CN111105597A - Locomotive crew member working state early warning reminding system - Google Patents
Locomotive crew member working state early warning reminding system Download PDFInfo
- Publication number
- CN111105597A CN111105597A CN201811470645.6A CN201811470645A CN111105597A CN 111105597 A CN111105597 A CN 111105597A CN 201811470645 A CN201811470645 A CN 201811470645A CN 111105597 A CN111105597 A CN 111105597A
- Authority
- CN
- China
- Prior art keywords
- alarm
- processor
- crew
- crew member
- locomotive
- 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
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/06—Alarms for ensuring the safety of persons indicating a condition of sleep, e.g. anti-dozing alarms
Abstract
The invention discloses a locomotive crew member working state early warning reminding system, which comprises: 1) the image collector is used for collecting the work image of the crew member; 2) the processor is used for processing and transmitting the images collected by the image collector; 3) and the alarm device gives an alarm according to the early warning information provided by the processor, and implements graded voice alarm. The invention is based on a high-speed DSP (digital signal processing) embedded processing system with mature technology, adopts advanced video image processing technology, monitors the working state of a locomotive crew member in real time in the whole course of train operation, provides a new technical means for supervision of the working process of the crew member and prevention of fatigue state, can store video data in a certain time period, and has the accident recollection function of driver fatigue state images.
Description
Technical Field
The invention belongs to the technical field of railway train operation safety, and particularly relates to a locomotive crew member working state early warning and reminding system.
Background
With the rapid development of railways in China, the problem of railway transportation safety is the subject of constant and unchangeable modern traffic, and more attention is paid to the problem. The locomotive crew member plays a central role in the operation of the locomotive, and directly determines whether the train can normally operate, avoids accidents and the like, so that the safety management of the locomotive crew member is important. In the aspect of safety management of locomotive crew members, a standardized requirement is made aiming at an operation standard of a safety key item in the process of taking a crew member by value, a safety node is listed by analyzing a one-time taking standard operation program of the crew member according to a method of system science, for example, serious potential safety hazard problems such as the locomotive crew member driving spirit is not vibration, the visual field is deviated, the crew member sleeps, and the intermittent observation possibly occur in the running of a train, and an alarm prompt is given while the problems occur through technical means, so that the driving state of the crew member is corrected in time, the prevention and control risk is in the early stage, and the driving safety is ensured.
At present, each locomotive crew recording pen and locomotive video monitoring are equipped in each locomotive crew department, a management method is formulated, an analysis team is established, and LKJ (train operation monitoring device) special analysis is combined to monitor the crew to implement daily operation standards. However, the video monitoring and recording pen device records continuously, namely how long and how long the car is driven by the crew, so that the workload of ground audio and video analysis personnel after dumping is huge, each crew section has few crew sections and hundreds of crew sections every day, and more crew members are multiplied on line to generate massive audio and video data, and the number capable of being analyzed by the analysts is simply nine cattle one hair compared with the data amount generated every day, so that full coverage analysis cannot be realized, and the problem is managed afterwards, and the problem is found to constitute a fact, even cause an accident, and cannot be recovered.
The technical problem restricting the popularization and the application of the real-time monitoring technology of the fatigue state of the personnel at present is mainly represented as follows:
1. the change of the ambient illumination in the vehicle is large, the light directly illuminates the face and other interference light sources when meeting in daytime, at night, in backlight, at night, and the change of the posture of the driver and the like when the driver wears sunglasses. It is a technical difficulty in these cases to obtain accurate and clear images of faces, particularly eyes, in real time.
2. Although image recognition techniques have been used in many fields, they basically store images for comparison in advance. When a driver gets on the bus, the driver needs to register in advance, and the system can work normally. This method has a great limitation in application since the drivers may not be the same person. It is a technical difficulty how to accurately locate the face and eyes without image registration in advance, and extract various dynamic parameters of the face in real time and correctly identify the parameters.
3. The real-time requirement of the crew mental state early warning reminding device makes some relatively stable image processing and identification algorithms unavailable, and the algorithm with higher running speed does not reach enough identification rate. How to realize stable and accurate identification under the algorithm with higher running speed is a technical difficulty.
Therefore, a method and a system capable of intelligently identifying and further standardizing the driving standardized operation of the crew in real time on line are very necessary, and especially, the method and the system capable of intelligently identifying and further standardizing the driving standardized operation of the crew in real time on line have the effect of preventing the crew from getting ill in advance.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a crew condition early warning and reminding system. The specific technical scheme of the invention is as follows.
