CN114137634A - Platform end intrusion detection method and system based on radar video fusion - Google Patents

Platform end intrusion detection method and system based on radar video fusion Download PDF

Info

Publication number
CN114137634A
CN114137634A CN202111479482.XA CN202111479482A CN114137634A CN 114137634 A CN114137634 A CN 114137634A CN 202111479482 A CN202111479482 A CN 202111479482A CN 114137634 A CN114137634 A CN 114137634A
Authority
CN
China
Prior art keywords
intrusion
radar
edge computing
intelligent terminal
data
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
Application number
CN202111479482.XA
Other languages
Chinese (zh)
Inventor
刘步荣
周凌峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yi Tai Fei Liu Information Technology LLC
Original Assignee
Yi Tai Fei Liu Information Technology LLC
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yi Tai Fei Liu Information Technology LLC filed Critical Yi Tai Fei Liu Information Technology LLC
Priority to CN202111479482.XA priority Critical patent/CN114137634A/en
Publication of CN114137634A publication Critical patent/CN114137634A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01VGEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V11/00Prospecting or detecting by methods combining techniques covered by two or more of main groups G01V1/00 - G01V9/00

Landscapes

  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Geophysics (AREA)
  • Alarm Systems (AREA)

Abstract

本发明提供一种基于雷达视频融合的站台端头入侵检测方法及系统,通过在前端将视频数据采集和雷达数据采集融合,规避单个传感器工作的弊端和不足,能够不受任何天气、光线环境的影响,即使在浓烟或浓雾的情况下,依然可以可靠的进行人体的检测与跟踪;实现快速发现,快速定位,监测功能不受任何影响,可全天候全方位站台端头入侵,识别效率和准确度均大大超过人工能力;通过对采集到的雷达数据、视频图像序列进行自动分析计算,通过识别各类光照阴影、雨雪天气、飞鸟来识别排除误报,能够准确实时识别出站台端头入侵的异常事件,并发出智能声光报警,帮助车站提升站台旅客安全管理的工作效率和用户服务体验。

Figure 202111479482

The present invention provides a method and system for intrusion detection at a station head based on radar video fusion. By integrating video data collection and radar data collection at the front end, the disadvantages and deficiencies of single sensor operation can be avoided, and the system can be protected from any weather and light environment. Even in the case of thick smoke or fog, the human body can still be detected and tracked reliably; it can realize rapid discovery, rapid positioning, and the monitoring function is not affected in any way. The accuracy greatly exceeds the human ability; through the automatic analysis and calculation of the collected radar data and video image sequences, and by identifying various types of light and shadows, rain and snow weather, and flying birds to identify and eliminate false alarms, the platform terminal can be accurately identified in real time. Abnormal events of intrusion, and intelligent sound and light alarms are issued to help stations improve the work efficiency and user service experience of platform passenger safety management.

