CN111080608A - Method for recognizing closing fault image of automatic brake valve plug handle of railway wagon in derailment - Google Patents
Method for recognizing closing fault image of automatic brake valve plug handle of railway wagon in derailment Download PDFInfo
- Publication number
- CN111080608A CN111080608A CN201911272335.8A CN201911272335A CN111080608A CN 111080608 A CN111080608 A CN 111080608A CN 201911272335 A CN201911272335 A CN 201911272335A CN 111080608 A CN111080608 A CN 111080608A
- Authority
- CN
- China
- Prior art keywords
- image
- layer
- data set
- brake valve
- automatic brake
- 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
- 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
The invention discloses a method for identifying a fault image of closing of a handle of an automatic brake valve of a railway wagon during derailment, and relates to a method for identifying a fault image of a railway wagon. The invention aims to solve the problem of low image recognition accuracy of closing fault images of the existing automatic brake valve plug handle for the railway wagon derailment. The process is as follows: firstly, respectively building equipment around a truck track to obtain two-dimensional images of two sides of a truck; secondly, roughly positioning the handle part of the derailed automatic brake valve plug valve on the obtained two-dimensional image; thirdly, establishing an original sample data set based on the image of the handle part of the derailment automatic braking valve plug door with rough positioning; fourthly, carrying out data set amplification on the original sample data set; fifthly, obtaining a trained positioning model SSD and a classification model VGG 16; and sixthly, generating a message according to the fault position and the fault category and uploading the message to an alarm platform. The method is used for the field of railway wagon fault image recognition.
Description
Technical Field
The invention relates to a fault image identification method for a railway wagon.
Background
Train derailment is a very serious accident in the running process of railway vehicles, and in order to avoid the occurrence of the accident, a derailment automatic braking device is often arranged on a railway wagon. After checking and confirming that the train cannot run due to air leakage of the derailing automatic brake device, the valve plug handle can be closed first, an air passage between the derailing brake valve and the train brake main pipe is cut off, a line is quickly opened, and the valve plug handle is opened after maintenance. However, in the application, the derailment valve plug handle closing fault of the derailment automatic braking device is found to occur due to the misoperation of related maintainers, so that the basic performance of the railway wagon is influenced, and the potential safety hazard is brought to the running of the wagon. The existing car inspection operation mode of manually looking at the pictures one by one has the problems of influence of personnel quality and responsibility, error and omission detection, difficulty in ensuring the operation quality, huge labor cost, low efficiency and the like.
Therefore, the automatic detection of the derailing valve plug handle closing fault of the derailing automatic braking device has important significance. By combining image processing and deep learning technologies, automatic fault identification and alarm are realized, and the quality and efficiency of vehicle inspection operation are effectively improved.
Disclosure of Invention
The invention aims to solve the problem of low accuracy of image recognition of closing fault images of the existing automatic derailment brake valve plug handle of the rail wagon, and provides an image recognition method of closing fault images of the automatic derailment brake valve plug handle of the rail wagon.
The method for identifying the closing fault image of the automatic brake valve plug handle during the derailment of the railway wagon comprises the following specific processes:
firstly, respectively building equipment around a truck track to obtain two-dimensional images of two sides of a truck;
step two, roughly positioning the handle part of the derailed automatic brake valve plug valve on the two-dimensional image obtained in the step one;
step three, establishing an original sample data set based on the image of the handle part of the derailment automatic braking valve plug door roughly positioned in the step two; the specific process is as follows:
dividing the image data sets of the derailment automatic brake valve plug handle part roughly positioned in the second step into two categories, wherein one category corresponds to the type of the railway wagon GQ70, and the other category corresponds to the types of the railway wagon NX70A, X70 and X6K;
each category data set includes: an original image set and a mark information set;
the original image set is a local area image which contains the handle part of the derail automatic brake valve and is roughly positioned in the step two;
the marking information set is information of a rectangular sub-area containing a handle part of the derailment automatic brake valve and is obtained in a manual marking mode;
the original image set and the marking information data set are in one-to-one correspondence, namely each image corresponds to one marking data;
obtaining an original sample data set based on the original image set and the marking information data set;
fourthly, performing data set amplification on the original sample data set;
step five, respectively establishing a positioning model SSD and a classification model VGG16 for GQ70 data sets based on the amplified wagon model, respectively establishing a positioning model SSD and a classification model VGG16 for NX70A, X70 and X6K data sets based on the amplified wagon model, and training to obtain a trained positioning model SSD and a classification model VGG 16;
sixthly, positioning the components in the sub-area image of the image to be detected by using the positioning model;
utilizing the classification model to perform fault judgment on the part image positioned by the positioning model;
and generating a message according to the fault information and the fault position and fault category, and uploading the message to an alarm platform.
