CN117768610A - High-speed railway perimeter intrusion risk monitoring method and system based on multi-target recognition - Google Patents
High-speed railway perimeter intrusion risk monitoring method and system based on multi-target recognition Download PDFInfo
- Publication number
- CN117768610A CN117768610A CN202311572266.9A CN202311572266A CN117768610A CN 117768610 A CN117768610 A CN 117768610A CN 202311572266 A CN202311572266 A CN 202311572266A CN 117768610 A CN117768610 A CN 117768610A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- sub
- region
- intrusion
- perimeter
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 387
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000002159 abnormal effect Effects 0.000 claims abstract description 115
- 230000009545 invasion Effects 0.000 claims abstract description 85
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000000007 visual effect Effects 0.000 claims abstract description 8
- 238000001514 detection method Methods 0.000 claims description 19
- 238000005286 illumination Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000004891 communication Methods 0.000 claims description 8
- 230000001740 anti-invasion Effects 0.000 claims description 6
- 238000012800 visualization Methods 0.000 claims description 6
- 230000003213 activating effect Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 11
- 230000000694 effects Effects 0.000 abstract description 5
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 8
- 238000012502 risk assessment Methods 0.000 description 7
- 230000005856 abnormality Effects 0.000 description 6
- 230000004044 response Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 229910052742 iron Inorganic materials 0.000 description 4
- 230000002093 peripheral effect Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000002265 prevention Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000002329 infrared spectrum Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Landscapes
- Alarm Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a high-speed railway perimeter intrusion risk monitoring method and system based on multi-target identification, belonging to the field of railway security, wherein the method comprises the following steps: distributing a distributed high-speed rail perimeter monitoring network, acquiring a sub-region real-time state through a photographing unit, and acquiring a sub-region state result; when the state result is abnormal, recording the video of the target subarea to obtain an abnormal subarea video stream; transmitting the sub-region state result and the sub-region abnormal video stream to a processing center, and analyzing the sub-region state result and the sub-region abnormal video stream to obtain a terminal monitoring result; and transmitting the terminal monitoring result to a railway control center for visual display. The method and the device solve the technical problems that in the prior art, the perimeter region of the high-speed railway is not comprehensively monitored, the monitoring efficiency is low and the risk judgment is inaccurate, and achieve the technical effects of realizing real-time comprehensive, accurate and efficient monitoring of perimeter invasion risk of the high-speed railway through distributed network monitoring and intelligent analysis processing.
Description
Technical Field
The invention relates to the field of railway security, in particular to a high-speed railway perimeter intrusion risk monitoring method and system based on multi-target identification.
Background
With the rapid development of high-speed railway construction, the high-speed railway is taken as public infrastructure, and the safe operation of the high-speed railway has great significance for civilian life. The important boundary area around the high-speed railway is used as an important protection barrier for stations and lines, and once various risk events such as illegal invasion and the like occur, the operation safety of the high-speed railway is seriously jeopardized. Therefore, how to comprehensively, real-timely and efficiently monitor and prevent various invasion risks in the peripheral area of the high-speed railway is an important link for ensuring the safety of the high-speed railway. At present, the monitoring and precaution means of the perimeter area of the high-speed railway mainly comprise a monitoring camera and manual inspection. The monitoring means cannot conduct omnibearing supervision on the perimeter environment, meanwhile, the monitoring means are prone to overlooking by means of manual judgment, and real-time, efficient and accurate intrusion detection is difficult to achieve.
Disclosure of Invention
The application aims to solve the technical problems of incomplete monitoring, low monitoring efficiency and inaccurate risk judgment of the perimeter region of the high-speed railway by providing the perimeter intrusion risk monitoring method and system of the high-speed railway based on multi-target identification.
In view of the above problems, the application provides a high-speed railway perimeter intrusion risk monitoring method and system based on multi-objective recognition.
