CN111080600A - Fault identification method for split pin on spring supporting plate of railway wagon - Google Patents
Fault identification method for split pin on spring supporting plate of railway wagon Download PDFInfo
- Publication number
- CN111080600A CN111080600A CN201911272261.8A CN201911272261A CN111080600A CN 111080600 A CN111080600 A CN 111080600A CN 201911272261 A CN201911272261 A CN 201911272261A CN 111080600 A CN111080600 A CN 111080600A
- Authority
- CN
- China
- Prior art keywords
- cotter pin
- image
- images
- fault
- lost
- 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
- 238000000034 method Methods 0.000 title claims abstract description 17
- 230000011218 segmentation Effects 0.000 claims abstract description 41
- 238000012549 training Methods 0.000 claims abstract description 22
- 239000000523 sample Substances 0.000 claims description 34
- 238000012360 testing method Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 230000007423 decrease Effects 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 239000012468 concentrated sample Substances 0.000 claims description 2
- 230000002708 enhancing effect Effects 0.000 claims description 2
- 238000005192 partition Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 11
- 230000003247 decreasing effect Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 208000003464 asthenopia Diseases 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
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
- 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
- G06T2207/20—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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
A fault identification method for a split pin on a spring supporting plate of a railway wagon solves the problem that missing detection is easily caused when the split pin is lost and fault detection is carried out in a mode of manually observing images, and belongs to the field of fault identification of railway wagons. The invention comprises the following steps: acquiring images of two sides of a railway wagon, and constructing a sample set; inputting the images in the sample set into a U-Net segmentation network for training to obtain a segmentation model; the U-Net segmentation network reduces two convolutions and downsampling of a first layer and two convolutions and upsampling of a last layer on the basis of the existing U-Net network, and changes the number of feature maps of each layer into half of the original number; inputting a railway wagon image to be detected into a segmentation model, and acquiring a segmentation area of the cotter pin; and analyzing the fault of the divided area, judging whether the cotter pin is lost, if so, judging that the cotter pin is lost, and alarming the lost cotter pin.
Description
Technical Field
The invention relates to a fault identification method, in particular to a fault identification method for loss of a cotter pin on a spring supporting plate of a railway wagon, and belongs to the field of fault identification of railway wagons.
Background
If the open pin loss fault of the upper folding head bolt of the spring supporting plate of the truck is not discovered in time, hidden danger can be buried in the lost nut, and then the driving safety is endangered. In the existing cotter loss fault detection, the cotter loss fault detection is carried out in a mode of manually observing images, the cotter target is small, the color difference with other parts is small, visual fatigue is easily caused to naked eye detection, the eyesight of workers is influenced, the appearance of missing detection is caused, and the driving safety is influenced.
Disclosure of Invention
The invention provides a fault identification method for automatically identifying a cotter pin on a spring supporting plate of a railway wagon, aiming at the problem that missing detection is easily caused when the cotter pin loss fault detection is carried out in the conventional mode of manually observing images.
The invention discloses a fault identification method for a cotter pin on a spring supporting plate of a railway wagon, which comprises the following steps of:
s1, collecting images of two sides of the railway wagon, and constructing a sample set;
the concentrated sample image is a subimage obtained by cutting out a cotter pin from the wagon image, and the cotter pin is marked in the subimage and is a cotter pin of a dog-ear bolt on the spring supporting plate;
s2, inputting the images in the sample set into a U-Net segmentation network for training to obtain a segmentation model;
the U-Net segmentation network reduces two convolutions and downsampling of a first layer and two convolutions and upsampling of a last layer on the basis of the existing U-Net network, and changes the number of feature maps of each layer into half of the original number;
s3, inputting the railway wagon image to be detected into the segmentation model, and acquiring the segmentation area of the cotter pin;
and S4, carrying out fault analysis on the divided areas, judging whether the cotter pin is lost or not, if so, judging that the cotter pin is lost, and alarming the lost cotter pin.