A locomotive crew member working state early warning reminding system is characterized by comprising:
1) the image collector is used for collecting the work image of the crew member;
2) the processor is used for processing and transmitting the images collected by the image collector;
3) and the alarm device gives an alarm according to the early warning information provided by the processor, and implements graded voice alarm.
Furthermore, the image collector adopts a full-shielding structure and comprises a CMOS (complementary metal oxide semiconductor) camera, a rotary adjusting shaft and a collector shell; the CMOS camera comprises a lens and a low-pass narrow-band filter; the rotary adjusting shaft adopts a tooth meshing structure; the collector shell is connected with the cable shielding layer by adopting a double-layer compression joint structure.
Further, the processor is compatible with electricity and magnetism prevention; the connection part of the case is embedded by adopting copper foil, so that the electromagnetic shielding problem at the joint of the case is solved; the signal outgoing line of the processor is provided with a magnetic core inductor, the power supply is provided with double filtering, the anti-jamming capability of the whole machine is improved, the processor adopts an all-metal shell, the heat dissipation adopts the close fit of a machine shell and a heating element, and the whole machine box has the heat dissipation effect.
Further, the processor comprises a circular frequency filter, a neural network algorithm model, an iris recognition instrument, a DSP (digital signal processing) circuit and an FPGA (field programmable gate array) image recognition device.
The graded voice alarm of the alarm device comprises the steps of reminding a crew member on line to keep the crew member operating correctly, and finishing the voice alarm immediately after the abnormal operation phenomenon is eliminated; and meanwhile, the alarm information is stored and transmitted to a management department. The classification comprises primary alarm, secondary alarm, tertiary alarm, driving posture alarm and shielding alarm; the primary alarm means that the visual field of a crew does not continuously focus on the operation front for more than 10s, the secondary alarm means that the visual field of the crew does not continuously focus on the operation front for more than 15s, the tertiary alarm means that the visual field of the crew does not continuously focus on the operation front for more than 20s, the driving posture alarm means that the face of the crew exceeds the range of the image collector for more than 60s, and the blocking alarm means that the image collector is blocked for more than 600 s.
Further, the image collector comprises an active light source and an auxiliary light source; the active light source comprises a laser, and the laser is an infrared laser which generates laser beams with the wavelength of 940nm in a stimulated emission manner under the driving of a power supply and is used as the active light source; the auxiliary light source is provided with a facial PWM (pulse width modulation) driving circuit and comprises an infrared LED array with the wavelength of 850 nm.
Further, the infrared LED array comprises a plurality of infrared LED lamps with the wavelength of 850nm, the infrared LED lamps are uniformly distributed around the lens of the CMOS camera according to the light supplement requirement, and the vertical distance between the brightness position of each infrared LED lamp and the axis of the lens of the CMOS camera is 4-6 cm.
Further, the low-pass narrow-band filter is a film system structure infrared 940nm narrow-band low-pass filter with an optical thin film and is used for filtering all light rays except 940nm in ambient light including sunlight, fluorescent lamps and incandescent lamps; the center wave band of the low-pass narrow-band filter is 940nm, the pass-band pass rate is 70%, the bandwidth is +/-20 nm, and the pass-band positioning accuracy is +/-5 nm.
Further, the locomotive is provided with a GPS (global positioning system) locator, so that the position of the locomotive can be located in real time.
The invention has the beneficial effects that: the invention provides a brand-new unique detection system for solving the driving spirit and lookout states of locomotive crew members, and the system reaches the level of practical application. The invention is based on a high-speed DSP (digital signal processing) embedded processing system with mature technology, adopts advanced video image processing technology, monitors the working state of a locomotive crew member in real time in the whole course of train operation, provides a new technical means for supervision of the working process of the crew member and prevention of fatigue state, can store video data in a certain time period, and has the accident recollection function of driver fatigue state images. The invention uses GPS (satellite positioning system) positioning technology and 4G wireless transmission technology to realize real-time positioning of the locomotive, and can transmit crew value riding state video and remote transmission alarm data in real time, so that a manager can master crew value riding dynamic in real time, and a basis is provided for effective management through accumulation and intelligent analysis of big data.
Detailed Description
The present invention is described in further detail below.
The invention discloses a locomotive crew member working state early warning reminding system.A warning device automatically implements graded voice warning when abnormal operation occurs to a crew member when a locomotive runs, and reminds the crew member on line to keep correct operation, and the voice warning is immediately finished after the abnormal operation phenomenon is eliminated; the abnormal operations include visual field deviation, lassitude, and intermittent lookouts.
