CN114363582A - Integrated track inspection vehicle image processing system - Google Patents
Integrated track inspection vehicle image processing system Download PDFInfo
- Publication number
- CN114363582A CN114363582A CN202210252884.4A CN202210252884A CN114363582A CN 114363582 A CN114363582 A CN 114363582A CN 202210252884 A CN202210252884 A CN 202210252884A CN 114363582 A CN114363582 A CN 114363582A
- Authority
- CN
- China
- Prior art keywords
- inspection vehicle
- control
- unit
- module
- detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Train Traffic Observation, Control, And Security (AREA)
- Image Processing (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention relates to an integrated track inspection vehicle image processing system, which comprises: the camera shooting part is arranged on the rail inspection vehicle and used for acquiring images of the rail inspection vehicle in the running process under the control command sent by the control end according to the set period; the processor is used for receiving images of the rail inspection vehicle in the running process in real time and controlling the running of the rail inspection vehicle according to the processing result of the images; the invention provides an integrated track inspection vehicle image processing system which is additionally provided with image acquisition and processing on the basis of the original track inspection vehicle and judges the implementation of an operation plan through image processing.
Description
Technical Field
The invention relates to the technical field of rail inspection, in particular to an integrated rail inspection vehicle image processing system.
Background
After an urban subway or a high-speed railway is built, a test stage is generally carried out, before the test stage, a rail inspection vehicle is generally adopted to detect the flatness of a rail, and then the running speed and the like of each stage are measured through trial running. The construction process of the track is complex, so that the bending degree of some road sections is large, the road sections need to be judged by manual experience during trial operation, and reasonable operation correction is carried out, so that the test period is prolonged.
Disclosure of Invention
In view of the above, the present invention provides an integrated track inspection vehicle image processing system to solve the above problems in the prior art.
In order to achieve the purpose, the invention provides the following technical scheme:
integrated form track inspection car image processing system includes:
the camera shooting part is arranged on the rail inspection vehicle and used for acquiring images of the rail inspection vehicle in the running process under the control command sent by the control end according to the set period;
the processor is used for receiving images of the rail inspection vehicle in the running process in real time and controlling the running of the rail inspection vehicle according to the processing result of the images;
wherein the processor has:
the processing module is used for receiving the image of the current period in real time, carrying out abnormity detection on the image, and converting the image and a detection result contained in the image into a reference signal for controlling the operation of the rail inspection vehicle through the conversion unit after the abnormity detection;
the configuration module is used for configuring control parameters for operation control of the track inspection vehicle based on the reference signals;
the control module is used for planning the moving speed and the moving track of the track inspection vehicle based on the control parameters of the configuration module so as to control the operation of the track inspection vehicle;
and the regulation and control module is closed when the reference signals of two continuous periods contain abnormal signals, takes the copy of the reference signal of the previous period as the differential signal regulated and controlled by the current period, sends the differential signal and the reference signal of the current period to the correction module to form a correction signal, and configures the correction signal into single-line control to respond to the closing of the control module.
Further, the processing module has:
the extraction unit is used for extracting the image of the current period received in real time by taking a frame as a unit to form a frame image;
a detection unit configured to perform abnormality detection on the frame image;
a machine learning system configured to acquire a preset scheme of operation control of the rail inspection vehicle set corresponding to the abnormality detection based on the abnormality detection by the detection unit; wherein the preset scheme is updated based on a machine learning system;
and the conversion unit is used for converting the preset scheme acquired by the machine learning system into a reference signal for controlling the operation of the rail inspection vehicle.
Further, the machine learning system adds the abnormal detection obtained by the detection unit to a learning library, and determines whether to update the preset scheme according to the newly added abnormal detection as a sample through training of the machine learning system.
Further, the machine learning system carries out statistical marking on the newly added abnormal detection added to the learning library, and when the statistical marking of the newly added abnormal detection reaches an accumulative set threshold value, the machine learning system updates the preset scheme by taking the newly added abnormal detection as a sample.
Further, the configuration module has:
the analysis unit is used for analyzing the configuration parameters in the reference signals;
and the logic unit is used for matching the control logic based on the configuration parameters and writing the configuration parameters into the control logic to form the control parameters.
Further, the control module has:
and the trigger is used for switching on the connection with the control circuit of the rail patrol car based on the conduction of the logic control element and is used for driving the rail patrol car to control the operation of the rail patrol car according to the output of the control circuit.