Preferably, the S1 includes:
s11, acquiring images of the rail wagon;
s12, intercepting the image of the railway wagon by using the frame to obtain a sub-image comprising the cotter pin and the position identification component, marking the cotter pin and the position identification component in the sub-image, and constructing a sample set by using the marked sub-image, wherein the position identification component is a position relation required by the cotter pin fault judgment;
the S3 is as follows: intercepting a railway wagon image to be detected by using a frame, intercepting a subimage comprising a cotter pin, inputting the subimage into a segmentation model, and acquiring segmentation areas of the cotter pin and a position identification component;
and S4, judging whether the cotter pin is lost or not according to the position relation between the cotter pin and the position identification component in the partition area, if so, judging that the cotter pin is lost, and alarming the lost cotter pin.
Preferably, in S2, the learning rate during training is decreased as the number of training steps increases, and the training is stopped if the loss of the sample set continues for a plurality of iterations and does not decrease any more.
Preferably, the S2 further includes:
and (3) carrying out fault recognition on the new sample by using the obtained segmentation model, if the new sample has poor recognition effect, selecting and marking the new sample, putting the new sample into a training sample, reading the segmentation model to train continuously, and continuously testing and updating the weight until the result is satisfied, thereby completely training the segmentation model.
Preferably, the wagon images acquired at S1 include various images having different cotter shapes and different image gradations.
Preferably, in S12, the method further includes enhancing, flipping, and translating the sub-image with the failed cotter pin, so as to increase the number of sub-images in the sample set.
Preferably, the S11 further includes a gray scale value of the acquired image of the railway wagon, and if the gray scale value is lower than a set threshold, the image is subjected to adaptive histogram equalization to improve the contrast.
The invention has the advantages that the invention applies deep learning to fault detection, replaces the traditional human eye to identify faults, has more efficient operation and saves cost. According to the invention, for the open pin loss fault, the fault recognition is carried out by using a deep learning algorithm, so that the problems of external interference, image gray level difference and variable target forms can be effectively solved, and the missing detection is avoided. The invention has the advantages of small target, model compression and network acceleration.
Drawings
FIG. 1 is a schematic diagram of a U-Net segmentation network of the present invention;
FIG. 2 is a schematic flow chart of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments 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.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
As shown in fig. 2, the method for identifying a failure of a cotter pin in a spring retainer plate of a railway wagon according to the present embodiment includes:
s1, building a high-speed imaging system at a fixed detection station, and when a truck passes through the station, the imaging system works to obtain complete images of two sides of the truck and construct a sample set;
in the embodiment, the image in the sample set is a sub-image obtained by cutting out a cotter pin from the railway wagon image, and the cotter pin is marked in the sub-image and is a cotter pin of a folding bolt on a spring supporting plate;
s2, inputting the images in the sample set into a U-Net segmentation network for training to obtain a segmentation model;
as shown in fig. 1, the U-Net segmentation network is to reduce two convolutions and downsamplings of the first layer and two convolutions and upsamplings of the last layer on the basis of the existing U-Net network, and change the number of feature maps of each layer to half of the original number;
due to the fact that the fault target of the cotter pin is small, the target interception subgraph is also small, the last layer of the network is reduced, down-sampling and up-sampling are reduced, the number of feature graphs of each layer of the U-Net is reduced to be half of the original number, and the running speed of the network is further accelerated. Truck fault online identification has high requirements for identification speed, but the same part image is many, and much acceleration of model identification is necessary.
In the embodiment, the trained segmentation model is added into the subsequent image processing operation, and after the algorithm is completed, the segmentation model is uploaded to the online waiting for vehicle passing. When a wagon passes through the base station, the camera collects images to obtain railway wagon images to be detected;
s3, inputting the railway wagon image to be detected into the segmentation model, and acquiring the segmentation area of the cotter pin;
s4, analyzing the fault of the divided area, judging whether the cotter pin is lost, if so, judging that the cotter pin is lost,
and a program calling fault identification module is used for judging the loss fault of the split pin of the folded bolt of the spring supporting plate. If the fault is identified, fault information is output to the platform for manual confirmation and processing. If no fault is identified, the program is closed without alarming. And continuing waiting for the arrival of the next truck, acquiring the next image to be detected, and switching to S3.