In a first aspect of the present disclosure, a method for monitoring perimeter intrusion risk of a high-speed railway based on multi-objective recognition is provided, the method comprising: the distributed high-speed rail perimeter monitoring network is distributed in a target high-speed rail perimeter area, and an initial monitoring state of the target high-speed rail perimeter area is set for the distributed high-speed rail perimeter monitoring network; shooting a first target subarea in the target Gao Tiezhou boundary area through a first photographing unit in a first terminal monitoring node in the distributed high-speed rail perimeter monitoring network to obtain a first subarea real-time state; based on the real-time state of the first sub-region, a first initial monitoring state of the first target sub-region is called in the initial monitoring state; comparing the real-time state of the first subarea with the first initial monitoring state to obtain a first subarea state result; when the state result of the first sub-region is abnormal, a first video recording unit in the first terminal monitoring node is activated, video recording is carried out on the first target sub-region, and a first sub-region abnormal video stream is obtained; transmitting the first sub-region state result and the first sub-region abnormal video stream to a first processing center in the first terminal monitoring node, and analyzing the first sub-region state result and the first sub-region abnormal video stream based on the first processing center to obtain a first terminal monitoring result, wherein the first terminal monitoring result comprises a first invasion main body set, a first invasion route set and a first invasion risk level set; and transmitting the first terminal monitoring result to the railway control center in real time through the wireless ad hoc network, and visually displaying the first terminal monitoring result by the railway control center.
In another aspect of the present disclosure, a high-speed railway perimeter intrusion risk monitoring system based on multi-objective identification is provided, the system comprising: the monitoring network layout module is used for layout the distributed high-speed rail perimeter monitoring network in a target high-speed rail perimeter area, and setting an initial monitoring state of the target high-speed rail perimeter area for the distributed high-speed rail perimeter monitoring network; the real-time state acquisition module is used for shooting a first target subarea in the target Gao Tiezhou boundary area through a first photographing unit in a first terminal monitoring node in the distributed high-speed rail perimeter monitoring network to acquire a first subarea real-time state; the initial state calling module is used for calling a first initial monitoring state of the first target subarea in the initial monitoring state based on the real-time state of the first subarea; the regional state result module is used for comparing the real-time state of the first subregion with the first initial monitoring state to obtain a first subregion state result; the abnormal video stream module is used for activating a first video recording unit in the first terminal monitoring node when the state result of the first sub-region is abnormal, recording the first target sub-region and obtaining a first sub-region abnormal video stream; the terminal monitoring result module is used for transmitting the first sub-region state result and the first sub-region abnormal video stream to a first processing center in the first terminal monitoring node, analyzing the first sub-region state result and the first sub-region abnormal video stream based on the first processing center, and acquiring a first terminal monitoring result, wherein the first terminal monitoring result comprises a first invasion main body set, a first invasion route set and a first invasion risk level set; and the monitoring visualization module is used for transmitting the monitoring result of the first terminal to the railway control center in real time through the wireless ad hoc network, and the railway control center performs visualization display on the monitoring result of the first terminal.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
because the distributed high-speed rail perimeter monitoring network is distributed in the target high-speed rail perimeter area, the network comprises a plurality of terminal monitoring nodes, and the full-direction monitoring coverage of the perimeter area is realized; setting an initial monitoring state of the area for the network as a reference state for judging the abnormality of the area; shooting a sub-region responsible for the first monitoring node by a camera unit of the first monitoring node, and acquiring real-time state information of the sub-region, wherein the real-time state information is a basis for comparison and judgment; based on the real-time state of the subarea, the corresponding initial state is called for comparison, and the deviation between the real-time state and the initial state is judged; if the comparison result is abnormal, a video recording unit of the node is activated to record video of the subarea, an abnormal video stream is obtained, and detailed information of occurrence of the abnormality is provided; the processing center analyzes the state result and the video stream, acquires the monitoring result including an intrusion body, a route, a risk level and the like, and performs intelligent identification and judgment; the monitoring result is transmitted to the control center in real time through the wireless network, centralized management is realized, visual display is performed, quick response is facilitated, the comprehensive, real-time and intelligent monitoring of the invasion risk of the perimeter region of the high-speed railway is realized, the technical problems of incomplete monitoring, low monitoring efficiency and inaccurate risk judgment of the perimeter region of the high-speed railway in the prior art are solved, and the technical effects of real-time comprehensive, accurate and efficient monitoring of the perimeter invasion risk of the high-speed railway through distributed network monitoring and intelligent analysis processing are achieved.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Fig. 1 is a schematic flow chart of a method for monitoring perimeter intrusion risk of a high-speed railway based on multi-objective recognition according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a method for deploying a target intrusion monitoring model in a high-speed railway perimeter intrusion risk monitoring method based on multi-target recognition according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a high-speed railway perimeter intrusion risk monitoring system based on multi-objective recognition according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a monitoring network layout module 11, a real-time state acquisition module 12, an initial state retrieval module 13, a regional state result module 14, an abnormal video stream module 15, a terminal monitoring result module 16 and a monitoring visualization module 17.