The invention has the beneficial effects that:
the high-definition imaging equipment on the two sides of the truck track is used for shooting the truck moving at high speed to obtain high-definition images on the two sides of the truck. And acquiring a coarse positioning image containing the component according to the wheel base information, the bogie type and other prior knowledge. And collecting, sorting and amplifying data of the images to obtain a training image sample set. And establishing a proper deep neural network according to the fault type, and training for multiple times until the model converges to obtain corresponding parameters. And in the identification stage, loading parameters, inputting the shot images into a network to obtain a prediction result, judging whether the images are in failure or not according to the prediction result, and alarming a failure area if the images are in failure.
1. The automatic identification technology is introduced into truck fault detection, automatic fault identification and alarm are realized, only the alarm result needs to be confirmed manually, the labor cost is effectively saved, and the operation quality and the operation efficiency are improved.
2. The deep learning algorithm is applied to automatic identification of the closing fault of the derailment automatic brake valve plug handle, and compared with the traditional machine vision detection method, the method has higher accuracy and stability.
3. The positioning network structure is improved, and the loss function is modified, so that the convergence speed of model training is increased.
Drawings
FIG. 1 is a flow chart of model training according to the present invention.
Detailed Description
The first embodiment is as follows: the method for identifying the closing fault image of the automatic brake valve plug handle during the derailment of the railway wagon comprises the following specific steps:
firstly, respectively building high-definition equipment around a truck track, shooting a truck passing at a high speed, and acquiring two-dimensional images of two sides of the truck; by adopting line scanning, seamless splicing of images can be realized, and a two-dimensional image with a large visual field and high precision is generated.
Step two, roughly positioning the handle part of the derailed automatic brake valve plug valve on the two-dimensional image obtained in the step one;
step three, establishing an original sample data set based on the image of the handle part of the derailment automatic braking valve plug door roughly positioned in the step two; the specific process is as follows:
the truck parts can be influenced by natural conditions such as rainwater, mud, oil, black paint and the like or artificial conditions. Also, there may be differences in the images taken at different sites. Therefore, there is a certain difference between the handle images of the derailed automatic brake valve plug. Therefore, in the process of collecting the training image data set, the diversity is ensured, and the images of different sites under various conditions are collected as much as possible.
The shape of the handle of the derailing automatic brake valve plug in the image can be different for different vehicle types due to different installation modes. According to the morphological characteristics of the image data sets of the handle parts of the derailment automatic brake valve in the second step, roughly positioned, are divided into two categories, wherein one category corresponds to the type of the railway wagon GQ70, and the other category corresponds to the types of the railway wagon NX70A, X70 and X6K, and the image data sets of the parts of different types are further distinguished so as to enhance the part positioning accuracy and the fault recognition rate;
each category data set includes: an original image set and a mark information set;
the original image set is a local area image which contains the handle part of the derail automatic brake valve and is roughly positioned in the step two;
the marking information set is information of a rectangular sub-area containing a handle part of the derailment automatic brake valve and is obtained in a manual marking mode;
the original image set and the marking information data set are in one-to-one correspondence, namely each image corresponds to one marking data;
obtaining an original sample data set based on the original image set and the marking information data set;
fourthly, performing data set amplification on the original sample data set;
step five, because the forms and the fault performances of the plug door handles of the derailment automatic braking valves of different vehicle types are different, respectively establishing a positioning model SSD and a classification model VGG16 for a GQ70 data set based on the amplified railway wagon type, respectively establishing a positioning model SSD and a classification model VGG16 for a NX70A, an X70 and an X6K data set based on the amplified railway wagon type, and training to obtain a trained positioning model SSD and a classification model VGG16, wherein as shown in FIG. 1, the GQ70 vehicle type data set is taken as an example for explanation;
sixthly, positioning the components in the sub-area image of the image to be detected by using the positioning model;
utilizing the classification model to perform fault judgment on the part image positioned by the positioning model;
and generating a message according to the fault information and the fault position and fault category, and uploading the message to an alarm platform.