In a preferred embodiment, S1 of the present embodiment includes:
s11, collecting images of two sides of the railway wagon;
actual vehicle passing images are continuously acquired through the online equipment. Deep learning needs support of a large number of images, if the data set is too small, the robustness of the model is affected, and the effect on the form of an unseen target for subsequent recognition is poor. And selecting original images with different target forms, image gray level differences, rainwater and other conditions from the sample images, wherein the more the types of the original images are, the better the test effect is. And adjusting the collected image as a whole, performing whole gray level analysis, and if the gray level value is too low, performing self-adaptive histogram equalization on the gray level value to improve the contrast and increase the segmentation accuracy.
S12, intercepting images on two sides of the railway wagon by using a frame to obtain sub-images comprising cotter pins and position identification components, marking the cotter pins and the position identification components in the sub-images, and constructing a sample set by using the marked sub-images, wherein the position identification components are position relations required by cotter pin fault judgment;
and intercepting the sub-image through a framework combining a priori knowledge and hardware, wherein the size of the sub-image is fixed and the target area of the cotter pin is completely included. In the embodiment, the image is cut into 256 × 256 images, and the edge gray scale is further influenced if the difference between the target and the background is small and the image is directly captured by an original image without being adjusted in size. The number of normal images in the sample is far larger than that of fault images, so that the fault images are enhanced, and the fault samples are increased by adding the transformation of overturning, translation and the like.
Because the cotter pin target is small, and the image gray difference is small, the fault recognition is carried out by selecting the segmentation model, even if the segmentation result has a small amount of interference, the image processing can be used for removing, and the fault recognition accuracy is improved.
In the embodiment, the collected sample images are marked to generate label images corresponding to the sample images one by one, and the target image is an image to be generated in the subsequent prediction. The grayscale image of the label image includes cotter pins and other types of positional relationships required for failure determination. Obtaining a target binary image alone is often difficult to determine faults, so a reference object having a fixed positional relationship with the target binary image needs to be selected. The same gray values in the gray map are of one type.
S3 of the present embodiment is: intercepting a railway wagon image to be detected by using a frame, intercepting a subimage comprising a cotter pin, inputting the subimage into a segmentation model, and acquiring segmentation areas of the cotter pin and a position identification component;
in S4 of the present embodiment, it is determined whether or not the cotter pin is lost based on the positional relationship between the cotter pin and the position indicator in the divided region, and if the cotter pin is lost, it is determined that a cotter pin loss failure has occurred, and a lost cotter pin is warned.
The sample set of the embodiment scrambles the fault image and the normal image, selects partial images as a test set, and adds a regularization dropout layer to prevent over-fitting training. In the step S2, the learning rate during training is decreased with the increase of the number of training steps, an advance stop condition is added during training, and the training is stopped if the loss of the test set continues and the number of iteration rounds is not decreased any more. Data enhancement is added during training, and the generalization capability of the model is enhanced by adding turnover, deviation and the like.
S2 of the present embodiment further includes: and (3) carrying out fault recognition on the new sample by using the obtained segmentation model, if the new sample has poor recognition effect, selecting and marking the new sample, putting the new sample into a training sample, reading the segmentation model to train continuously, and continuously testing and updating the weight until the result is satisfied, thereby completely training the segmentation model.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that features described in different dependent claims and herein may be combined in ways different from those described in the original claims. It is also to be understood that features described in connection with individual embodiments may be used in other described embodiments.