Figure 202111479482

Description

Platform end intrusion detection method and system based on radar video fusion
Technical Field
The invention relates to the technical field of radar, in particular to a platform end intrusion detection method and system based on radar video fusion.
Background
At present, although a video camera of an existing railway platform end intrusion alarm system adopting video detection can accurately sense personnel intrusion, due to the optical characteristics of the video camera, the video camera is very easily interfered by the surrounding environment, for example, severe weather such as dense fog, smoke, storm, rain, snow, glare, total darkness and the like or illumination weather conditions can obviously influence the normal work of the video camera, so that the acquired image information result is not accurate enough, the detection accuracy of intelligent events is low, and all-weather detection cannot be supported.
Although the existing railway platform end intrusion alarm system adopting infrared temperature sensing equipment can overcome the problem of weak light at night, the detection distance is generally less than 10 meters, and meanwhile, the system is particularly easy to be interfered by the environment, so that missing reports can be generated when the temperature of a human body is low in winter, and false reports can be easily generated when small animals pass by.
The existing platform end detection system is generally a data isolated island, although each system has more data collection amount, the data application rate is low or even no application exists, the data lacks systematic analysis, and the system can not perform version iteration upgrade optimization according to the used condition.
Disclosure of Invention
In view of the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a method for detecting station end intrusion based on radar video fusion, which is used to solve the problems existing in the conventional system for detecting station end intrusion.
To achieve the above and other related objects, the present invention provides a system for detecting intrusion at a station end based on radar video fusion, comprising: the system comprises a millimeter wave radar, a camera, a network switch, an RFID reader-writer and a solar power supply system; the millimeter wave radar and the camera are both connected with the network switch, and the network switch and the RFID reader-writer are both connected with the solar power supply system;
the millimeter wave radar is used for acquiring radar data of the railway platform and judging whether intruders exist according to the radar data;
the video camera is used for acquiring video data of the railway platform and judging whether intruders exist according to the video data;
the RFID reader-writer is used for verifying whether the intruder is a worker when the millimeter wave radar and the camera output the intruder.
Optionally, the system further comprises an LED display screen, wherein the LED display screen is connected to the solar power supply system and used for flashing an LED when the result output by the RFID reader is a non-worker.
Optionally, the system further comprises a network sound, wherein the network sound is connected with the solar power supply system and used for outputting alarm voice when the result output by the RFID reader-writer is a non-working person.
Optionally, the system further comprises an edge computing intelligent terminal, and the edge computing intelligent terminal is connected with the solar power supply system and used for carrying out data transmission with the cloud platform.
The invention also provides a platform end intrusion detection method based on radar video fusion, which comprises the following steps:
step 1: judging whether the self-checking of the detection system passes or not; if the self-checking is passed, entering the step 2; if the self-test is not passed, repeating the step 1;
step 2: judging whether the millimeter wave radar detects radar data of the intrusion signal; if the radar data of the intrusion signal is detected, entering step 3; if the radar data of the intrusion signal is not detected, repeating the step 2;
and step 3: judging whether the classification attribute of the intrusion signal is human or not by using the millimeter wave radar; if the classification attribute of the intrusion signal is judged to be personnel intrusion, entering step 4; if the classification attribute of the intrusion signal is judged to be a non-personnel condition, entering step 5;
and 4, step 4: collecting and analyzing the collected video data by using a camera, and carrying out intelligent detection on the invasion of the personnel for the second time; if the analysis result output by the camera is that personnel intrusion exists, confirming that the personnel intrusion exists for the second time, and entering the step 6; if the analysis result output by the camera is that no personnel invade exists, the second confirmation is false alarm, and the step 5 is carried out;
and 5: judging whether the false alarm is discarded or not, and re-entering a detection state;
step 6: the RFID reader-writer detects whether the intruder works or not; if the result of the determination is that the staff is available, entering a step 7; if the determination result is not that the staff is available, entering step 8;
and 7: automatically alarming, recording the detection result, and reentering the detection state;
and 8: carrying out intrusion alarm of an end area, carrying out LED flicker by using an LED display screen and outputting prompt alarm voice by using a network sound;
and step 9: judging whether the person to be detected leaves the detection area or not; if the person to be detected leaves the detection area, the initial detection state of the detection system is recovered; and if the person to be detected is still in the detection area, repeating the step 8.
Optionally, the method further comprises:
step 11: continuously monitoring whether the detection system is in a data return window or not in the normal operation process of the detection system; if not, continuously circulating the step 11;
step 12: establishing a data uploading and downloading session link by using the edge computing intelligent terminal and the cloud platform;
step 13: judging whether the test session link of the edge computing intelligent terminal is established successfully or not, and if so, entering step 14; if the establishment fails, judging whether the failure times exceed 3 times, and if the failure times exceed 3 times, giving up the session and entering the step 11; if the failure times are not more than 3, entering step 12, and restarting the uploading and downloading session link;
step 14: and uploading the data to a cloud platform by using the edge computing intelligent terminal to perform model labeling training processing.
Optionally, the method further comprises:
step 21: continuously monitoring whether the detection system is in a data return window or not in the normal operation process of the detection system; if not, continuously looping step 21;
step 22: establishing a data uploading and downloading session link by using the edge computing intelligent terminal and the cloud platform;
step 23: judging whether the test session link of the edge computing intelligent terminal is established successfully or not, and if so, entering step 24; if the establishment fails, judging whether the failure times exceed 3 times, and if the failure times exceed 3 times, giving up the session and entering step 21; if the failure times are not more than 3, entering step 22 and re-initiating the upload and download session link;
step 24: downloading the current version from the cloud platform to the location of the edge computing intelligent terminal;
step 25: verifying and confirming the downloaded version packet by using the edge computing intelligent terminal, and if the verification is successful, entering the step 26; if the verification fails, the upgrade is abandoned;
step 26: carrying out version updating operation by utilizing the edge computing intelligent terminal;
step 27: confirming whether the monitoring system is updated or not by using the edge computing intelligent terminal, and entering step 28 if the monitoring system is updated; if the update fails, go to step 29;
step 28: after the current version is upgraded, the edge computing intelligent terminal is used for operating the current version;
step 29: and returning the version to the previous version, and operating the previous version by using the edge computing intelligent terminal.