Detailed Description
The technical scheme provided by the application has the following overall thought:
the embodiment of the application provides a high-speed railway perimeter intrusion risk monitoring method and system based on multi-target identification. Firstly, a distributed monitoring network is distributed in a target peripheral area, and the network comprises a plurality of intelligent terminal nodes capable of photographing and video recording so as to realize omnibearing monitoring of the whole peripheral area. Then, an initial state is set for each sub-area, and the terminal node is enabled to acquire real-time state information of the corresponding sub-area in real time, and abnormal conditions can be quickly found through comparing and judging the real-time state with the initial state, so that real-time monitoring is realized. Then, once the abnormality is judged, a recording unit of the terminal node is activated to record the abnormal area in detail. And the state judgment result and the video are transmitted into a processing center in the terminal node for deep analysis, so that key information such as an intrusion main body, a route and the like can be accurately obtained, and intelligent monitoring and analysis can be realized. And finally, transmitting the monitoring analysis result to a control center through a wireless network, carrying out centralized management and visual display so that safety personnel can respond quickly, and thus, comprehensive, real-time, accurate and intelligent perimeter intrusion risk monitoring of the high-speed railway is realized.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Example 1
As shown in fig. 1, the embodiment of the application provides a high-speed railway perimeter intrusion risk monitoring method based on multi-target recognition, which is applied to a high-speed railway perimeter intrusion risk monitoring system based on multi-target recognition, wherein the system comprises a distributed high-speed railway perimeter monitoring network, a wireless ad hoc network and a railway control center, the distributed high-speed railway perimeter monitoring network comprises a plurality of terminal monitoring nodes, each terminal monitoring node comprises a photographing unit, a video recording unit and a processing center, and the plurality of terminal monitoring nodes are in communication connection with the railway control center through the wireless ad hoc network.
Specifically, the embodiment of the application discloses a high-speed railway perimeter intrusion risk monitoring method based on multi-target recognition, which is applied to a high-speed railway perimeter intrusion risk monitoring system based on multi-target recognition, wherein the system comprises a distributed high-speed railway perimeter monitoring network, a wireless ad hoc network and a railway control center, and the distributed high-speed railway perimeter monitoring network is in communication connection with the railway control center through the wireless ad hoc network. The distributed high-speed rail perimeter monitoring network is used for monitoring a target high-speed rail perimeter region and comprises a plurality of terminal monitoring nodes arranged in the target high-speed rail perimeter region; the wireless ad hoc network is used for realizing information transmission between a plurality of terminal monitoring nodes and the railway control center; the railway control center is used for receiving the monitoring result information output by the plurality of terminal monitoring nodes and displaying the monitoring result information. The terminal monitoring node comprises a photographing unit, a video recording unit and a processing center; the photographing unit is used for collecting images of the corresponding target subareas of the terminal monitoring nodes; the video recording unit is used for starting video recording of the target subarea when the terminal monitoring node detects an abnormal condition; the processing center is used for analyzing and processing the monitoring information of the terminal monitoring node and outputting a monitoring result.