The second embodiment is as follows: the difference between the first embodiment and the second embodiment is that in the second step, the two-dimensional image obtained in the first step is subjected to rough positioning of the handle part of the derailment automatic brake valve plug; the specific process is as follows:
the position of the component is roughly positioned according to the truck wheel base information and the truck type information, and a local area image containing the derailment automatic brake valve plug handle component is intercepted from the side two-dimensional image, so that the time required by fault identification can be effectively reduced, and the identification accuracy can be improved.
Other steps and parameters are the same as those in the first embodiment.
The third concrete implementation mode: in this embodiment, which differs from the first or second embodiment, the data set is expanded in the fourth step; the specific process is as follows:
although the creation of the sample data set includes images under various conditions, data amplification of the sample data set is still required to improve the stability of the algorithm. The amplification form comprises operations of rotation, translation, zooming, mirror image and the like of the image, and each operation is performed under random conditions, so that the diversity and applicability of the sample can be ensured to the greatest extent.
Other steps and parameters are the same as those in the first or second embodiment.
The fourth concrete implementation mode: the difference between the present embodiment and the first to the third embodiment is that, in the fifth step, because the shapes and the failure performances of the plug handles of the derailment automatic braking valves of different vehicle types are different, a positioning model SSD and a classification model VGG16 are respectively established for a GQ70 type data set based on the amplified vehicle type of the railway wagon, and a positioning model SSD and a classification model VGG16 are respectively established for a NX70A type data set, an X70 type data set and an X6K type data set based on the amplified vehicle type of the railway wagon, and training is performed to obtain a trained positioning model SSD and a trained classification model VGG16, as shown in fig. 1, the positioning model SSD and the classification model VGG16 are explained by taking a GQ70 vehicle type data set as an example; the specific process is as follows:
step five, first: training a positioning model SSD to obtain a trained positioning model SSD; the specific process is as follows:
the invention adopts SSD (Single Shot MultiBox Detector) detection algorithm to accurately position the handle of the derailed automatic braking valve plug in the image;
the SSD algorithm is based on a VGG-16 network to extract high-level features in an original image, and feature maps with different scales are used for detection. Because the feature extraction network can generate a plurality of convolution feature maps with different scales in the operation process, the convolution feature maps contain different semantic features and position sensitivity, and are suitable for multi-scale detection. The SSD algorithm usually uses 6 layers of feature maps with different sizes for detection, and under the condition that the input image is 300 × 300, the feature map sizes of the layers are different from 38 × 38 to 1 × 1.
As can be seen from the SSD network structure, the more advanced feature maps in the network should learn both the high-level features for detection and the local information for transmission to the next layer of feature maps. This forms a seemingly contradictory learning task: a top feature layer is required to maintain both underlying information and learn high-level abstract features. In order to meet the contradictory requirements, the invention introduces a prediction supplementary module based on a residual error structure, which is responsible for learning higher-level abstract information, and a backbone network can keep more original bottom information and naturally transmit the information to the next layer.
According to the invention, a double-branch depth residual error structure is added behind the feature map of each scale, the structure is simple, the problem of gradient disappearance can be effectively avoided, and a better training effect is easily obtained. The improved SSD network structure is as follows.