Under the condition that the running speed of the locomotive is not 0km/h or a demonstration mode is started, the locomotive crew member working state early warning reminding system starts to run, and the system comprises:
1) the image collector is used for collecting the work image of the crew member;
2) the processor is used for processing and transmitting the images collected by the image collector;
3) and the alarm device gives an alarm according to the early warning information provided by the processor, and implements graded voice alarm.
The image collector adopts a full shielding structure; comprises a CMOS camera, a rotary adjusting shaft and a collector shell; the CMOS camera comprises a lens and a low-pass narrow-band filter; the rotary adjusting shaft adopts a tooth meshing structure; the collector shell is connected with the cable shielding layer by adopting a double-layer compression joint structure.
The processor is compatible with electricity and magnetism prevention; the connection part of the case is embedded by adopting copper foil, so that the electromagnetic shielding problem at the joint of the case is solved; a signal outgoing line of the processor is provided with a magnetic core inductor; the power supply of the processor is a double-filtering power supply, so that the anti-interference capability of the whole processor is improved; the processor adopts an all-metal shell, and has a firm and simple structure; automatic operation and basically maintenance-free; small volume, light weight and convenient installation and use. The processor adopts the tight fit of the casing and the heating element for heat dissipation, and the whole case has a heat dissipation effect.
The processor comprises a circular frequency filter, a neural network algorithm model, an iris recognition instrument, a DSP circuit and an FPGA image recognition device.
The graded voice alarm is set as a first-level alarm, a second-level alarm, a third-level alarm, a driving posture alarm and a shielding alarm, and the voice alarm is immediately ended after the abnormal operation phenomenon is eliminated; the primary alarm means that the visual field of a crew does not continuously focus on the operation front for more than 10s, the secondary alarm means that the visual field of the crew does not continuously focus on the operation front for more than 15s, the tertiary alarm means that the visual field of the crew does not continuously focus on the operation front for more than 20s, the driving posture alarm means that the face of the crew exceeds the range of the image collector for more than 60s, and the blocking alarm means that the image collector is blocked for more than 600 s.
The image collector comprises an active light source and an auxiliary light source; the active light source comprises a laser, the laser is an infrared laser which generates a laser beam with a wavelength of 940nm in a stimulated emission mode under the drive of a power supply and serves as the active light source, the interference of outside light is weakened, a video image with stable brightness is obtained all the day long, accurate and clear facial features are obtained, meanwhile, strong eye iris reflection features are obtained, the eye positioning and recognition process is greatly simplified, the recognition accuracy is improved, and the misjudgment rate is smaller than 1%; the auxiliary light source is provided with a face PWM driving circuit and comprises 20 arrays of infrared LED lamps with the wavelength of 850 nm. The LED lamps are enclosed into two circles according to the requirement of light supplement and are uniformly distributed around the CMOS camera lens, and the vertical distance between the brightness position of the infrared LED lamps and the axis of the CMOS camera lens is 5 cm.
The processor uses three parallel judgment methods for judging the eye state of the crew member, and the method comprises the following steps: iris feature judgment, geometric feature judgment and neural network judgment; the false alarm rate is effectively reduced; firstly, the eyes are positioned by using the interframe correlation, so that the running time of the system is greatly reduced, and the speed of video processing is reached. The system judges the state of human eyes by utilizing a neural network discrimination algorithm, the geometric outline characteristics of the eyes and the iris reflection characteristics of the eyes, analyzes and judges the driving spirit and the observation state of a locomotive attendant by integrating factors such as head position change and the like, and sends out humanized prompt tones to remind the locomotive attendant to keep being concentrated on and observe.
The low-pass narrow-band optical filter of the CMOS camera is a film system structure infrared 940nm narrow-band low-pass optical filter with an optical film and is used for filtering all light rays except 940nm in ambient light including sunlight, fluorescent lamps and incandescent lamps; the center wave band of the low-pass narrow-band filter is 940nm, the pass-band pass rate is 70%, the bandwidth is +/-20 nm, and the pass-band positioning accuracy is +/-5 nm.