As described above, the present invention provides a method and a system for detecting platform end intrusion based on radar video fusion, which have the following advantages: the invention integrates the video data acquisition and the radar data acquisition at the front end, avoids the defects and shortcomings of the work of a single sensor, is not influenced by any weather and light environment, and can still reliably detect and track the human body even under the condition of dense smoke or dense fog; the system realizes quick discovery, quick positioning, the monitoring function is not influenced, all-weather all-round platform end invasion can be realized, the recognition efficiency and the accuracy greatly exceed the manual capability, the artificial intelligence algorithm in the field of computer vision and radar is utilized, under the condition of almost no need of manual intervention, radar data acquired by the system and a video image sequence are automatically analyzed and calculated, false alarms are recognized and eliminated by recognizing various illumination shadows, rain and snow weather and flying birds, abnormal events of platform end invasion can be accurately recognized in real time, an intelligent sound-light alarm is given, meanwhile, on-site early warning prompts and event records are given out in the Web and the page of a comprehensive management and control platform, and the work efficiency and the user service experience of platform passenger safety management are improved. Meanwhile, the invention realizes the periodic acquisition and uploading of data to a training optimization model and the periodic version iterative updating through a side cloud cooperative system architecture, and improves the processing real-time performance and accuracy of the railway platform end intrusion detection system; the problems of bandwidth pressure, event time delay and the like of the high-definition camera on a video private network are solved through a side cloud cooperative framework, so that the operation management efficiency of a station is improved, the service level is improved, and the emergency handling capacity of station emergencies is enhanced. In addition, the invention can detect, track and identify and detect suspicious objects aiming at moving objects in radar data and video images, and carry out intelligent analysis and comparison on the objects on the basis, can identify whether personnel enter a platform end detection area, distinguish working personnel and non-working personnel, can shield some false alarms caused by illumination shadows, rain and snow weather, birds and the like, can accurately identify abnormal events invading the platform end in real time, can generate a customized voice alarm system according to the characteristic content detected by the current event, can push the system to a field area for playing in real time, carries out real-time voice alarm prompt, and pushes control information to flash and alarm of an LED screen of a corresponding area in real time. The invention forms all-weather signal acquisition function complementation with high-definition video detection by adding millimeter wave radar detection on the front-end sensing equipment layer, and can normally work under the conditions of dense fog, smoke dust, storm, rain and snow, glare, complete blackness and the like; in a network layer, a side cloud cooperative system architecture is formed by adding an edge computing intelligent event analysis server and a cloud intelligent event analysis server, occupation of a core network and a backbone transmission network is reduced by local distribution of high-bandwidth services, the utilization rate of a communication network is improved, front-end bandwidth pressure is relieved, processing real-time performance of an intrusion detection system at the end of a railway platform is improved by sinking content and computing capacity, and meanwhile, the detection accuracy of the existing system can be timely improved by training and updating a large number of samples and regular version iteration updating based on an AI technology.
Drawings
Fig. 1 is a schematic hardware configuration diagram of a platform end intrusion detection system based on radar video fusion according to an embodiment;
fig. 2 is a schematic flowchart illustrating a method for detecting station end intrusion based on radar video fusion according to an embodiment of the present invention;
fig. 3 is a schematic flow chart illustrating data uploading performed by the platform end intrusion detection system and the cloud platform according to an embodiment of the present invention;
fig. 4 is a flowchart illustrating a version download performed by the platform end intrusion detection system and the cloud platform according to an embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Referring to fig. 1, the present invention provides a platform end intrusion detection system based on radar video fusion, including: the system comprises a millimeter wave radar, a camera, a network switch, an RFID reader-writer and a solar power supply system; the millimeter wave radar and the camera are both connected with the network switch, and the network switch and the RFID reader-writer are both connected with the solar power supply system;
the millimeter wave radar is used for acquiring radar data of the railway platform and judging whether intruders exist according to the radar data;
the video camera is used for acquiring video data of the railway platform and judging whether intruders exist according to the video data;
the RFID reader-writer is used for verifying whether the intruder is a worker when the millimeter wave radar and the camera output the intruder.
According to the record, the platform end intrusion detection system further comprises an LED display screen, wherein the LED display screen is connected with the solar power supply system and used for carrying out LED flashing when the result output by the RFID reader-writer is a non-working person.
According to the record, the platform end intrusion detection system further comprises a network sound which is connected with the solar power supply system and used for outputting alarm voice when the result output by the RFID reader-writer is a non-working person.
According to the record, the platform end intrusion detection system further comprises an edge computing intelligent terminal, and the edge computing intelligent terminal is connected with the solar power supply system and used for carrying out data transmission with the cloud platform.
Therefore, the platform end intrusion detection edge computing system based on radar video fusion provided by the invention has the advantages that the video data acquisition and the radar data acquisition are fused at the front end, the defects and the defects of the work of a single sensor are avoided, the influence of any weather and light environment can be avoided, and the detection and the tracking of a human body can be reliably carried out even under the condition of dense smoke or dense fog; the system realizes quick discovery, quick positioning, the monitoring function is not influenced, all-weather all-round platform end invasion can be realized, the recognition efficiency and the accuracy greatly exceed the manual capability, the artificial intelligence algorithm in the field of computer vision and radar is utilized, under the condition of almost no need of manual intervention, radar data acquired by the system and a video image sequence are automatically analyzed and calculated, false alarms are recognized and eliminated by recognizing various illumination shadows, rain and snow weather and flying birds, abnormal events of platform end invasion can be accurately recognized in real time, an intelligent sound-light alarm is given, meanwhile, on-site early warning prompts and event records are given out in the Web and the page of a comprehensive management and control platform, and the work efficiency and the user service experience of platform passenger safety management are improved.
Meanwhile, the system realizes the periodic acquisition and uploading of data to a training optimization model and the periodic version iterative updating through a side cloud cooperative system architecture, and improves the processing real-time performance and accuracy of the railway platform end intrusion detection system; the problems of bandwidth pressure, event time delay and the like of the high-definition camera on a video private network are solved through a side cloud cooperative framework, so that the operation management efficiency of a station is improved, the service level is improved, and the emergency handling capacity of station emergencies is enhanced.
As shown in fig. 2, the present invention further provides a method for detecting station end intrusion based on radar video fusion by using any of the above detection systems, the method comprising the following steps:
step 1: judging whether the self-checking of the detection system passes or not; if the self-checking is passed, entering the step 2; if the self-test is not passed, repeating the step 1;
step 2: judging whether the millimeter wave radar detects radar data of the intrusion signal; if the radar data of the intrusion signal is detected, entering step 3; if the radar data of the intrusion signal is not detected, repeating the step 2;
and step 3: judging whether the classification attribute of the intrusion signal is human or not by using the millimeter wave radar; if the classification attribute of the intrusion signal is judged to be personnel intrusion, entering step 4; if the classification attribute of the intrusion signal is judged to be a non-personnel condition, entering step 5;