The positioning model SSD target detection network structure comprises an input layer, a traditional VGG-16 layer, a residual layer, conv5, a residual layer, a conv6, a residual layer, a conv7, a residual layer, a conv8, a residual layer, a conv9, a residual layer, a conv10 and a residual layer;
after a network is constructed, the amplified marker information data set is scaled to a fixed size of 512 x 512, and is input into a positioning model SSD target detection network; outputting a rectangular frame coordinate containing a target by the marking information data set through an SSD target detection network, and calculating a loss value with the marking information data set in the amplified original sample data set through a loss function;
when the SSD calculates the loss value, all candidate blocks can be divided into two categories, positive and negative: in all the prior frames, the frame with the largest overlapping rate with each labeling frame is regarded as a positive sample, in the rest prior frames, the frame with the overlapping rate more than 0.5 with any labeling frame is also regarded as a positive sample, and the rest is regarded as a negative sample. Since the target fraction is usually much smaller than the background fraction in most training images, there are two problems arising from the fact that the negative samples are much larger than the positive ones:
1) the error loss proportion of the negative sample is too large due to excessive negative samples, the error loss of the positive sample is easily submerged, and the convergence of the model is not facilitated;
2) most negative examples are not on the transition areas of the foreground and background, which are relatively easy to classify and are called easy negative examples. Such negative examples are effective for convergence of model training parameters. The model most requires samples with large loss values and large influence on parameter convergence, which are also called difficult samples. Difficult sample bowden is usually required to make parameter updates more efficient.
According to the invention, the cross entropy loss function is improved, so that the contribution of the difficult sample to parameter convergence is increased, the difficult sample is mined, and the model convergence is accelerated. The conventional cross entropy loss function is of the form:
Lce＝-log(Pt)
wherein P istRepresenting the relative confidence of the prediction box with respect to the annotation box. When P is presenttThe larger the value, the more accurate the classification is, the easier the sample is classified, the smaller its contribution to the loss value. When P is presenttThe smaller the value, the less accurate the classification, the less easily the sample is classified, the greater its contribution to the loss value. Since the background samples existing in a large amount are all simple negative samples, when a large amount of the samples are superposed, the sum of the generated loss values is large, and the classification balance of the background is influenced.
The invention will be (1-P)t)αAdding the modulation factor into a traditional cross entropy loss function as a modulation factor, wherein the cross entropy loss function after the modulation factor is added is as follows:
MLce＝-(1-Pt)αlog(Pt)
wherein, PtIs the relative confidence of the predicted frame relative to the labeled frame, p is the probability of predicting as a foreground object, y is the sample class, α is the modulation factor, ML is the probability of the foreground objectceIs a cross entropy loss function after adding the modulation factor;
it has the following properties: when a sample is a difficult sample, the classification result is easily misjudged, then (1-P)t) Is closer to 1, indicating that its loss value will be maximally preserved; when a sample is an easy sample, (1-P)t) Is close to 0, indicating that its loss value is greatly reduced, making its contribution to the overall loss value smaller. Therefore, the loss values of a large number of easy samples can be compressed in a small interval by using the modulation factor, so that the loss value weight of the difficult samples is highlighted, and the effect of mining the difficult samples is achieved.
And the convergence speed of the model parameter training is accelerated by improving the loss value calculation function.
Calculating a loss value through a loss function, and performing weight optimization through an optimizer Adam; the Adam optimizer has the advantages of high efficiency, small occupied memory, suitability for large-scale data and the like.
After passing through a loss function and an optimizer, calculating a new weight coefficient, updating the weight coefficient, completing one training iteration, repeatedly executing the step five, completing iteration of all images for a fixed number of times, but not updating the weight for each iteration, and only updating the weight with a lower loss function until an optimal weight coefficient is found to obtain a trained positioning model SSD;
step five two: training a classification model to obtain a trained classification model VGG 16; the specific process is as follows:
after the precision training of the component positioning network meets the expected requirement, a trained positioning model SSD can be used for collecting a large number of local area images containing the handle component of the derailment automatic braking valve plug in the step two in rough positioning;
training a local area image containing the handle part of the derailment automatic brake valve and the amplified label information data set obtained in the step four as a training set of a classification model VGG16 until the cross entropy loss gradually converges;
the method adopts a Visual Geometry Group Network (VGG-16) algorithm to judge the fault of the derailed automatic brake valve plug handle in the image.
The invention does not directly and generally use the model parameters of the VGG-16, but carries out fine tuning multiplexing on the VGG-16 according to the characteristics of the identification task, abandons the full connection layers of the last layers, and replaces the full connection layers with the output layers which accord with the identification task, and mainly has the following reasons:
1) the object of class 1000 can be identified in the VGG-16, however, the invention only needs to identify the handle part of the derailment automatic brake valve plug, the part is not concentrated in the traditional VGG-16 training sample, and the retraining should be carried out to make the part more pertinent.