The perimeter intrusion risk monitoring method for the high-speed railway comprises the following steps:
the distributed high-speed rail perimeter monitoring network is distributed in a target high-speed rail perimeter area, and an initial monitoring state of the target high-speed rail perimeter area is set for the distributed high-speed rail perimeter monitoring network;
in the embodiment of the application, the target high-speed railway perimeter area refers to an area around a high-speed railway line needing intrusion monitoring, and the area is divided into a plurality of target subareas according to the trend of the high-speed railway line. In order to cover the target iron perimeter area to the greatest extent, a plurality of terminal monitoring nodes are arranged in the target area, so that each terminal monitoring node can monitor one target subarea. The camera unit of the terminal monitoring node is used for shooting a still image, the video recording unit is used for capturing a dynamic video, and the processing center analyzes and processes the acquired image and video. And a plurality of terminal monitoring nodes form a distributed high-speed railway perimeter monitoring network, so that the monitoring of the whole target iron perimeter area is realized. In addition, after the distributed high-speed railway perimeter monitoring network is laid, an initial monitoring state of a target iron perimeter area is set for the monitoring network, including related information such as landform, vegetation distribution, obstacle arrangement and the like of the target subarea, and the initial monitoring state is used for subsequent intrusion risk assessment, detection and analysis and lays a foundation for subsequent intrusion monitoring.
Shooting a first target subarea in the target Gao Tiezhou boundary area through a first photographing unit in a first terminal monitoring node in the distributed high-speed rail perimeter monitoring network to obtain a first subarea real-time state;
in the embodiment of the application, after the distributed high-speed rail perimeter monitoring network is distributed, each terminal monitoring node in the monitoring network is started to monitor in real time. And shooting the first target subarea which is responsible for monitoring by starting a first photographing unit in the first terminal monitoring node, and acquiring the image of the first target subarea in real time. The first terminal monitoring node is any one of a plurality of terminal monitoring nodes in the distributed high-speed rail perimeter monitoring network and is responsible for monitoring a first target subarea, the first photographing unit is an image acquisition device arranged on the first terminal monitoring node and is a high-resolution camera and other devices and is responsible for acquiring images in the first target subarea.
The first terminal monitoring node starts the first photographing unit to automatically photograph the first target subarea according to a certain time interval, acquires a real-time image in the subarea, reflects information such as landforms, target object distribution and the like in the first target subarea, forms a real-time state of the first subarea, and is a basis for realizing subsequent intrusion detection and risk assessment.
Based on the real-time state of the first sub-region, a first initial monitoring state of the first target sub-region is called in the initial monitoring state;
in the embodiment of the present application, after the first sub-area real-time state of the first target sub-area is obtained, a first initial monitoring state corresponding to the first target sub-area is called in a preset initial monitoring state based on the first sub-area real-time state. The first initial monitoring state is initial state information corresponding to the first target subarea.
When the obtained real-time state is analyzed later, the real-time state is compared with the initial state to judge whether an abnormal condition occurs in the target subarea. Therefore, after the real-time state of the first sub-region of the first target sub-region is obtained, the corresponding first initial monitoring state is firstly called to prepare for the subsequent comparative analysis. The information contained in the first initial monitoring state is matched with the type and format of the information acquired when the real-time state of the first sub-region is generated, so that the comparability between the two states is ensured, and the abnormal situation is accurately judged.
And the first initial monitoring state corresponding to the real-time state of the first sub-region is called in the initial monitoring state, so that the corresponding association between the real-time state and the initial state is realized, and basic data is provided for subsequent abnormality detection and intrusion risk assessment.
Comparing the real-time state of the first subarea with the first initial monitoring state to obtain a first subarea state result;
in the embodiment of the application, after the first sub-region real-time state and the corresponding first initial monitoring state of the first target sub-region are obtained. Firstly, preprocessing the image corresponding to the real-time state and the first initial monitoring state of the first sub-region, including image denoising, enhancement and the like, so as to improve the image quality. Then, the two images are aligned through image registration, so that the two images correspond to the same scene and view angle, and the image content is aligned. On the basis of image alignment, traversing all pixels of two images, calculating the difference of color or gray values of the two images at the same position, and generating the calculated difference value of each pixel point into a new difference image. Then, threshold segmentation is performed on the difference image based on a difference threshold value preset based on the monitoring sensitivity, and pixels having a difference value higher than the threshold value are determined as abnormal pixels. And then, marking the abnormal region on the original image in the real-time state of the first sub-region according to the communicated region of the abnormal pixel, and outputting a first sub-region state result according to the area of the abnormal region and the pixel duty ratio, wherein the first sub-region state result is normal or abnormal.