Further, the regulatory module has:
the identification unit is coupled to the detection unit to acquire a detection result, judges the continuously acquired detection result and executes the closing operation of the control module when the reference signals of two continuous periods contain abnormal signals;
the clock unit is used for configuring a reference clock for acquiring the detection result for the identification unit;
the correction unit is used for taking a copy of the reference signal of the previous period as a differential signal regulated and controlled by the current period, and sending the differential signal and the reference signal of the current period to the correction module to form a correction signal;
and the execution unit is coupled to the analysis unit, analyzes the configuration parameters in the correction signal, matches the control logic based on the configuration parameters, writes the configuration parameters into the control logic to form control parameters, and drives the execution unit to access the control circuit based on the control parameters so as to execute a preset protection plan.
The invention provides an integrated track inspection vehicle image processing system, which adds image acquisition and processing on the basis of the original track inspection vehicle, judges the implementation of an operation plan through image processing, concretely, in a test stage, a plurality of plans are set manually, such as sharp curves, large-amplitude curves, viaducts and slope sections in the test stage respectively, in the test stage, images in the operation process are acquired by the track inspection vehicle in 5 seconds as a period to deduce basic information of a front road section, a corresponding operation control scheme is adopted according to the setting of the basic information to check whether the operation control scheme can be stably executed, then in the next test stage, the test speed is gradually increased, if the operation can be stably executed, a machine learning system is required to optimize and learn the manually set operation control scheme, and updating the operation control schemes, and repeating the test run for multiple times in sequence.
The invention combines the track inspection and the test operation process into a whole, and can apply the well-learned scheme to the test operation in the test operation stage so as to reduce the test operation period.
Drawings
FIG. 1 is a schematic diagram of the general framework of the present invention;
fig. 2 is a detailed framework principle schematic diagram of the present invention.
Detailed Description
The present invention is described in detail below with reference to the accompanying drawings, which refer to fig. 1 to 2.
Referring to fig. 1 to 2, the present invention provides an integrated rail inspection vehicle image processing system, including:
the camera shooting part is arranged on the rail inspection vehicle and used for acquiring images of the rail inspection vehicle in the running process under the control command sent by the control end according to the set period;
the processor is used for receiving images of the rail inspection vehicle in the running process in real time and controlling the running of the rail inspection vehicle according to the processing result of the images;
wherein the processor has:
the processing module is used for receiving the image of the current period in real time, carrying out abnormity detection on the image, and converting the image and a detection result contained in the image into a reference signal for controlling the operation of the rail inspection vehicle through the conversion unit after the abnormity detection;
the configuration module is used for configuring control parameters for operation control of the track inspection vehicle based on the reference signals;
the control module is used for planning the moving speed and the moving track of the track inspection vehicle based on the control parameters of the configuration module so as to control the operation of the track inspection vehicle;
and the regulation and control module is closed when the reference signals of two continuous periods contain abnormal signals, takes the copy of the reference signal of the previous period as the differential signal regulated and controlled by the current period, sends the differential signal and the reference signal of the current period to the correction module to form a correction signal, and configures the correction signal into single-line control to respond to the closing of the control module. The invention provides an integrated track inspection vehicle image processing system, which adds image acquisition and processing on the basis of the original track inspection vehicle, judges the implementation of an operation plan through image processing, concretely, in a test stage, a plurality of plans are set manually, such as sharp curves, large-amplitude curves, viaducts and slope sections in the test stage respectively, in the test stage, images in the operation process are acquired by the track inspection vehicle in 5 seconds as a period to deduce basic information of a front road section, a corresponding operation control scheme is adopted according to the setting of the basic information to check whether the operation control scheme can be stably executed, then in the next test stage, the test speed is gradually increased, if the operation can be stably executed, a machine learning system is required to optimize and learn the manually set operation control scheme, and updating the operation control schemes, and repeating the test run for multiple times in sequence. The invention combines the track inspection and the test operation process into a whole, and can apply the well-learned scheme to the test operation in the test operation stage so as to reduce the test operation period.
In the above, the processing module has:
the extraction unit is used for extracting the image of the current period received in real time by taking a frame as a unit to form a frame image;
a detection unit configured to perform abnormality detection on the frame image;
a machine learning system configured to acquire a preset scheme of operation control of the rail inspection vehicle set corresponding to the abnormality detection based on the abnormality detection by the detection unit; wherein the preset scheme is updated based on a machine learning system;
and the conversion unit is used for converting the preset scheme acquired by the machine learning system into a reference signal for controlling the operation of the rail inspection vehicle.
In the above, the machine learning system adds the abnormal detection obtained by the detection unit to the learning library, and determines whether to update the preset scheme according to the newly added abnormal detection as a sample through training of the machine learning system.