and 4, step 4: collecting and analyzing the collected video data by using a camera, and carrying out intelligent detection on the invasion of the personnel for the second time; if the analysis result output by the camera is that personnel intrusion exists, confirming that the personnel intrusion exists for the second time, and entering the step 6; if the analysis result output by the camera is that no personnel invade exists, the second confirmation is false alarm, and the step 5 is carried out;
and 5: judging whether the false alarm is discarded or not, and re-entering a detection state;
step 6: the RFID reader-writer detects whether the intruder works or not; if the result of the determination is that the staff is available, entering a step 7; if the determination result is not that the staff is available, entering step 8;
and 7: automatically alarming, recording the detection result, and reentering the detection state;
and 8: carrying out intrusion alarm of an end area, carrying out LED flicker by using an LED display screen and outputting prompt alarm voice by using a network sound;
and step 9: judging whether the person to be detected leaves the detection area or not; if the person to be detected leaves the detection area, the initial detection state of the detection system is recovered; and if the person to be detected is still in the detection area, repeating the step 8.
According to the record, the invention provides a platform end intrusion detection method based on radar video fusion, which is used for detecting, tracking and identifying moving objects in radar data and video images and detecting suspicious objects, intelligently analyzing and comparing the moving objects on the basis, identifying whether personnel enter a platform end detection area or not, distinguishing working personnel and non-working personnel, shielding false alarms caused by illumination shadows, rain and snow weather, flying birds and the like, accurately identifying abnormal events of platform end intrusion in real time, generating a customized voice alarm system according to the characteristic content of current event detection, pushing the system to a field area for playing in real time, carrying out real-time voice alarm prompt, and pushing control information to LED screen flashing alarms of corresponding areas in real time.
In an exemplary embodiment, as shown in fig. 3, the method for detecting station head intrusion further includes:
step 11: continuously monitoring whether the detection system is in a data return window or not in the normal operation process of the detection system; if not, continuously circulating the step 11;
step 12: establishing a data uploading and downloading session link by using the edge computing intelligent terminal and the cloud platform;
step 13: judging whether the test session link of the edge computing intelligent terminal is established successfully or not, and if so, entering step 14; if the establishment fails, judging whether the failure times exceed 3 times, and if the failure times exceed 3 times, giving up the session and entering the step 11; if the failure times are not more than 3, entering step 12, and restarting the uploading and downloading session link;
step 14: and uploading the data to a cloud platform by using the edge computing intelligent terminal to perform model labeling training processing.
In an exemplary embodiment, as shown in fig. 3, the method for detecting station head intrusion further includes:
step 21: continuously monitoring whether the detection system is in a data return window or not in the normal operation process of the detection system; if not, continuously looping step 21;
step 22: establishing a data uploading and downloading session link by using the edge computing intelligent terminal and the cloud platform;
step 23: judging whether the test session link of the edge computing intelligent terminal is established successfully or not, and if so, entering step 24; if the establishment fails, judging whether the failure times exceed 3 times, and if the failure times exceed 3 times, giving up the session and entering step 21; if the failure times are not more than 3, entering step 22 and re-initiating the upload and download session link;
step 24: downloading the current version from the cloud platform to the location of the edge computing intelligent terminal;
step 25: verifying and confirming the downloaded version packet by using the edge computing intelligent terminal, and if the verification is successful, entering the step 26; if the verification fails, the upgrade is abandoned;
step 26: carrying out version updating operation by utilizing the edge computing intelligent terminal;
step 27: confirming whether the monitoring system is updated or not by using the edge computing intelligent terminal, and entering step 28 if the monitoring system is updated; if the update fails, go to step 29;
step 28: after the current version is upgraded, the edge computing intelligent terminal is used for operating the current version;
step 29: and returning the version to the previous version, and operating the previous version by using the edge computing intelligent terminal.
In summary, the invention provides a platform end intrusion detection method and system based on radar video fusion, which avoids the defects and shortcomings of single sensor operation by fusing video data acquisition and radar data acquisition at the front end, can not be affected by any weather and light environment, and can still reliably detect and track human body even under the condition of dense smoke or dense fog; the system realizes quick discovery, quick positioning, the monitoring function is not influenced, all-weather all-round platform end invasion can be realized, the recognition efficiency and the accuracy greatly exceed the manual capability, the artificial intelligence algorithm in the field of computer vision and radar is utilized, under the condition of almost no need of manual intervention, radar data acquired by the system and a video image sequence are automatically analyzed and calculated, false alarms are recognized and eliminated by recognizing various illumination shadows, rain and snow weather and flying birds, abnormal events of platform end invasion can be accurately recognized in real time, an intelligent sound-light alarm is given, meanwhile, on-site early warning prompts and event records are given out in the Web and the page of a comprehensive management and control platform, and the work efficiency and the user service experience of platform passenger safety management are improved. Meanwhile, the invention realizes the periodic acquisition and uploading of data to a training optimization model and the periodic version iterative updating through a side cloud cooperative system architecture, and improves the processing real-time performance and accuracy of the railway platform end intrusion detection system; the problems of bandwidth pressure, event time delay and the like of the high-definition camera on a video private network are solved through a side cloud cooperative framework, so that the operation management efficiency of a station is improved, the service level is improved, and the emergency handling capacity of station emergencies is enhanced. In addition, the invention can detect, track and identify and detect suspicious objects aiming at moving objects in radar data and video images, and carry out intelligent analysis and comparison on the objects on the basis, can identify whether personnel enter a platform end detection area, distinguish working personnel and non-working personnel, can shield some false alarms caused by illumination shadows, rain and snow weather, birds and the like, can accurately identify abnormal events invading the platform end in real time, can generate a customized voice alarm system according to the characteristic content detected by the current event, can push the system to a field area for playing in real time, carries out real-time voice alarm prompt, and pushes control information to flash and alarm of an LED screen of a corresponding area in real time. The invention forms all-weather signal acquisition function complementation with high-definition video detection by adding millimeter wave radar detection on the front-end sensing equipment layer, and can normally work under the conditions of dense fog, smoke dust, storm, rain and snow, glare, complete blackness and the like; in a network layer, a side cloud cooperative system architecture is formed by adding an edge computing intelligent event analysis server and a cloud intelligent event analysis server, occupation of a core network and a backbone transmission network is reduced by local distribution of high-bandwidth services, the utilization rate of a communication network is improved, front-end bandwidth pressure is relieved, processing real-time performance of an intrusion detection system at the end of a railway platform is improved by sinking content and computing capacity, and meanwhile, the detection accuracy of the existing system can be timely improved by training and updating a large number of samples and regular version iteration updating based on an AI technology.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.