Providing a judging basis for judging whether the intrusion exists in the target subarea or not by acquiring a first subarea state result, wherein a normal state result indicates no intrusion risk, and monitoring can be continued; the abnormal state result will start the subsequent abnormal processing flow, such as video recording, risk assessment, etc. By accurately and effectively detecting the abnormal condition of the target subarea, a foundation is laid for subsequent intrusion monitoring and response.
When the state result of the first sub-region is abnormal, a first video recording unit in the first terminal monitoring node is activated, video recording is carried out on the first target sub-region, and a first sub-region abnormal video stream is obtained;
in the embodiment of the application, when the real-time state of the first sub-region and the first initial monitoring state are compared and analyzed to obtain the abnormal state result, it is indicated that the intrusion risk may exist in the first target sub-region corresponding to the real-time state of the first sub-region. To further analyze the risk situation, dynamic anomaly information for the sub-region needs to be obtained.
When detecting that the state result of the first sub-area is abnormal, immediately activating a first video recording unit in a corresponding first terminal monitoring node, starting dynamic video recording on the first target sub-area, and acquiring video information of abnormal conditions, namely abnormal video streams of the first sub-area. The first video unit adopts high-definition digital cameras and other equipment to record dynamically without dead angles; the obtained abnormal video stream of the first sub-region comprises a real-time image sequence, and the whole dynamic process of the possible abnormal condition of the first target sub-region can be recorded.
By acquiring the abnormal video stream of the first subarea, information is provided for locating an abnormal source, judging an intrusion mode, evaluating the severity of an event and the like, so that the comprehensive monitoring of abnormal conditions is ensured, and a basis is provided for intrusion risk monitoring.
Transmitting the first sub-region state result and the first sub-region abnormal video stream to a first processing center in the first terminal monitoring node, and analyzing the first sub-region state result and the first sub-region abnormal video stream based on the first processing center to obtain a first terminal monitoring result, wherein the first terminal monitoring result comprises a first invasion main body set, a first invasion route set and a first invasion risk level set;
in the embodiment of the application, after the first sub-area state result and the first sub-area abnormal video stream are obtained, the first sub-area state result and the first sub-area abnormal video stream are input into a first processing center of a corresponding first terminal monitoring node for comprehensive analysis, so that a detailed intrusion monitoring result is obtained. The first processing center is embedded with a target intrusion monitoring model, and can conduct intelligent analysis processing on the sub-region state result and the sub-region abnormal video stream. The first processing center performs target detection, tracking, behavior analysis and the like on the state result of the first subarea and the abnormal video stream of the first subarea so as to position an intrusion main body, judge an intrusion route and evaluate a risk level, generates a monitoring result of a first terminal monitoring node, and the result comprises a first intrusion main body set, a first intrusion route set and a first intrusion risk level set which respectively reflect intrusion objects, intrusion lines and risk judging results of abnormal video monitoring and provide basis for subsequent decisions.
And transmitting the first terminal monitoring result to the railway control center in real time through the wireless ad hoc network, and visually displaying the first terminal monitoring result by the railway control center.
In the embodiment of the application, after a first processing center of a first terminal monitoring node obtains a first terminal monitoring result, the first terminal monitoring result is transmitted to a railway control center in real time through a wireless ad hoc network. The wireless ad hoc network can enable terminal monitoring nodes in the distributed high-speed rail perimeter monitoring network to be autonomously networked, and an elastic and reliable wireless relay link is established, so that multi-hop wireless transmission of monitoring data is realized. The monitoring results of the first terminal are transmitted to a monitoring station of the railway control center through the wireless ad hoc network, the monitoring station is integrated with an information visualization platform, and the received monitoring results of the terminals can be summarized and integrated to perform vivid visual presentation. The abnormal detection and invasion risk distribution conditions of the peripheral area of the target high-speed railway line can be visually displayed through the visual results, monitoring information is reflected through the patterns, the animations and the like, security personnel can conveniently and rapidly grasp the abnormal conditions and risk dynamics to carry out corresponding decisions, information interconnection sharing between a monitoring network and a control center is realized, the monitoring results can be timely fed back to decision-making parties, and security response efficiency of railway facilities is improved.