The film system of the optical film of the narrow-band low-pass filter is formed by combining a long-wave-pass filter film and a short-wave-pass filter film, and the film material comprises SiO which can obtain a low-refractive-index film layer2Film material and TiO capable of obtaining high-refractive-index film layer2Film material and Si film material. The narrow-band low-pass filter is used for filtering all light rays except 940nm in ambient light including sunlight, fluorescent lamps and incandescent lamps, effectively stopping the light rays from entering the image sensor, remarkably enhancing the performance of resisting ambient light interference, weakening the adverse effect of the ambient light, particularly the ambient light in different periods, on images, and emitting laser light to form a high-brightness uniform surface light sourceAfter the reflection of the face, the light is filtered by the narrow-band low-pass filter to obtain a relatively real face gray image, and the image is a basically consistent perfect image no matter what time period and under the ambient light conditions with different illumination intensities.
The locomotive is provided with a GPS positioner, and the position of the locomotive can be positioned in real time by adopting an wgs84 coordinate system, so that the work of dispatching personnel is facilitated. The monitoring center can play back the history travel and track record of the vehicle in the user-defined period within nearly 60 days at any time; the real-time position, the driving direction and the driving speed of the vehicle can be monitored 24 hours all day long, so that the condition of the vehicle can be mastered most timely; the vehicle can be remotely cut off oil and electricity at any time, and the vehicle can be locked.
Claims (9)
1. A locomotive crew member working state early warning reminding system is characterized by comprising:
1) the image collector is used for collecting the work image of the crew member;
2) the processor is used for processing and transmitting the images collected by the image collector;
3) and the alarm device gives an alarm according to the early warning information provided by the processor, and implements graded voice alarm.
2. The system of claim 1, wherein the image collector adopts a full shielding structure; comprises a CMOS camera, a rotary adjusting shaft and a collector shell; the CMOS camera comprises a lens and a low-pass narrow-band filter; the rotary adjusting shaft adopts a tooth meshing structure; the collector shell is connected with the cable shielding layer by adopting a double-layer compression joint structure.
3. The system of claim 1, wherein the processor is electrically and magnetically compatible; the connection part of the case is embedded by copper foil; a signal outgoing line of the processor is provided with a magnetic core inductor; the power supply of the processor is a double-filtering power supply; the processor adopts an all-metal shell.
4. The system of claim 1, wherein the processor comprises a circular frequency filter, a neural network algorithm model, an eye iris recognition unit, a DSP circuit, and an FPGA image recognition device.
5. The system of claim 1, wherein the graded voice alarm is set to a primary alarm, a secondary alarm, a tertiary alarm, a driving posture alarm and a blocking alarm, and the voice alarm is immediately ended after the abnormal operation phenomenon is eliminated; the primary alarm means that the visual field of a crew does not continuously focus on the operation front for more than 10s, the secondary alarm means that the visual field of the crew does not continuously focus on the operation front for more than 15s, the tertiary alarm means that the visual field of the crew does not continuously focus on the operation front for more than 20s, the driving posture alarm means that the face of the crew exceeds the range of the image collector for more than 60s, and the blocking alarm means that the image collector is blocked for more than 600 s.
6. The system of claim 1, wherein the image collector comprises an active light source and an auxiliary light source; the active light source comprises a laser, and the laser is an infrared laser which generates laser beams with the wavelength of 940nm in a stimulated emission manner under the driving of a power supply and is used as the active light source; the auxiliary light source is provided with a face PWM driving circuit and comprises an infrared LED array with the wavelength of 850 nm.
7. The system according to claim 6, wherein the infrared LED array comprises a plurality of infrared LED lamps with the wavelength of 850nm, the infrared LED lamps are uniformly distributed around the lens of the CMOS camera according to the need of supplementary lighting, and the vertical distance between the brightness positions of the infrared LED lamps and the axis of the lens of the CMOS camera is 4-6 cm.
8. The system of claim 2, wherein the low-pass narrow-band filter is a film system structure infrared 940nm narrow-band low-pass filter with optical thin films, the central wave band is 940nm, the pass-band pass rate is 70%, the bandwidth is +/-20 nm, and the pass-band positioning accuracy is +/-5 nm.