In the above, the machine learning system performs statistical marking on the newly added abnormal detections added to the learning library, and when the statistical marking of the newly added abnormal detections reaches the accumulation set threshold, the machine learning system updates the preset scheme by using the newly added abnormal detections as samples.
In the above, the configuration module has:
the analysis unit is used for analyzing the configuration parameters in the reference signals;
and the logic unit is used for matching the control logic based on the configuration parameters and writing the configuration parameters into the control logic to form the control parameters.
In the above, the control module has:
and the trigger is used for switching on the connection with the control circuit of the rail patrol car based on the conduction of the logic control element and is used for driving the rail patrol car to control the operation of the rail patrol car according to the output of the control circuit.
In the above, the regulatory module has:
the identification unit is coupled to the detection unit to acquire a detection result, judges the continuously acquired detection result and executes the closing operation of the control module when the reference signals of two continuous periods contain abnormal signals;
the clock unit is used for configuring a reference clock for acquiring the detection result for the identification unit;
the correction unit is used for taking a copy of the reference signal of the previous period as a differential signal regulated and controlled by the current period, and sending the differential signal and the reference signal of the current period to the correction module to form a correction signal;
and the execution unit is coupled to the analysis unit, analyzes the configuration parameters in the correction signal, matches the control logic based on the configuration parameters, writes the configuration parameters into the control logic to form control parameters, and drives the execution unit to access the control circuit based on the control parameters so as to execute a preset protection plan.
The principle of the invention is as follows: the image pickup part acquires images of the rail inspection vehicle in the running process according to a control command sent by a set period; the extraction unit extracts the image of the current period received in real time according to the frame as a unit to form a frame image; sending the frame image to a detection unit, and carrying out abnormity detection on the frame image by the detection unit; the machine learning system acquires a preset scheme of operation control of the rail inspection vehicle, which is set corresponding to the abnormal detection, based on the abnormal detection; the machine learning system adds the anomaly detection acquired by the detection unit to the learning library, carries out statistical marking on the newly added anomaly detection added to the learning library, and updates the preset scheme by taking the newly added anomaly detection as a sample through the machine learning system when the statistical marking of the newly added anomaly detection reaches an accumulative set threshold value.
And the conversion unit is used for converting the preset scheme acquired by the machine learning system into a reference signal for controlling the operation of the rail inspection vehicle. Sending the reference signal to an analysis unit, and analyzing the configuration parameters in the reference signal by the analysis unit; the logic unit matches the control logic based on the configuration parameters and writes the configuration parameters to the control logic to form the control parameters. At least one logic control element is connected with the logic unit, each logic control element is correspondingly provided with a trigger, and the triggers are used for being connected with a control circuit of the rail patrol car based on the conduction of the logic control elements and used for driving the rail patrol car to control the operation of the rail patrol car according to the output of the control circuit. If the reference signals of two continuous periods contain abnormal signals, the control module is closed, the copy of the reference signal of the previous period is used as a differential signal regulated and controlled by the current period, the differential signal and the reference signal of the current period are sent to the correction module to form a correction signal, and the correction signal is configured into single-line control to respond to the closing of the control module.
Specifically, the identification unit is coupled to the detection unit to acquire a detection result, and judges the continuously acquired detection result, and when the reference signal of two continuous periods contains an abnormal signal, the identification unit executes the closing operation of the control module; the correction unit takes the copy of the reference signal of the previous period as a differential signal regulated and controlled by the current period, and sends the differential signal and the reference signal of the current period to the correction module to form a correction signal; the execution unit is coupled to the analysis unit, analyzes the configuration parameters in the correction signal, matches the control logic based on the configuration parameters, writes the configuration parameters into the control logic to form control parameters, and drives the execution unit to access the control circuit based on the control parameters to execute a preset protection plan.
The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts of the present invention. The foregoing is only a preferred embodiment of the present invention, and it should be noted that there are objectively infinite specific structures due to the limited character expressions, and it will be apparent to those skilled in the art that a plurality of modifications, decorations or changes may be made without departing from the principle of the present invention, and the technical features described above may be combined in a suitable manner; such modifications, variations, combinations, or adaptations of the invention using its spirit and scope, as defined by the claims, may be directed to other uses and embodiments.