Further, the embodiment of the application further includes:
defining a perimeter anti-invasion area for the target high-speed rail perimeter area, defining invasion risk levels based on the perimeter anti-invasion area, and acquiring an area invasion risk distribution map;
dividing the regional invasion risk distribution map based on the monitoring regions of the plurality of terminal monitoring nodes to obtain a subarea invasion risk distribution map corresponding to the plurality of terminal monitoring nodes;
acquiring a first subarea intrusion risk distribution map based on the first terminal monitoring node;
and determining the invasion risk level of the invasion main body according to the first subarea invasion risk distribution map.
In a preferred embodiment, first, an intrusion prevention area of a railway perimeter is defined in a target high-speed railway perimeter area according to factors such as railway facility distribution, topography, traffic conditions, and the like, and the perimeter intrusion prevention area is obtained. The area includes a banded region on either side of the railway line, as well as other sensitive areas connected to or adjacent to the region. And secondly, in the delimited perimeter intrusion prevention area, different intrusion risk levels are divided according to the conditions of the topographic features, monitoring deployment, response time and the like of different positions, and the closer to the railway facilities, the highest corresponding intrusion risk level is. Then, according to the distribution condition of the invasion risk grades, an area invasion risk distribution map is formed, invasion risk grades of different positions in the invasion preventing area of the perimeter are visually displayed, a foundation is provided for refining the invasion risk grades according to the geographic positions, and analysis refinement of risks is achieved.
After the overall intrusion risk profile of the target iron perimeter area is obtained, the intrusion risk profile is subdivided into each terminal monitoring node to guide risk analysis of each node. Firstly, reasonably planning a monitoring subarea of each terminal monitoring node in a target high-speed rail perimeter area according to the positions and monitoring ranges of a plurality of arranged terminal monitoring nodes; on the regional intrusion risk distribution map, dividing according to the monitoring subareas of the terminal monitoring nodes, enabling each subarea to correspond to one terminal monitoring node, and extracting intrusion risk level information in the subareas. In this way, each terminal monitoring node obtains a corresponding sub-area intrusion risk distribution diagram, and the diagram contains intrusion risk level information of each position in the sub-area in charge of monitoring by the node. According to the subarea intrusion risk distribution diagram, each terminal monitoring node can clearly know intrusion risk conditions of different positions in the monitoring area of the terminal monitoring node, and a basis is provided for intrusion risk judgment of each node. By subdividing the corresponding region information, the refinement of monitoring network risk analysis is realized, and the pertinence and the effectiveness of protection are improved.
After the regional invasion risk distribution map is divided, each terminal monitoring node obtains a corresponding regional invasion risk distribution map. Taking a first terminal monitoring node as an example, extracting the sub-region corresponding to the first terminal monitoring node in the region intrusion risk distribution map according to the specific position of the node in the target high-speed rail perimeter region, and acquiring the first sub-region intrusion risk distribution map. The subarea intrusion risk distribution map comprises intrusion risk levels of each geographical position point in the subarea which is responsible for monitoring by the first terminal monitoring node, and areas with different risk levels have clear boundaries. After the first terminal monitoring node acquires the intrusion risk distribution map of the corresponding subarea, when the first terminal monitoring node detects an intrusion main body through the first subarea in real-time state, the position coordinates of the main body can be determined, and on the first subarea intrusion risk distribution map, according to the position coordinates of the intrusion main body, the area where the first terminal monitoring node is located is judged to belong to which risk level subarea in the risk level distribution map. Thus, the invasion risk level of the invasion main body is the level of the risk subarea, so that the invasion risk level of the invasion main body is accurately obtained. When the first terminal monitoring node monitors a plurality of invasion main bodies, a first invasion main body set is formed, each invasion main body in the first invasion main body set has a corresponding invasion risk level, and a first invasion risk level set is obtained. The first intrusion risk level set dynamically changes according to real-time position coordinates of each intrusion body in the first intrusion body set.