Claims (7)

1.一种基于雷达视频融合的站台端头入侵检测系统,其特征在于,包括有:毫米波雷达、摄像机、网络交换机、RFID读写器、太阳能供电系统;其中,所述毫米波雷达、所述摄像机均与所述网络交换机连接,网络交换机、RFID读写器均与所述太阳能供电系统连接;1. a platform end intrusion detection system based on radar video fusion, is characterized in that, comprises: millimeter wave radar, video camera, network switch, RFID reader, solar power supply system; Wherein, described millimeter wave radar, all The cameras are all connected to the network switch, and the network switch and the RFID reader are all connected to the solar power supply system; 所述毫米波雷达用于采集铁路站台的雷达数据,并根据所述雷达数据判断是否存在入侵人员;The millimeter-wave radar is used to collect radar data of the railway platform, and judge whether there is an intruder according to the radar data; 所述摄像机用于采集铁路站台的视频数据,并根据所述视频数据判断是否存在入侵人员;The camera is used to collect video data of the railway platform, and judge whether there is an intruder according to the video data; 所述RFID读写器用于在所述毫米波雷达和所述摄像机均输出存在入侵人员时,验证所述入侵人员是否为工作人员。The RFID reader/writer is used to verify whether the intruder is a worker when both the millimeter-wave radar and the camera output that there is an intruder. 2.根据权利要求1所述的基于雷达视频融合的站台端头入侵检测系统,其特征在于,所述系统还包括LED显示屏,所述LED显示屏与所述太阳能供电系统连接,用于在所述RFID读写器输出的结果为非工作人员时,进行LED闪烁。2 . The platform end intrusion detection system based on radar video fusion according to claim 1 , wherein the system further comprises an LED display screen, and the LED display screen is connected with the solar power supply system and is used for intrusion detection. 2 . When the result output by the RFID reader is non-workers, the LED flashes. 3.根据权利要求1所述的基于雷达视频融合的站台端头入侵检测系统,其特征在于,所述系统还包括网络音响,所述网络音响与所述太阳能供电系统连接,用于在所述RFID读写器输出的结果为非工作人员时,输出告警语音。3. The station head intrusion detection system based on radar video fusion according to claim 1, characterized in that, the system further comprises a network sound, the network sound is connected with the solar power supply system, and is used for When the result output by the RFID reader is non-staff, it outputs an alarm voice. 4.根据权利要求1所述的基于雷达视频融合的站台端头入侵检测系统,其特征在于,所述系统还包括边缘计算智能终端,所述边缘计算智能终端与所述太阳能供电系统连接,用于与云平台进行数据传输。4. The station head intrusion detection system based on radar video fusion according to claim 1, wherein the system further comprises an edge computing intelligent terminal, and the edge computing intelligent terminal is connected with the solar power supply system, using For data transfer with cloud platform. 5.一种应用上述如权利要求1至4中任一所述检测系统的基于雷达视频融合的站台端头入侵检测方法,其特征在于,所述方法包括以下步骤:5. A method for intrusion detection of a station head based on radar video fusion using the above-mentioned detection system according to any one of claims 1 to 4, wherein the method comprises the following steps: 步骤1:判断所述检测系统自检是否通过;若自检通过,则进入步骤2;若自检未通过,则重复步骤1;Step 1: determine whether the self-check of the detection system has passed; if the self-check passes, go to Step 2; if the self-check fails, repeat Step 1; 步骤2:判断毫米波雷达是否检测到入侵信号的雷达数据;若检测到入侵信号的雷达数据,则进入步骤3;若未检测到入侵信号的雷达数据,则重复步骤2;Step 2: determine whether the millimeter-wave radar has detected radar data of the intrusion signal; if the radar data of the intrusion signal is detected, go to step 3; if the radar data of the intrusion signal is not detected, repeat step 2; 步骤3:利用所述毫米波雷达判断所述入侵信号的分类属性是否为人;若判断入侵信号的分类属性是人员入侵,则进入步骤4;若判断所述入侵信号的分类属性是非人员情况,则进入步骤5;Step 3: Use the millimeter-wave radar to determine whether the classification attribute of the intrusion signal is human; if it is determined that the classification attribute of the intrusion signal is a human intrusion, then proceed to step 4; if it is determined that the classification attribute of the intrusion signal is a non-person situation, then Go to step 5; 步骤4:利用摄像头采集并分析所采集的视频数据,进行第二次人员入侵智能检测;若所述摄像头输出的分析结果为存在人员入侵,则第二次确认是人员入侵,并进入步骤6;若所述摄像头输出的分析结果为不存在人员入侵,则第二次确认是误报,并进入步骤5;Step 4: use the camera to collect and analyze the collected video data, and carry out the second intelligent detection of human intrusion; if the analysis result output by the camera is that there is a human intrusion, then confirm that it is a human intrusion for the second time, and enter step 6; If the analysis result output by the camera is that there is no human intrusion, the second confirmation is a false alarm, and the process goes to step 5; 步骤5:判断是否为误报丢弃,并重新进入检测状态;Step 5: Determine whether it is a false positive discard, and re-enter the detection state; 步骤6:RFID读写器检测入侵人员是否工作人员;若确定结果是工作人员,则进入步骤7;若确定结果不是工作人员,则进入步骤8;Step 6: The RFID reader detects whether the intruder is a staff member; if it is determined that the result is a staff member, go to Step 7; if it is determined that the result is not a staff member, then go to Step 8; 步骤7:自动报警,并记录检测结果,重新进入检测状态;Step 7: Automatically alarm, record the test results, and re-enter the test state; 步骤8:进行端头区域入侵告警,并利用LED显示屏进行LED闪烁以及利用网络音响输出提示告警语音;Step 8: Perform the intrusion alarm of the terminal area, and use the LED display screen to perform LED flashing and use the network audio output to prompt the alarm voice; 步骤9:判断待检测人员是否已经离开检测区域;若待检测人员已经离开检测区域,则恢复所述检测系统的初始检测状态;若待检测人员还在检测区域内,则重复步骤8。