Claims (7)
1. Integrated form track inspection car image processing system, its characterized in that includes:
the camera shooting part is arranged on the rail inspection vehicle and used for acquiring images of the rail inspection vehicle in the running process under the control command sent by the control end according to the set period;
the processor is used for receiving images of the rail inspection vehicle in the running process in real time and controlling the running of the rail inspection vehicle according to the processing result of the images;
wherein the processor has:
the processing module is used for receiving the image of the current period in real time, carrying out abnormity detection on the image, and converting the image and a detection result contained in the image into a reference signal for controlling the operation of the rail inspection vehicle through the conversion unit after the abnormity detection;
the configuration module is used for configuring control parameters for operation control of the track inspection vehicle based on the reference signals;
the control module is used for planning the moving speed and the moving track of the track inspection vehicle based on the control parameters of the configuration module so as to control the operation of the track inspection vehicle;
and the regulation and control module is closed when the reference signals of two continuous periods contain abnormal signals, takes the copy of the reference signal of the previous period as the differential signal regulated and controlled by the current period, sends the differential signal and the reference signal of the current period to the correction module to form a correction signal, and configures the correction signal into single-line control to respond to the closing of the control module.
2. The integrated rail inspection vehicle image processing system of claim 1, wherein the processing module has:
the extraction unit is used for extracting the image of the current period received in real time by taking a frame as a unit to form a frame image;
a detection unit configured to perform abnormality detection on the frame image;
a machine learning system configured to acquire a preset scheme of operation control of the rail inspection vehicle set corresponding to the abnormality detection based on the abnormality detection by the detection unit; wherein the preset scheme is updated based on a machine learning system;
and the conversion unit is used for converting the preset scheme acquired by the machine learning system into a reference signal for controlling the operation of the rail inspection vehicle.
3. The integrated rail inspection vehicle image processing system according to claim 2, wherein the machine learning system adds the abnormality detection obtained by the detection unit to a learning library, and performs training of the machine learning system to determine whether to perform updating of the preset scheme based on the newly added abnormality detection as a sample.
4. The integrated rail inspection vehicle image processing system according to claim 3, wherein the machine learning system performs statistical labeling of newly added abnormal detections added to the learning library, and when the statistical labeling of the newly added abnormal detections reaches an accumulative setting threshold, the machine learning system updates the preset scheme by using the newly added abnormal detections as samples.
5. The integrated rail inspection vehicle image processing system of claim 1, wherein the configuration module has:
the analysis unit is used for analyzing the configuration parameters in the reference signals;
and the logic unit is used for matching the control logic based on the configuration parameters and writing the configuration parameters into the control logic to form the control parameters.
6. The integrated rail inspection vehicle image processing system of claim 1, wherein the control module has:
and the trigger is used for switching on the connection with the control circuit of the rail patrol car based on the conduction of the logic control element and is used for driving the rail patrol car to control the operation of the rail patrol car according to the output of the control circuit.
7. The integrated rail inspection vehicle image processing system of claim 1, wherein the conditioning module has:
the identification unit is coupled to the detection unit to acquire a detection result, judges the continuously acquired detection result and executes the closing operation of the control module when the reference signals of two continuous periods contain abnormal signals;
the clock unit is used for configuring a reference clock for acquiring the detection result for the identification unit;
the correction unit is used for taking a copy of the reference signal of the previous period as a differential signal regulated and controlled by the current period, and sending the differential signal and the reference signal of the current period to the correction module to form a correction signal;
and the execution unit is coupled to the analysis unit, analyzes the configuration parameters in the correction signal, matches the control logic based on the configuration parameters, writes the configuration parameters into the control logic to form control parameters, and drives the execution unit to access the control circuit based on the control parameters so as to execute a preset protection plan.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210252884.4A CN114363582B (en) | 2022-03-15 | 2022-03-15 | Integrated track inspection vehicle image processing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210252884.4A CN114363582B (en) | 2022-03-15 | 2022-03-15 | Integrated track inspection vehicle image processing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114363582A true CN114363582A (en) | 2022-04-15 |