9. The system of claim 1 wherein the locomotive has a GPS locator for locating locomotive location in real time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811470645.6A CN111105597A (en) | 2018-12-04 | 2018-12-04 | Locomotive crew member working state early warning reminding system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811470645.6A CN111105597A (en) | 2018-12-04 | 2018-12-04 | Locomotive crew member working state early warning reminding system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111105597A true CN111105597A (en) | 2020-05-05 |
Family
ID=70420105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811470645.6A Pending CN111105597A (en) | 2018-12-04 | 2018-12-04 | Locomotive crew member working state early warning reminding system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111105597A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201927051U (en) * | 2010-09-30 | 2011-08-10 | 深圳市中控生物识别技术有限公司 | Face identifying device with function of resisting hard light |
| CN102393989A (en) * | 2011-07-28 | 2012-03-28 | 山西智济电子科技有限公司 | Real-time monitoring system of driver working state |
| CN202217366U (en) * | 2011-07-28 | 2012-05-09 | 山西智济电子科技有限公司 | Real-time monitor system of driver working state |
| US20140025257A1 (en) * | 2012-07-23 | 2014-01-23 | Denso Corporation | Drive Assist Apparatus Assisting Driver To Drive Vehicle Based On Degree Of Fatigue Of Driver |
| CN106781284A (en) * | 2016-12-31 | 2017-05-31 | 中国铁道科学研究院电子计算技术研究所 | Trainman's driver fatigue monitor system based on Multi-information acquisition |
| CN206561805U (en) * | 2017-02-22 | 2017-10-17 | 深圳市百智康科技有限公司 | A kind of anti-fatigue-driving car-mounted device |
-
2018
- 2018-12-04 CN CN201811470645.6A patent/CN111105597A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201927051U (en) * | 2010-09-30 | 2011-08-10 | 深圳市中控生物识别技术有限公司 | Face identifying device with function of resisting hard light |
| CN102393989A (en) * | 2011-07-28 | 2012-03-28 | 山西智济电子科技有限公司 | Real-time monitoring system of driver working state |
| CN202217366U (en) * | 2011-07-28 | 2012-05-09 | 山西智济电子科技有限公司 | Real-time monitor system of driver working state |
| US20140025257A1 (en) * | 2012-07-23 | 2014-01-23 | Denso Corporation | Drive Assist Apparatus Assisting Driver To Drive Vehicle Based On Degree Of Fatigue Of Driver |
| CN106781284A (en) * | 2016-12-31 | 2017-05-31 | 中国铁道科学研究院电子计算技术研究所 | Trainman's driver fatigue monitor system based on Multi-information acquisition |
| CN206561805U (en) * | 2017-02-22 | 2017-10-17 | 深圳市百智康科技有限公司 | A kind of anti-fatigue-driving car-mounted device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106251701B (en) | Vehicle rearview monitor and alarm system and method based on rotation zoom multi-cam | |
| CN102393989B (en) | Real-time monitoring system of driver working state | |
| CN103315754B (en) | A kind of fatigue detection method and device | |
| CN111325977A (en) | Tunnel intelligence edge calculation management and control system | |
| CN103116996B (en) | Tunnel illumination and security system | |
| CN105185160B (en) | The pavement detection method that distorted region mobile trend identifies in virtual grid | |
| KR101967610B1 (en) | Multi lane monitoring system that can recognize vehicle velocity and license plate number of multi lane | |
| CN105336105A (en) | Method, intelligent device and system for preventing fatigue driving | |
| CN106205143A (en) | Intelligent bayonet networked system | |
| CN105719486A (en) | Intelligent warning control system for sharp turning vehicle passage on road and method | |
| CN105163014B (en) | Road monitoring device and method | |
| WO2007149629A1 (en) | Intelligent railyard monitoring system | |
| CN104008663A (en) | Intelligent identification, caution and early-warning system for running tracks in low-visibility environment | |
| DE102016119343A1 (en) | Object monitoring with infrared image recording and infrared pulse illumination | |
| CN109348179A (en) | A kind of road monitoring detection system and method based on artificial intelligence | |
| CN212624073U (en) | Safe driving early warning system | |
| CN114360210A (en) | Vehicle fatigue driving early warning system | |
| CN110807929A (en) | Traffic police auxiliary command information analysis and research and judgment system | |
| CN111105597A (en) | Locomotive crew member working state early warning reminding system | |
| EP3857528A1 (en) | Traffic monitoring and evidence collection system | |
| CN110708830B (en) | Intelligent lamp inspection system | |
| CN205354289U (en) | All -weather intelligent alarm system of geological disasters based on multisource video monitoring technique | |
| CN112216106A (en) | Intelligent traffic management system based on BIM technology | |
| CN204810453U (en) | Road video monitor system | |
| CN111105596A (en) | Locomotive crew member working state early warning reminding method |
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: 20200505 |
|
| RJ01 | Rejection of invention patent application after publication |