Step 9: Determine whether the person to be detected has left the detection area; if the person to be detected has left the detection area, restore the initial detection state of the detection system; if the person to be detected is still in the detection area, repeat step 8. 6.根据权利要求5所述的基于雷达视频融合的站台端头入侵检测方法,其特征在于,所述方法还包括:6. The station head intrusion detection method based on radar video fusion according to claim 5, is characterized in that, described method also comprises: 步骤11:在所述检测系统正常运行过程中,持续监测所述检测系统是否在数据回传窗口;如果不在回传窗口,则持续循环步骤11;Step 11: During the normal operation of the detection system, continuously monitor whether the detection system is in the data return window; if it is not in the return window, continue to loop step 11; 步骤12:利用边缘计算智能终端和云平台建立数据上传下载会话链接;Step 12: Use the edge computing intelligent terminal and the cloud platform to establish a data upload and download session link; 步骤13:判断所述边缘计算智能终端的测试会话链接是否建立成功,如果建立成功,则进入步骤14;如果建立失败,则判断失败次数是否超过3次,以及在失败次数超过3次时,放弃本次会话,并进入步骤11;若失败次数未超过3次,则进入步骤12,并重新发起上传下载会话链接;Step 13: judge whether the test session link of the edge computing intelligent terminal is successfully established, if the establishment is successful, then go to step 14; if the establishment fails, then judge whether the number of failures exceeds 3 times, and when the number of failures exceeds 3 times, give up This session, and go to step 11; if the number of failures does not exceed 3 times, go to step 12, and re-initiate the upload and download session link; 步骤14:利用所述边缘计算智能终端将数据上传到云平台进行模型标注训练处理。Step 14: Use the edge computing intelligent terminal to upload the data to the cloud platform for model labeling training processing. 7.根据权利要求5或6所述的基于雷达视频融合的站台端头入侵检测方法,其特征在于,所述方法还包括:7. The station head intrusion detection method based on radar video fusion according to claim 5 or 6, wherein the method further comprises: 步骤21:在所述检测系统正常运行过程中,持续监测所述检测系统是否在数据回传窗口;如果不在回传窗口,则持续循环步骤21;Step 21: During the normal operation of the detection system, continuously monitor whether the detection system is in the data return window; if it is not in the return window, continue to loop step 21; 步骤22:利用边缘计算智能终端和云平台建立数据上传下载会话链接;Step 22: Use the edge computing intelligent terminal and the cloud platform to establish a data upload and download session link; 步骤23:判断所述边缘计算智能终端的测试会话链接是否建立成功,如果建立成功,则进入步骤24;如果建立失败,则判断失败次数是否超过3次,以及在失败次数超过3次时,放弃本次会话,并进入步骤21;若失败次数未超过3次,则进入步骤22,并重新发起上传下载会话链接;Step 23: judge whether the test session link of the edge computing intelligent terminal is successfully established, if the establishment is successful, then go to step 24; if the establishment fails, judge whether the number of failures exceeds 3 times, and when the number of failures exceeds 3 times, give up For this session, go to step 21; if the number of failures does not exceed 3 times, go to step 22, and re-initiate the upload and download session link; 步骤24:将当前版本从云平台下载到所述边缘计算智能终端所在地;Step 24: Download the current version from the cloud platform to the location of the edge computing intelligent terminal; 步骤25:利用所述边缘计算智能终端对下载的版本包进行校验确认,如果校验成功,则进入步骤26;如果校验失败,则放弃升级;Step 25: Use the edge computing intelligent terminal to verify and confirm the downloaded version package, if the verification is successful, go to Step 26; if the verification fails, give up the upgrade; 步骤26:利用所述边缘计算智能终端进行版本更新操作;Step 26: use the edge computing intelligent terminal to perform a version update operation; 步骤27:利用所述边缘计算智能终端确认所述监测系统是否完成更新,若完成更新,则进入步骤28;若更新失败,则进入步骤29;Step 27: Use the edge computing intelligent terminal to confirm whether the monitoring system has completed the update, if the update is completed, go to Step 28; if the update fails, go to Step 29; 步骤28:在当前版本完成升级后,利用所述边缘计算智能终端运行当前版本;Step 28: After the current version is upgraded, use the edge computing intelligent terminal to run the current version; 步骤29:将版本回退至上一个版本,并利用所述边缘计算智能终端运行所述上一个版本。Step 29: Roll back the version to the previous version, and use the edge computing intelligent terminal to run the previous version.
CN202111479482.XA 2021-12-07 2021-12-07 Platform end intrusion detection method and system based on radar video fusion Pending CN114137634A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111479482.XA CN114137634A (en) 2021-12-07 2021-12-07 Platform end intrusion detection method and system based on radar video fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111479482.XA CN114137634A (en) 2021-12-07 2021-12-07 Platform end intrusion detection method and system based on radar video fusion