| CN114363582B CN114363582B (en) | 2022-06-10 |
Family
ID=81094451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210252884.4A Active CN114363582B (en) | 2022-03-15 | 2022-03-15 | Integrated track inspection vehicle image processing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114363582B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070217670A1 (en) * | 2006-03-02 | 2007-09-20 | Michael Bar-Am | On-train rail track monitoring system |
| JP2015010456A (en) * | 2013-07-02 | 2015-01-19 | 日本電気株式会社 | Rail inspection device, and rail inspection method |
| US20190039633A1 (en) * | 2017-08-02 | 2019-02-07 | Panton, Inc. | Railroad track anomaly detection |
| CN110969082A (en) * | 2019-10-28 | 2020-04-07 | 中国信息通信研究院 | Clock synchronization test inspection method and system |
| WO2020103533A1 (en) * | 2018-11-20 | 2020-05-28 | 中车株洲电力机车有限公司 | Track and road obstacle detecting method |
| CN112441064A (en) * | 2019-08-30 | 2021-03-05 | 比亚迪股份有限公司 | Rail flaw detection method, device and system and automatic inspection vehicle |
| CN113023293A (en) * | 2021-02-08 | 2021-06-25 | 精锐视觉智能科技（深圳）有限公司 | Inspection method, device, equipment and system for belt conveyor |
| CN113112501A (en) * | 2021-05-11 | 2021-07-13 | 上海市东方海事工程技术有限公司 | Vehicle-mounted track inspection device and method based on deep learning |
-
2022
- 2022-03-15 CN CN202210252884.4A patent/CN114363582B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070217670A1 (en) * | 2006-03-02 | 2007-09-20 | Michael Bar-Am | On-train rail track monitoring system |
| JP2015010456A (en) * | 2013-07-02 | 2015-01-19 | 日本電気株式会社 | Rail inspection device, and rail inspection method |
| US20190039633A1 (en) * | 2017-08-02 | 2019-02-07 | Panton, Inc. | Railroad track anomaly detection |
| WO2020103533A1 (en) * | 2018-11-20 | 2020-05-28 | 中车株洲电力机车有限公司 | Track and road obstacle detecting method |
| CN112441064A (en) * | 2019-08-30 | 2021-03-05 | 比亚迪股份有限公司 | Rail flaw detection method, device and system and automatic inspection vehicle |
| CN110969082A (en) * | 2019-10-28 | 2020-04-07 | 中国信息通信研究院 | Clock synchronization test inspection method and system |
| CN113023293A (en) * | 2021-02-08 | 2021-06-25 | 精锐视觉智能科技（深圳）有限公司 | Inspection method, device, equipment and system for belt conveyor |
| CN113112501A (en) * | 2021-05-11 | 2021-07-13 | 上海市东方海事工程技术有限公司 | Vehicle-mounted track inspection device and method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114363582B (en) | 2022-06-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110262463B (en) | Rail transit platform door fault diagnosis system based on deep learning | |
| CN113870260B (en) | Welding defect real-time detection method and system based on high-frequency time sequence data | |
| CN102053563A (en) | Flight training data acquisition and quality evaluation system of analog machine | |
| CN108776452B (en) | Special equipment field maintenance monitoring method and system | |
| CN109711285B (en) | Training and testing method and device for recognition model | |
| CN108280542A (en) | A kind of optimization method, medium and the equipment of user's portrait model | |
| CN105185118A (en) | Vehicle peccancy detection device and method based on video detection | |
| CN108672316A (en) | A kind of micro parts quality detecting system based on convolutional neural networks | |
| CN109800717A (en) | Activity recognition video frame sampling method and system based on intensified learning | |
| CN111753653B (en) | High-speed rail contact net fastener identification and positioning method based on attention mechanism | |
| CN109543732A (en) | Assembly system and method based on category feature knowledge base | |
| CN114363582B (en) | Integrated track inspection vehicle image processing system | |
| CN114066804A (en) | Curved surface fault layer tooth position identification method based on deep learning | |
| CN111563886B (en) | Unsupervised feature learning-based tunnel steel rail surface disease detection method and device | |
| CN116524297B (en) | Weak supervision learning training method based on expert feedback | |
| CN219015651U (en) | Test system based on vehicle dynamic traffic automatic driving | |
| CN104615535B (en) | Method and device for generating test case based on extended data flow model | |
| CN113099175B (en) | Multi-model handheld cloud detection transmission system and detection method based on 5G | |
| CN110108510A (en) | Based on embedded system automobile electronics intelligent checking system and its method | |
| CN106569840A (en) | Method for machine vision driving assistance system to automatically obtain sample to improve recognition accuracy | |
| CN113076235B (en) | Time sequence abnormity detection method based on state fusion | |
| He et al. | Intelligent vehicle design based on paddlepaddle and deep learning | |
| CN114463151A (en) | Sub-health diagnosis method for rail vehicle door system | |
| CN207067770U (en) | Connecting bridge simulation detection system | |
| CN112116195A (en) | Railway beam production process identification method based on example segmentation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 518000 706, No.2 workshop, Baolong factory area, Anbo technology, No.2, Baolong 4th Road, Baolong community, Baolong street, Longgang District, Shenzhen City, Guangdong Province Patentee after: Shenzhen Zhonghui Innovation Technology Co.,Ltd. Address before: 518000 706, No.2 workshop, Baolong factory area, Anbo technology, No.2, Baolong 4th Road, Baolong community, Baolong street, Longgang District, Shenzhen City, Guangdong Province Patentee before: Shenzhen Zhonghui rail Intelligent Technology Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |