scope of the disclosure.
Example embodiments are provided so that this disclosure will be thorough and will fully convey the scope to those skilled in the art. Numerous details are set forth, such as examples of specific parts, devices, and methods, in order to provide a thorough understanding of embodiments of the present disclosure. It will be apparent to those skilled in the art that specific details need not be employed, that example embodiments may be embodied in many different forms and that neither should be construed to limit the scope of the disclosure. In certain example embodiments, well-known processes, well-known device structures, and well-known technologies are not described in detail.
The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises" and "comprising" are intended to be inclusive and therefore specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring their performance in the particular order discussed and illustrated, unless explicitly indicated as an order of performance. It should also be understood that additional or alternative steps may be employed.
When an element or layer is referred to as being "on" … … "," engaged with "… …", "connected to" or "coupled to" another element or layer, it can be directly on, engaged with, connected to or coupled to the other element or layer, or intervening elements or layers may also be present. In contrast, when an element or layer is referred to as being "directly on … …," "directly engaged with … …," "directly connected to" or "directly coupled to" another element or layer, there may be no intervening elements or layers present. Other words used to describe the relationship of elements should be interpreted in a similar manner (e.g., "between … …" and "directly between … …", "adjacent" and "directly adjacent", etc.). As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items. Although the terms first, second, third, etc. may be used herein to describe various elements, components, regions, layers and/or sections, these elements, components, regions, layers and/or sections should not be limited by these terms. These terms may be only used to distinguish one element, component, region or section from another element, component, region or section. Unless clearly indicated by the context, use of terms such as the terms "first," "second," and other numerical values herein does not imply a sequence or order. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the teachings of the example embodiments.
Spatially relative terms, such as "inner," "outer," "below," "… …," "lower," "above," "upper," and the like, may be used herein for ease of description to describe a relationship between one element or feature and one or more other elements or features as illustrated in the figures. Spatially relative terms may be intended to encompass different orientations of the device in addition to the orientation depicted in the figures. For example, if the device in the figures is turned over, elements described as "below" or "beneath" other elements or features would then be oriented "above" the other elements or features. Thus, the example term "below … …" can encompass both an orientation of facing upward and downward. The device may be otherwise oriented (rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein interpreted.
Claims (10)
1. A method for detecting an image of a scuffing on a tread of a wheel of a train, the method comprising:
acquiring a tread image of a train wheel;
finding out a suspected scratch area in the tread image by adopting a defect segmentation network;
judging whether real scratches exist in the suspected scratch area by adopting a defect classification network;
if there is a real scratch, outputting a tread surface image in which the real scratch exists.
2. The method of detecting an image of a tread scuff on a train wheel according to claim 1 wherein the step of using a defect segmentation network to find a suspected scuff area in the tread image includes:
preprocessing the tread image and determining a tread detection area;
and finding out a suspected scratch area in the tread detection area by adopting a defect segmentation network.
3. The method of detecting an image of train wheel tread scuffing according to claim 1, further comprising:
and pre-establishing and training the defect segmentation network.
4. The method of detecting a train wheel tread scratch image according to claim 3, wherein said step of pre-establishing and training said defect segmentation network comprises;
pre-establishing the defect segmentation network;
collecting tread images for training;
marking a suspected scratch area in the training tread image;
and inputting the marked tread image for training to the defect segmentation network so as to train the defect segmentation network.
5. The method of detecting an image of train wheel tread scuffing according to claim 1, further comprising:
and establishing and training the defect classification network in advance.
6. The method of detecting images of train wheel tread scratches as in claim 5, wherein said step of pre-establishing and training said defect classification network comprises;
pre-establishing the defect classification network;
collecting defect images for training;
marking real scratches in the defect images for training;
and inputting the labeled defect image for training to the defect classification network so as to train the defect classification network.
7. The method of detecting an image of a train wheel tread scuff according to claim 1, wherein after the step of using a defect classification network to make a determination of whether there is a real scuff in the suspected scuff area, the method further comprises:
and if the real scratch does not exist, collecting the tread images without the real scratch as training tread images to train the defect segmentation network.
8. The method of detecting an image of a train wheel tread scuff according to claim 1, wherein after the step of outputting a tread image with a true scuff if present, the method further comprises:
collecting tread surface images with the real scratches as defect images for training to train the defect classification network.