Publications (1)

Publication Number Publication Date
CN114137634A true CN114137634A (en) 2022-03-04

Family

ID=80384545

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111479482.XA Pending CN114137634A (en) 2021-12-07 2021-12-07 Platform end intrusion detection method and system based on radar video fusion

Country Status (1)

Country Link
CN (1) CN114137634A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI858709B (en) * 2023-05-16 2024-10-11 大鵬科技股份有限公司 Intruder detection system

Citations (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102799454A (en) * 2012-07-13 2012-11-28 中广传播集团有限公司 Air upgrading and uploading method of terminal software
CN104318732A (en) * 2014-10-27 2015-01-28 国网冀北电力有限公司张家口供电公司 Transformer substation field worker monitoring and management system and method based on video analysis and RFID
CN206258991U (en) * 2016-12-23 2017-06-16 四川先导中通光电科技有限公司 It is a kind of that boundary's warning system is enclosed based on millimeter wave
CN107888601A (en) * 2017-11-21 2018-04-06 国云科技股份有限公司 A cloud platform server intelligent detection virus intrusion system and method thereof
CN108833539A (en) * 2018-06-14 2018-11-16 武汉阿迪克电子股份有限公司 Collector, method and system based on edge calculations
CN109360361A (en) * 2018-12-14 2019-02-19 辽宁奇辉电子系统工程有限公司 A kind of railroad platform end intrusion detection early warning system
CN109920185A (en) * 2019-04-16 2019-06-21 中科九度(北京)空间信息技术有限责任公司 One kind merging the mobile mesh calibration method of detection with video data based on millimetre-wave radar
CN110278644A (en) * 2019-06-28 2019-09-24 山东建筑大学 A multi-building lighting monitoring method and system based on edge computing
CN110910183A (en) * 2019-12-04 2020-03-24 山东建筑大学 Rapid processing system and method for smart community meter data based on edge computing
CN211236276U (en) * 2019-12-09 2020-08-11 昆明铁路局集团科学技术研究所 Intelligent monitoring system for railway platform end intrusion
CN112019610A (en) * 2020-08-21 2020-12-01 广东德圳智能技术有限公司 Cloud edge coordination method and system
CN112466077A (en) * 2020-12-09 2021-03-09 山东铁投能源有限公司 Early warning method, device and system
CN112463882A (en) * 2020-11-20 2021-03-09 广东省安心加科技有限公司 Data synchronization method, cloud platform and edge terminal
CN112766192A (en) * 2021-01-25 2021-05-07 北京市地铁运营有限公司地铁运营技术研发中心 Intelligent train monitoring system
CN112887326A (en) * 2021-02-23 2021-06-01 昆明理工大学 Intrusion detection method based on edge cloud cooperation
CN112968885A (en) * 2021-02-02 2021-06-15 中国信息通信研究院 Edge computing platform safety protection method and device
CN113037722A (en) * 2021-02-26 2021-06-25 济南浪潮高新科技投资发展有限公司 Intrusion detection method and device for edge calculation scene
CN213634753U (en) * 2020-12-09 2021-07-06 山东铁投能源有限公司 Early warning device
CN113179190A (en) * 2021-06-29 2021-07-27 深圳智造谷工业互联网创新中心有限公司 Edge controller, edge computing system and configuration method thereof
CN113219454A (en) * 2021-04-22 2021-08-06 九州云(北京)科技发展有限公司 System and method for preventing vehicles and personnel from invading runway based on millimeter wave radar
CN113296163A (en) * 2021-05-24 2021-08-24 深圳绿米联创科技有限公司 Human body detection method, device and system, composite sensor and storage medium
CN113329205A (en) * 2021-04-09 2021-08-31 成都中科创达软件有限公司 Internet of things video data processing system, intelligent retail system, method and device