2) The invention selects to extract the characteristics through VGG-16 without using an output layer thereof, changes 1000 output nodes of the original Softmax layer into 2 (turning off and turning on a handle) through self-defining the output layer, extracts the more universal characteristics by utilizing the convolutional neural network, obviously saves a large amount of training time and improves the training efficiency.
And training the modified VGG-16 network by using a sub-set containing components until the cross entropy loss & 1 is gradually converged.
Other steps and parameters are the same as those in one of the first to third embodiments.
The fifth concrete implementation mode: the difference between this embodiment and one of the first to the fourth embodiments is that the classification model VGG16 in the second step includes:
convolutional layer 1, max-pooling layer 1, convolutional layer 2, max-pooling layer 2, convolutional layer 3, max-pooling layer 3, convolutional layer 4, max-pooling layer 4, convolutional layer 5, average-pooling layer 1, full-link layer, Dropout layer, and full-link layer.
Other steps and parameters are the same as in one of the first to fourth embodiments.
The present invention is capable of other embodiments and its several details are capable of modifications in various obvious respects, all without departing from the spirit and scope of the present invention.
Claims (5)
1. The method for identifying the closing fault image of the automatic brake valve plug handle during the derailment of the railway wagon is characterized by comprising the following steps of: the method comprises the following specific processes:
firstly, respectively building equipment around a truck track to obtain two-dimensional images of two sides of a truck;
step two, roughly positioning the handle part of the derailed automatic brake valve plug valve on the two-dimensional image obtained in the step one;
step three, establishing an original sample data set based on the image of the handle part of the derailment automatic braking valve plug door roughly positioned in the step two; the specific process is as follows:
dividing the image data sets of the derailment automatic brake valve plug handle part roughly positioned in the second step into two categories, wherein one category corresponds to the type of the railway wagon GQ70, and the other category corresponds to the types of the railway wagon NX70A, X70 and X6K;
each category data set includes: an original image set and a mark information set;
the original image set is a local area image which contains the handle part of the derail automatic brake valve and is roughly positioned in the step two;
the marking information set is information of a rectangular sub-area containing a handle part of the derailment automatic brake valve and is obtained in a manual marking mode;
the original image set and the marking information data set are in one-to-one correspondence, namely each image corresponds to one marking data;
obtaining an original sample data set based on the original image set and the marking information data set;
fourthly, performing data set amplification on the original sample data set;
step five, respectively establishing a positioning model SSD and a classification model VGG16 for GQ70 data sets based on the amplified wagon model, respectively establishing a positioning model SSD and a classification model VGG16 for NX70A, X70 and X6K data sets based on the amplified wagon model, and training to obtain a trained positioning model SSD and a classification model VGG 16;
sixthly, positioning the components in the sub-area image of the image to be detected by using the positioning model;
utilizing the classification model to perform fault judgment on the part image positioned by the positioning model;
and generating a message according to the fault information and the fault position and fault category, and uploading the message to an alarm platform.
2. The method for identifying the closing fault image of the handle of the automatic brake valve of the derailment of railway wagon according to claim 1, wherein the method comprises the following steps: in the second step, roughly positioning the handle part of the derailed automatic brake valve plug valve on the two-dimensional image obtained in the first step; the specific process is as follows:
and roughly positioning the position of the component according to the truck wheel base information and the vehicle type information, and intercepting a local area image containing the handle part of the derailment automatic brake valve from a side two-dimensional image.
3. The method for identifying the closing fault image of the handle of the automatic brake valve of the derailment of railway wagon according to claim 2, wherein the image comprises: amplifying the data set in the fourth step; the specific process is as follows:
and performing data amplification on the sample data set. The augmentation modality includes rotation, translation, scaling, and mirroring operations of the image.
4. The method for identifying the closing fault image of the handle of the automatic brake valve of the derailment of railway wagon according to claim 3, wherein the image comprises: in the fifth step, a positioning model SSD and a classification model VGG16 are respectively established for GQ70 data sets based on the amplified wagon models, and a positioning model SSD and a classification model VGG16 are respectively established for NX70A, X70 and X6K data sets based on the amplified wagon models, and are trained to obtain a trained positioning model SSD and a trained classification model VGG 16; the specific process is as follows:
step five, first: training a positioning model SSD to obtain a trained positioning model SSD; the specific process is as follows:
the positioning model SSD target detection network structure comprises an input layer, a VGG-16 layer, a residual layer, a conv5 layer, a residual layer, a conv6 layer, a residual layer, a conv7 layer, a residual layer, a conv8 layer, a residual layer, a conv9 layer, a residual layer, a conv10 layer and a residual layer;
scaling the amplified tag information data set to a fixed size of 512 x 512, and inputting the scaled tag information data set into a positioning model SSD target detection network; outputting a rectangular frame coordinate containing a target by the marking information data set through an SSD target detection network, and calculating a loss value with the marking information data set in the amplified original sample data set through a loss function;
will be (1-P)t)αAdding the modulation factor into a cross entropy loss function, wherein the cross entropy loss function after the modulation factor is added is as follows:
MLce＝-(1-Pt)αlog(Pt)
wherein, PtIs the relative confidence of the predicted frame relative to the labeled frame, p is the probability of predicting as a foreground object, y is the sample class, α is the modulation factor, ML is the probability of the foreground objectceIs a cross entropy loss function after adding the modulation factor;
calculating a new weight coefficient after passing through a loss function and an optimizer, updating the weight coefficient, completing one training iteration, repeatedly executing the step five or one, and completing iteration of all images for a fixed number of times until an optimal weight coefficient is found to obtain a trained positioning model SSD;
step five two: training a classification model to obtain a trained classification model VGG 16; the specific process is as follows:
collecting the local area image containing the handle part of the derail automatic brake valve plug door roughly positioned in the step two by using a trained positioning model SSD;
and training a local area image containing the derailment automatic brake valve handle part and the amplified label information data set obtained in the step four as a training set of a classification model VGG16 until the cross entropy loss is converged.
5. The method for identifying the closing fault image of the handle of the automatic brake valve of the railway wagon derailment according to claim 4, wherein the image comprises: the structure of the classification model VGG16 in the step five and the step two comprises the following steps:
convolutional layer 1, max-pooling layer 1, convolutional layer 2, max-pooling layer 2, convolutional layer 3, max-pooling layer 3, convolutional layer 4, max-pooling layer 4, convolutional layer 5, average-pooling layer 1, full-link layer, Dropout layer, and full-link layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911272335.8A CN111080608A (en) | 2019-12-12 | 2019-12-12 | Method for recognizing closing fault image of automatic brake valve plug handle of railway wagon in derailment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911272335.8A CN111080608A (en) | 2019-12-12 | 2019-12-12 | Method for recognizing closing fault image of automatic brake valve plug handle of railway wagon in derailment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111080608A true CN111080608A (en) | 2020-04-28 |
Family
ID=70314070
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911272335.8A Pending CN111080608A (en) | 2019-12-12 | 2019-12-12 | Method for recognizing closing fault image of automatic brake valve plug handle of railway wagon in derailment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111080608A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN112489040A (en) * | 2020-12-17 | 2021-03-12 | 哈尔滨市科佳通用机电股份有限公司 | Truck auxiliary reservoir falling fault identification method |