Claims (7)
1. A fault identification method for a cotter pin on a spring supporting plate of a railway wagon is characterized by comprising the following steps:
s1, collecting images of two sides of the railway wagon, and constructing a sample set;
the concentrated sample image is a subimage obtained by cutting out a cotter pin from the wagon image, and the cotter pin is marked in the subimage and is a cotter pin of a dog-ear bolt on the spring supporting plate;
s2, inputting the images in the sample set into a U-Net segmentation network for training to obtain a segmentation model;
the U-Net segmentation network reduces two convolutions and downsampling of a first layer and two convolutions and upsampling of a last layer on the basis of the existing U-Net network, and changes the number of feature maps of each layer into half of the original number;
s3, inputting the railway wagon image to be detected into the segmentation model, and acquiring the segmentation area of the cotter pin;
and S4, carrying out fault analysis on the divided areas, judging whether the cotter pin is lost or not, if so, judging that the cotter pin is lost, and alarming the lost cotter pin.
2. The fault identification method according to claim 1, wherein the S1 includes:
s11, acquiring images of the rail wagon;
s12, intercepting the image of the railway wagon by using the frame to obtain a sub-image comprising the cotter pin and the position identification component, marking the cotter pin and the position identification component in the sub-image, and constructing a sample set by using the marked sub-image, wherein the position identification component is a position relation required by the cotter pin fault judgment;
the S3 is as follows: intercepting a railway wagon image to be detected by using a frame, intercepting a subimage comprising a cotter pin, inputting the subimage into a segmentation model, and acquiring segmentation areas of the cotter pin and a position identification component;
and S4, judging whether the cotter pin is lost or not according to the position relation between the cotter pin and the position identification component in the partition area, if so, judging that the cotter pin is lost, and alarming the lost cotter pin.
3. The method according to claim 2, wherein in step S2, the learning rate during training decreases as the number of training steps increases, and the training is stopped if the loss of the sample set does not decrease any more for a plurality of iterations.
4. The fault identification method according to claim 2, wherein the S2 further includes:
and (3) carrying out fault recognition on the new sample by using the obtained segmentation model, if the new sample has poor recognition effect, selecting and marking the new sample, putting the new sample into a training sample, reading the segmentation model to train continuously, and continuously testing and updating the weight until the result is satisfied, thereby completely training the segmentation model.
5. The fault identification method according to claim 2, wherein the railway wagon images acquired at S1 comprise various images with different cotter pin shapes and different image gray levels.
6. The method according to claim 2, wherein in S12, the method further comprises enhancing, flipping, and translating the sub-images of the cotter pin failure to increase the number of sub-images in the sample set.
7. The fault identification method according to claim 2, wherein the S11 further includes acquiring a gray level of the image of the collected railway wagon, and if the gray level is lower than a set threshold, performing adaptive histogram equalization on the image to improve contrast.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911272261.8A CN111080600A (en) | 2019-12-12 | 2019-12-12 | Fault identification method for split pin on spring supporting plate of railway wagon |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911272261.8A CN111080600A (en) | 2019-12-12 | 2019-12-12 | Fault identification method for split pin on spring supporting plate of railway wagon |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111080600A true CN111080600A (en) | 2020-04-28 |
Family
ID=70314011
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911272261.8A Pending CN111080600A (en) | 2019-12-12 | 2019-12-12 | Fault identification method for split pin on spring supporting plate of railway wagon |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111080600A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832562A (en) * | 2020-07-16 | 2020-10-27 | 哈尔滨市科佳通用机电股份有限公司 | Fault identification method of spring pallet based on image processing |
| CN112102297A (en) * | 2020-09-17 | 2020-12-18 | 哈尔滨市科佳通用机电股份有限公司 | Method for identifying breaking fault of spring supporting plate of railway wagon bogie |
| CN112132821A (en) * | 2020-09-30 | 2020-12-25 | 哈尔滨市科佳通用机电股份有限公司 | A method for detecting the loss of split pins based on image processing |
| CN112149676A (en) * | 2020-09-11 | 2020-12-29 | 中国铁道科学研究院集团有限公司 | Small target detection processing method for railway goods loading state image |
| CN112434695A (en) * | 2020-11-20 | 2021-03-02 | 哈尔滨市科佳通用机电股份有限公司 | Upper pull rod fault detection method based on deep learning |
| CN115170923A (en) * | 2022-07-19 | 2022-10-11 | 哈尔滨市科佳通用机电股份有限公司 | Fault identification method for loss of railway wagon supporting plate nut |
| CN116188449A (en) * | 2023-03-13 | 2023-05-30 | 哈尔滨市科佳通用机电股份有限公司 | Rail wagon relief valve pull rod split pin loss fault identification method and equipment |
| CN116452906A (en) * | 2023-03-03 | 2023-07-18 | 哈尔滨市科佳通用机电股份有限公司 | Railway wagon fault picture generation method based on text description |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354827A (en) * | 2015-10-08 | 2016-02-24 | 成都唐源电气有限责任公司 | Method and system for intelligently identifying clamp nut shedding in catenary image |
| CN106600581A (en) * | 2016-12-02 | 2017-04-26 | 北京航空航天大学 | Train operation fault automatic detection system and method based on binocular stereoscopic vision |
| US9836808B2 (en) * | 2015-06-23 | 2017-12-05 | Nxp Usa, Inc. | Apparatus and method for verifying image data comprising mapped texture image data |
| CN108805868A (en) * | 2018-05-25 | 2018-11-13 | 哈尔滨市科佳通用机电股份有限公司 | The image processing method and fault detection method that EEF bogie equipment fault detects under a kind of vehicle-mounted vehicle of electricity business |
| CN109300114A (en) * | 2018-08-30 | 2019-02-01 | 西南交通大学 | Detection method for tightness and missing of extremely small target parts of high-speed rail catenary support device |
| CN110136157A (en) * | 2019-04-09 | 2019-08-16 | 华中科技大学 | A Deep Learning-Based Method for Vessel Wall Segmentation in 3D Carotid Ultrasound Images |
-
2019
- 2019-12-12 CN CN201911272261.8A patent/CN111080600A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9836808B2 (en) * | 2015-06-23 | 2017-12-05 | Nxp Usa, Inc. | Apparatus and method for verifying image data comprising mapped texture image data |
| CN105354827A (en) * | 2015-10-08 | 2016-02-24 | 成都唐源电气有限责任公司 | Method and system for intelligently identifying clamp nut shedding in catenary image |
| CN106600581A (en) * | 2016-12-02 | 2017-04-26 | 北京航空航天大学 | Train operation fault automatic detection system and method based on binocular stereoscopic vision |
| CN108805868A (en) * | 2018-05-25 | 2018-11-13 | 哈尔滨市科佳通用机电股份有限公司 | The image processing method and fault detection method that EEF bogie equipment fault detects under a kind of vehicle-mounted vehicle of electricity business |
| CN109300114A (en) * | 2018-08-30 | 2019-02-01 | 西南交通大学 | Detection method for tightness and missing of extremely small target parts of high-speed rail catenary support device |
| CN110136157A (en) * | 2019-04-09 | 2019-08-16 | 华中科技大学 | A Deep Learning-Based Method for Vessel Wall Segmentation in 3D Carotid Ultrasound Images |
Non-Patent Citations (1)
| Title |
|---|
| 陈景文 等: "基于U-net网络的航拍绝缘子检测", 《陕西科技大学学报》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111832562B (en) * | 2020-07-16 | 2021-03-16 | 哈尔滨市科佳通用机电股份有限公司 | Spring supporting plate fault identification method based on image processing |
| CN111832562A (en) * | 2020-07-16 | 2020-10-27 | 哈尔滨市科佳通用机电股份有限公司 | Fault identification method of spring pallet based on image processing |
| CN112149676B (en) * | 2020-09-11 | 2024-04-30 | 中国铁道科学研究院集团有限公司 | A small target detection and processing method for railway cargo loading status images |
| CN112149676A (en) * | 2020-09-11 | 2020-12-29 | 中国铁道科学研究院集团有限公司 | Small target detection processing method for railway goods loading state image |