9. The method of detecting an image of a train wheel tread scuff according to claim 1, wherein the step of outputting the tread image with the true scuff present includes:
locating the location of the real scratches and calculating the area of the real scratches;
outputting a tread surface image in which the real abrasion exists, a location of the real abrasion, and an area of the real abrasion.
10. An image detection system for detecting scratches on a tread of a train wheel, the system comprising:
the image acquisition module is used for acquiring tread images of train wheels;
the suspected determining module is used for finding out a suspected scratch area in the tread image by adopting a defect segmentation network;
the scratch judging module is used for judging whether real scratches exist in the suspected scratch area by adopting a defect classification network;
and the image output module is used for outputting the tread image with the real scratch if the real scratch exists.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111124462.0A CN113838033A (en) | 2021-09-24 | 2021-09-24 | Train wheel tread scratch image detection method and image detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111124462.0A CN113838033A (en) | 2021-09-24 | 2021-09-24 | Train wheel tread scratch image detection method and image detection system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113838033A true CN113838033A (en) | 2021-12-24 |
Family
ID=78969967
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111124462.0A Pending CN113838033A (en) | 2021-09-24 | 2021-09-24 | Train wheel tread scratch image detection method and image detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113838033A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114572273A (en) * | 2022-03-15 | 2022-06-03 | 南京拓控信息科技股份有限公司 | 3D image detection method for wheel set tread of railway vehicle |
-
2021
- 2021-09-24 CN CN202111124462.0A patent/CN113838033A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114572273A (en) * | 2022-03-15 | 2022-06-03 | 南京拓控信息科技股份有限公司 | 3D image detection method for wheel set tread of railway vehicle |
| CN114572273B (en) * | 2022-03-15 | 2024-04-30 | 南京拓控信息科技股份有限公司 | Railway vehicle wheel set tread 3D image detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Automatic defect detection of metro tunnel surfaces using a vision-based inspection system | |
| Nayyeri et al. | Foreground–background separation technique for crack detection | |
| Vittorio et al. | Automated sensing system for monitoring of road surface quality by mobile devices | |
| CN103512762B (en) | Image processing method, device and train failure detection system | |
| Jahanshahi et al. | Unsupervised approach for autonomous pavement-defect detection and quantification using an inexpensive depth sensor | |
| CN101893580B (en) | Digital image based detection method of surface flaw of steel rail | |
| CN108280855A (en) | A kind of insulator breakdown detection method based on Fast R-CNN | |
| Lekshmipathy et al. | Vibration vs. vision: Best approach for automated pavement distress detection | |
| Shaghouri et al. | Real-time pothole detection using deep learning | |
| Wang et al. | A road quality detection method based on the mahalanobis-taguchi system | |
| CN103837087B (en) | Pantograph automatic testing method based on active shape model | |
| CN104608799A (en) | Information fusion technology based train wheel set tread damage online detection and recognition method | |
| CN104966049B (en) | Lorry detection method based on image | |
| Hadjidemetriou et al. | Vision-and entropy-based detection of distressed areas for integrated pavement condition assessment | |
| CN111899288A (en) | Tunnel leakage water area detection and identification method based on infrared and visible light image fusion | |
| CN107967681B (en) | Elevator compensation chain impact defect detection method based on machine vision | |
| CN111311567A (en) | Method for identifying fastener and steel rail diseases of track line image | |
| CN107273802A (en) | A kind of detection method and device of railroad train brake shoe drill ring failure | |
| CN113838033A (en) | Train wheel tread scratch image detection method and image detection system | |
| Doycheva et al. | Computer vision and deep learning for real-time pavement distress detection | |
| CN106951820A (en) | Passenger flow statistical method based on annular template and ellipse fitting | |
| CN112198170A (en) | Detection method for identifying water drops in three-dimensional detection of outer surface of seamless steel pipe | |
| Aggarwal et al. | Road crack detection and segmentation for autonomous driving | |
| CN105279488A (en) | Mileage signboard automatic recognition method for road video routing inspection | |
| Pundir et al. | POCONET: A Pathway to Safety |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: No. 10, Wansheng Road, Wanjiang Street, Dongguan City, Guangdong Province, 523039 Applicant after: Dongguan Nuoli Technology Co.,Ltd. Address before: 3 / F, building 10, Wanhong village, Wanjiang community, Wanjiang District, Dongguan City, Guangdong Province, 523039 Applicant before: DONGGUAN NANNAR ELECTRONICS TECHNOLOGY Co.,Ltd. |