Patent Citations (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102799454A (en) * 2012-07-13 2012-11-28 中广传播集团有限公司 Air upgrading and uploading method of terminal software
CN104318732A (en) * 2014-10-27 2015-01-28 国网冀北电力有限公司张家口供电公司 Transformer substation field worker monitoring and management system and method based on video analysis and RFID
CN206258991U (en) * 2016-12-23 2017-06-16 四川先导中通光电科技有限公司 It is a kind of that boundary's warning system is enclosed based on millimeter wave
CN107888601A (en) * 2017-11-21 2018-04-06 国云科技股份有限公司 A cloud platform server intelligent detection virus intrusion system and method thereof
CN108833539A (en) * 2018-06-14 2018-11-16 武汉阿迪克电子股份有限公司 Collector, method and system based on edge calculations
CN109360361A (en) * 2018-12-14 2019-02-19 辽宁奇辉电子系统工程有限公司 A kind of railroad platform end intrusion detection early warning system
CN109920185A (en) * 2019-04-16 2019-06-21 中科九度(北京)空间信息技术有限责任公司 One kind merging the mobile mesh calibration method of detection with video data based on millimetre-wave radar
CN110278644A (en) * 2019-06-28 2019-09-24 山东建筑大学 A multi-building lighting monitoring method and system based on edge computing
CN110910183A (en) * 2019-12-04 2020-03-24 山东建筑大学 Rapid processing system and method for smart community meter data based on edge computing
CN211236276U (en) * 2019-12-09 2020-08-11 昆明铁路局集团科学技术研究所 Intelligent monitoring system for railway platform end intrusion
CN112019610A (en) * 2020-08-21 2020-12-01 广东德圳智能技术有限公司 Cloud edge coordination method and system
CN112463882A (en) * 2020-11-20 2021-03-09 广东省安心加科技有限公司 Data synchronization method, cloud platform and edge terminal
CN112466077A (en) * 2020-12-09 2021-03-09 山东铁投能源有限公司 Early warning method, device and system
CN213634753U (en) * 2020-12-09 2021-07-06 山东铁投能源有限公司 Early warning device
CN112766192A (en) * 2021-01-25 2021-05-07 北京市地铁运营有限公司地铁运营技术研发中心 Intelligent train monitoring system
CN112968885A (en) * 2021-02-02 2021-06-15 中国信息通信研究院 Edge computing platform safety protection method and device
CN112887326A (en) * 2021-02-23 2021-06-01 昆明理工大学 Intrusion detection method based on edge cloud cooperation
CN113037722A (en) * 2021-02-26 2021-06-25 济南浪潮高新科技投资发展有限公司 Intrusion detection method and device for edge calculation scene
CN113329205A (en) * 2021-04-09 2021-08-31 成都中科创达软件有限公司 Internet of things video data processing system, intelligent retail system, method and device
CN113219454A (en) * 2021-04-22 2021-08-06 九州云(北京)科技发展有限公司 System and method for preventing vehicles and personnel from invading runway based on millimeter wave radar
CN113296163A (en) * 2021-05-24 2021-08-24 深圳绿米联创科技有限公司 Human body detection method, device and system, composite sensor and storage medium
CN113179190A (en) * 2021-06-29 2021-07-27 深圳智造谷工业互联网创新中心有限公司 Edge controller, edge computing system and configuration method thereof

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI858709B (en) * 2023-05-16 2024-10-11 大鵬科技股份有限公司 Intruder detection system

Similar Documents

Publication Publication Date Title
CN108037770B (en) UAV transmission line inspection system and method based on artificial intelligence
WO2019179024A1 (en) Method for intelligent monitoring of airport runway, application server and computer storage medium
US8614744B2 (en) Area monitoring using prototypical tracks
CN112633057B (en) Intelligent monitoring method for abnormal behavior in bus
CN108319926A (en) A kind of the safety cap wearing detecting system and detection method of building-site
CN105184258A (en) Target tracking method and system and staff behavior analyzing method and system
CN104394361A (en) Pedestrian crossing intelligent monitoring device and detection method
CN112776856A (en) Track foreign matter intrusion monitoring method, device and system and monitoring host equipment
CN108527382A (en) A kind of crusing robot
CN113484858A (en) Intrusion detection method and system
CN116935551A (en) Perimeter intrusion detection method, system, equipment and storage medium
JP2023524623A (en) Method, apparatus, equipment, storage medium and program for detecting traffic anomalies
CN110047272A (en) A kind of intelligent transportation pedestrian behavior monitoring and alarming system based on big data
CN112434827A (en) Safety protection identification unit in 5T fortune dimension
CN116630866B (en) Abnormal event monitoring method, device, equipment and medium for audio-video radar fusion
CN112132048A (en) Community patrol analysis method and system based on computer vision
CN118429952B (en) Intelligent license plate recognition platform for traffic management
CN111612815A (en) Infrared thermal imaging behavior intention analysis method and system
KR20210133827A (en) Apparatus for learning deep learning model and method thereof
CN116311727A (en) Intrusion response method, device, equipment and readable storage medium
CN111627224A (en) Vehicle speed abnormality detection method, device, equipment and storage medium
CN211630273U (en) Intelligent image recognition device for railway environment
CN114137634A (en) Platform end intrusion detection method and system based on radar video fusion
CN116597394A (en) Railway foreign matter intrusion detection system and method based on deep learning
CN112731375A (en) Area detector capable of automatically adjusting sensitivity according to environmental change and method thereof

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: 20220304

RJ01 Rejection of invention patent application after publication