| CN112766260A (en) * | 2021-01-15 | 2021-05-07 | 哈尔滨市科佳通用机电股份有限公司 | Image identification method and system for positioning air reservoir for accelerating and relieving railway train |
| CN113362302A (en) * | 2021-06-03 | 2021-09-07 | 西南交通大学 | Fault detection method of subway train electric box cover based on image recognition |
| CN113421246A (en) * | 2021-06-24 | 2021-09-21 | 长江存储科技有限责任公司 | Method for forming rail detection model and method for detecting rail abrasion |
| CN114755002A (en) * | 2022-04-06 | 2022-07-15 | 燕山大学 | Buffer balance valve fault diagnosis method based on full-connection neural network |
| CN115527170A (en) * | 2022-10-14 | 2022-12-27 | 哈尔滨市科佳通用机电股份有限公司 | Method and system for identifying closing fault of door stopper handle of automatic freight car derailing brake device |
| CN115661776A (en) * | 2022-10-25 | 2023-01-31 | 哈尔滨市科佳通用机电股份有限公司 | Method and system for identifying railway wagon brake beam safety chain falling fault image |
| CN115901299A (en) * | 2023-02-15 | 2023-04-04 | 慧铁科技有限公司 | Method for analyzing and processing faults of buffer component of train coupler hook |
| CN115973125A (en) * | 2023-02-15 | 2023-04-18 | 慧铁科技有限公司 | Method for processing fault of automatic derailment braking device of railway wagon |
| 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 |
| CN116403163A (en) * | 2023-04-20 | 2023-07-07 | 慧铁科技有限公司 | Method and device for identifying opening and closing states of handles of cut-off plug doors |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332089A (en) * | 2011-06-23 | 2012-01-25 | 北京康拓红外技术股份有限公司 | Railway wagon brake shoe key going-out fault recognition method based on artificial neural network |
| CN105389579A (en) * | 2015-11-12 | 2016-03-09 | 湖北工业大学 | Method for automatically identifying loss fault of cutout cock handle of train |
| CN107316007A (en) * | 2017-06-07 | 2017-11-03 | 浙江捷尚视觉科技股份有限公司 | A kind of monitoring image multiclass object detection and recognition methods based on deep learning |
| CN107563372A (en) * | 2017-07-20 | 2018-01-09 | 济南中维世纪科技有限公司 | A kind of license plate locating method based on deep learning SSD frameworks |
| CN108460382A (en) * | 2018-03-26 | 2018-08-28 | 西安电子科技大学 | Remote sensing image Ship Detection based on deep learning single step detector |
| CN109902697A (en) * | 2017-12-07 | 2019-06-18 | 展讯通信(天津)有限公司 | Multi-target detection method, device and mobile terminal |
| CN110232452A (en) * | 2019-06-12 | 2019-09-13 | 中国神华能源股份有限公司 | Repair method and system based on track car team state of the art |
-
2019
- 2019-12-12 CN CN201911272335.8A patent/CN111080608A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332089A (en) * | 2011-06-23 | 2012-01-25 | 北京康拓红外技术股份有限公司 | Railway wagon brake shoe key going-out fault recognition method based on artificial neural network |
| CN105389579A (en) * | 2015-11-12 | 2016-03-09 | 湖北工业大学 | Method for automatically identifying loss fault of cutout cock handle of train |
| CN107316007A (en) * | 2017-06-07 | 2017-11-03 | 浙江捷尚视觉科技股份有限公司 | A kind of monitoring image multiclass object detection and recognition methods based on deep learning |
| CN107563372A (en) * | 2017-07-20 | 2018-01-09 | 济南中维世纪科技有限公司 | A kind of license plate locating method based on deep learning SSD frameworks |
| CN109902697A (en) * | 2017-12-07 | 2019-06-18 | 展讯通信(天津)有限公司 | Multi-target detection method, device and mobile terminal |
| CN108460382A (en) * | 2018-03-26 | 2018-08-28 | 西安电子科技大学 | Remote sensing image Ship Detection based on deep learning single step detector |
| CN110232452A (en) * | 2019-06-12 | 2019-09-13 | 中国神华能源股份有限公司 | Repair method and system based on track car team state of the art |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN112489040A (en) * | 2020-12-17 | 2021-03-12 | 哈尔滨市科佳通用机电股份有限公司 | Truck auxiliary reservoir falling fault identification method |
| CN112766260A (en) * | 2021-01-15 | 2021-05-07 | 哈尔滨市科佳通用机电股份有限公司 | Image identification method and system for positioning air reservoir for accelerating and relieving railway train |
| CN112766260B (en) * | 2021-01-15 | 2021-09-14 | 哈尔滨市科佳通用机电股份有限公司 | Image identification method and system for positioning air reservoir for accelerating and relieving railway train |
| CN113362302B (en) * | 2021-06-03 | 2023-04-07 | 西南交通大学 | Fault detection method of subway train electric box cover based on image recognition |