| CN112102297B (en) * | 2020-09-17 | 2021-04-20 | 哈尔滨市科佳通用机电股份有限公司 | Method for identifying breaking fault of spring supporting plate of railway wagon bogie |
| CN112102297A (en) * | 2020-09-17 | 2020-12-18 | 哈尔滨市科佳通用机电股份有限公司 | Method for identifying breaking fault of spring supporting plate of railway wagon bogie |
| CN112132821A (en) * | 2020-09-30 | 2020-12-25 | 哈尔滨市科佳通用机电股份有限公司 | A method for detecting the loss of split pins based on image processing |
| CN112434695A (en) * | 2020-11-20 | 2021-03-02 | 哈尔滨市科佳通用机电股份有限公司 | Upper pull rod fault detection method based on deep learning |
| CN112434695B (en) * | 2020-11-20 | 2021-07-16 | 哈尔滨市科佳通用机电股份有限公司 | Upper pull rod fault detection method based on deep learning |
| CN115170923A (en) * | 2022-07-19 | 2022-10-11 | 哈尔滨市科佳通用机电股份有限公司 | Fault identification method for loss of railway wagon supporting plate nut |
| CN116452906A (en) * | 2023-03-03 | 2023-07-18 | 哈尔滨市科佳通用机电股份有限公司 | Railway wagon fault picture generation method based on text description |
| CN116452906B (en) * | 2023-03-03 | 2024-01-30 | 哈尔滨市科佳通用机电股份有限公司 | Railway wagon fault picture generation method based on text description |
| CN116188449A (en) * | 2023-03-13 | 2023-05-30 | 哈尔滨市科佳通用机电股份有限公司 | Rail wagon relief valve pull rod split pin loss fault identification method and equipment |
| CN116188449B (en) * | 2023-03-13 | 2023-08-08 | 哈尔滨市科佳通用机电股份有限公司 | Rail wagon relief valve pull rod split pin loss fault identification method and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111080600A (en) | Fault identification method for split pin on spring supporting plate of railway wagon | |
| CN112967243B (en) | Deep learning chip packaging crack defect detection method based on YOLO | |
| CN111079747B (en) | Railway wagon bogie side frame fracture fault image identification method | |
| CN111080598B (en) | Bolt and nut missing detection method for coupler yoke key safety crane | |
| CN111080620A (en) | Road disease detection method based on deep learning | |
| CN111080611A (en) | A method for image recognition of railway freight car bolster spring breakage faults | |
| CN111091545B (en) | Method for detecting loss fault of bolt at shaft end of rolling bearing of railway wagon | |
| CN111091541B (en) | Method for identifying fault of missing nut in cross beam assembly of railway wagon | |
| CN111079631A (en) | Method and system for identifying falling fault of hook lifting rod of railway wagon | |
| CN111091543B (en) | Railway wagon swing bolster spring loss fault target detection method | |
| CN111080608A (en) | Image recognition method of automatic brake valve plug door handle closing fault of railway freight car derailment | |
| CN109376768A (en) | A deep learning-based fault diagnosis method for aerial image tower signage | |
| CN111091548B (en) | Railway wagon adapter dislocation fault image identification method and system based on deep learning | |
| CN111079821A (en) | Derailment automatic braking pull ring falling fault image identification method | |
| CN111080601A (en) | Image recognition method for fault image of pull ring grinding shaft of railway freight car derailment braking device | |
| CN111080617A (en) | Railway wagon brake beam pillar round pin loss fault identification method | |
| CN113516629A (en) | TFDS passed the job intelligent detection system | |
| CN115797314A (en) | Part surface defect detection method, system, equipment and storage medium | |
| CN113221839A (en) | Automatic truck image identification method and system | |
| CN116228651A (en) | Cloth defect detection method, system, equipment and medium | |
| CN118379548A (en) | Image defect identification method and system | |
| CN115170923B (en) | Fault identification method for loss of railway wagon supporting plate nut | |
| CN111080599A (en) | A fault identification method for the hook lift rod of a railway freight car | |
| CN115937095A (en) | Printing defect detection method and system integrating image processing algorithm and deep learning | |
| CN112329858B (en) | Image recognition method for breakage fault of anti-loosening iron wire of railway motor car |
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: 20200428 |
|
| RJ01 | Rejection of invention patent application after publication |