Further, as shown in fig. 2, the embodiment of the present application further includes:
collecting high-speed rail perimeter historical image data, marking an invasive subject in the high-speed rail perimeter historical image, and obtaining a perimeter invasive image data set;
based on a YOLOv3 code frame, loading pre-training weights, and training a model by using a perimeter intrusion image data set to obtain an initial intrusion monitoring model;
an initial intrusion monitoring model is exported and converted into a node deployment format, and the initial intrusion monitoring model is optimized based on the node deployment format to obtain a target intrusion monitoring model;
and deploying the target intrusion monitoring model in the processing centers of the plurality of terminal monitoring nodes, and establishing wired communication connection with the photographing unit and the video recording unit.
In a preferred embodiment, first, high-speed rail perimeter historical image data is obtained by collecting monitoring images and video data archived in high-speed rail perimeter area history, and selecting high-speed rail perimeter historical images which clearly contain intrusion targets therein, wherein the images cover various scenes of the perimeter area. Secondly, by using image marking software, the professional mark staff accurately marks the position of an invasion target in the high-speed railway perimeter historical images, such as invaders, vehicles, explosives and the like, so as to obtain a large number of high-speed railway perimeter images with the invasion target marks, and form a perimeter invasion image data set. Then, YOLOv3 is adopted as a model frame, and pre-training weights are loaded on the basis of the YOLOv3 frame, namely model weight parameters which are trained in advance on a large-scale image dataset are used for initializing a model, so that the model can be quickly converged. And then, using the generated perimeter intrusion image data set as a new data set, and performing model training based on a migration learning strategy to obtain an initial intrusion monitoring model suitable for high-speed railway perimeter intrusion monitoring.
Then, in order to make the initial intrusion monitoring model obtained by training run in the terminal monitoring node with high efficiency, the initial intrusion monitoring model is firstly exported into a general format, such as ONNX format, and then the general model is converted into a model format suitable for deployment in the terminal monitoring node, such as TensorRT model, according to the hardware requirement of the terminal monitoring node. After an initial intrusion monitoring model in a node deployment format is obtained, model compression, quantification and other optimization technologies are utilized, model structures are simplified, model sizes are reduced, optimization processing is carried out on the models, and a target intrusion monitoring model with higher operation efficiency is generated. After the target intrusion detection model optimized for node deployment is obtained, the target intrusion detection model is deployed into a plurality of actual terminal monitoring nodes, the target intrusion detection model is loaded into a processing center of the plurality of terminal monitoring nodes in the distributed high-speed rail perimeter monitoring network, and the processing center is accessed to realize efficient operation of the model. In addition, the processing center is ensured to be in wired communication connection with the camera unit and the video recording unit of the terminal monitoring node, so that the sub-area state result and the sub-area abnormal video stream are transmitted to the processing center in real time, and intrusion judgment is carried out.
Further, the embodiment of the application further includes:
after the first sub-region state result and the first sub-region abnormal video stream are transmitted to a first processing center in the first terminal monitoring node, a first sub-region real-time image corresponding to the first sub-region real-time state is extracted from the first sub-region state result;
inputting the first sub-region real-time image into the target intrusion monitoring model to obtain a first intrusion main body set, wherein different intrusion main bodies in the first intrusion main body set have main body unique identifiers and initial coordinate points, and the initial coordinate points have coordinate point time identifiers;
and adding the first invasion main body set to the first terminal monitoring result.
In a preferred embodiment, after the first processing center receives the first sub-area state result and the first sub-area abnormal video stream, firstly, a static picture image corresponding to the first sub-area real-time state, that is, a first sub-area real-time image, is extracted from the first sub-area state result, so as to perform intrusion target detection, and provide static picture data support for subsequent intrusion identification and monitoring. Then, inputting the real-time image of the first subarea into a target intrusion detection model which is pre-deployed in a terminal monitoring node, and identifying each intrusion instance existing in the image, such as an illegal intruder, a vehicle and the like; for each identified intrusion instance, the model generates a main body unique identifier, and simultaneously gives the coordinate position of the instance as a starting coordinate point and the image acquisition time corresponding to the coordinate position as a coordinate point time identifier. And integrating information of all the detected intrusion bodies and the unique identification of the intrusion bodies, the initial coordinate point and the time identification of the coordinate point to form a first intrusion body set, and completing monitoring and marking of the intrusion bodies to obtain key initial information of the intrusion bodies so as to lay a foundation for subsequent behavior analysis. And then, adding the first invasive main set into a first terminal monitoring result, and laying a foundation for subsequent tracking identification and risk analysis.