| CN113362302A (en) * | 2021-06-03 | 2021-09-07 | 西南交通大学 | Fault detection method of subway train electric box cover based on image recognition |
| CN113421246A (en) * | 2021-06-24 | 2021-09-21 | 长江存储科技有限责任公司 | Method for forming rail detection model and method for detecting rail abrasion |
| CN114755002A (en) * | 2022-04-06 | 2022-07-15 | 燕山大学 | Buffer balance valve fault diagnosis method based on full-connection neural network |
| CN115527170A (en) * | 2022-10-14 | 2022-12-27 | 哈尔滨市科佳通用机电股份有限公司 | Method and system for identifying closing fault of door stopper handle of automatic freight car derailing brake device |
| CN115661776A (en) * | 2022-10-25 | 2023-01-31 | 哈尔滨市科佳通用机电股份有限公司 | Method and system for identifying railway wagon brake beam safety chain falling fault image |
| CN115901299A (en) * | 2023-02-15 | 2023-04-04 | 慧铁科技有限公司 | Method for analyzing and processing faults of buffer component of train coupler hook |
| CN115973125A (en) * | 2023-02-15 | 2023-04-18 | 慧铁科技有限公司 | Method for processing fault of automatic derailment braking device of railway wagon |
| CN115901299B (en) * | 2023-02-15 | 2023-06-06 | 慧铁科技有限公司 | Method for analyzing and processing faults of train coupler buffering component |
| 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 |
| CN116403163A (en) * | 2023-04-20 | 2023-07-07 | 慧铁科技有限公司 | Method and device for identifying opening and closing states of handles of cut-off plug doors |
| CN116403163B (en) * | 2023-04-20 | 2023-10-27 | 慧铁科技有限公司 | Method and device for identifying opening and closing states of handles of cut-off plug doors |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111080608A (en) | Method for recognizing closing fault image of automatic brake valve plug handle of railway wagon in derailment | |
| CN111091558B (en) | Railway wagon swing bolster spring jumping fault image identification method | |
| CN111489339B (en) | Method for detecting backup and main defects of high-speed railway locator bolts | |
| CN111091542B (en) | Image identification method for breakage fault of spring supporting plate of railway wagon bogie | |
| CN109948416A (en) | A kind of illegal occupancy bus zone automatic auditing method based on deep learning | |
| CN111507371B (en) | Method and device for automatically evaluating reliability of label on training image | |
| CN111079746B (en) | Railway wagon axle box spring fault image identification method | |
| CN111091545B (en) | Method for detecting loss fault of bolt at shaft end of rolling bearing of railway wagon | |
| CN111079817B (en) | Method for identifying fault image of cross beam of railway wagon | |
| CN111080617B (en) | Railway wagon brake beam pillar round pin loss fault identification method | |
| CN111652295B (en) | Railway wagon coupler yoke key joist falling fault identification method | |
| CN111079734B (en) | Method for detecting foreign matters in triangular holes of railway wagon | |
| CN111862029A (en) | Fault detection method for bolt part of vertical shock absorber of railway motor train unit | |
| CN111080621B (en) | Method for identifying railway wagon floor damage fault image | |
| CN111080600A (en) | Fault identification method for split pin on spring supporting plate of railway wagon | |
| CN113221839B (en) | Automatic truck image identification method and system | |
| CN111598855B (en) | 2C equipment high-speed rail contact net dropper defect detection method based on deep learning and transfer learning | |
| CN111080607B (en) | Rolling bearing oil slinging fault detection method based on image recognition | |
| CN113962951B (en) | Training method and device for detecting segmentation model, and target detection method and device | |
| CN111191604A (en) | Method, device and storage medium for detecting integrity of license plate | |
| CN111079822A (en) | Method for identifying dislocation fault image of middle rubber and upper and lower plates of axle box rubber pad | |
| CN110991447A (en) | Train number accurate positioning and identification method based on deep learning | |
| CN111091551A (en) | Method for detecting loss fault of brake beam strut opening pin of railway wagon | |
| CN115527170A (en) | Method and system for identifying closing fault of door stopper handle of automatic freight car derailing brake device | |
| Mohan et al. | Yolo v2 with bifold skip: a deep learning model for video based real time train bogie part identification and defect detection |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200428 |