Further, the embodiment of the application further includes:
after the first invasion main body set is obtained, carrying out image frame uniform interval extraction on the first subarea abnormal video stream to obtain a first abnormal image frame set, wherein each abnormal image frame in the first abnormal image frame set is provided with an image frame time mark;
sequentially inputting the abnormal image frames in the first abnormal image frame set into the target intrusion monitoring model according to a time sequence based on the image frame time identification, and sequentially outputting a first tracking monitoring result;
the first tracking monitoring result comprises moving coordinate points of different invasion bodies, wherein the moving coordinate points are provided with coordinate point time marks, and the coordinate point time marks are consistent with the image frame time marks.
In a preferred embodiment, after the first intrusion body set is acquired, a uniform image extraction is performed on the first sub-region abnormal video stream acquired by the first target sub-region, that is, a single frame image is extracted from the first sub-region abnormal video stream at a fixed time interval, and the selected time interval is determined according to the video frame rate and the actual requirement. And extracting to form a first abnormal image frame set containing multi-frame images, wherein the first abnormal image frame set is used for inputting a target intrusion monitoring model to track an intrusion target. Meanwhile, when each frame of abnormal image is extracted, the time identifier corresponding to the frame is carried, so that each frame of image in the first abnormal image frame set has the corresponding image frame time identifier.
Then, after a first abnormal image frame set containing the image frame time identifier is obtained, abnormal image frames in the image frame set are input into a target intrusion monitoring model one by one according to the time identifier sequence, and the model can identify new coordinate positions of different intrusion bodies in each frame. The frame images are sequentially processed, the model analyzes the moving track of the invasive main body at different times, the tracking and monitoring of the invasive main body are realized, a first tracking and monitoring result is output, and the result contains the coordinate position of each invasive target at different frames and has the coordinate point time identifier of the coordinate, and the coordinate point time identifier is consistent with the corresponding image frame time identifier. The coordinate movement information of each invasive body in the first subarea abnormal video stream is obtained, and information is provided for completing route tracking identification of the invasive body.
Further, the embodiment of the application further includes:
after a first tracking monitoring result is obtained, matching the first invasion main body set with the same invasion main body in the first tracking monitoring result based on the main body unique identifier, and obtaining an invasion coordinate point set corresponding to the same invasion main body;
based on the intrusion coordinate point set, taking the same intrusion main body as a fitting main body, taking the initial coordinate point as an intrusion starting point, taking the moving coordinate point as an intrusion moving point, and performing curve fitting according to the coordinate point time mark to obtain a first intrusion route set;
And adding the first intrusion route set to the first terminal monitoring result.
In one possible implementation, based on the image frame time identifier, the abnormal image frames in the first abnormal image frame set are sequentially input into the target intrusion monitoring model according to the time sequence, and each abnormal image frame can obtain a corresponding first tracking monitoring result. After the first tracking monitoring results are obtained, the first invasion main body set is matched with the same invasion main body in the first tracking monitoring results according to the main body unique identification, an invasion coordinate point set corresponding to each invasion main body is obtained, the initial coordinate point of the invasion main body set and the corresponding tracking coordinate points in each first tracking monitoring result are contained, and a data base is provided for invasion route generation.
And then, acquiring an intrusion route of each intrusion body on the basis of the previous intrusion coordinate point set to form a dynamic first intrusion route set every time a first tracking monitoring result is obtained. Specifically, an intrusion main body is set as a fitting main body, an initial coordinate point of the obtained intrusion main body is used as an intrusion starting point, a tracking coordinate point in a first tracking monitoring result is used as an intrusion moving point, and an intrusion route from the starting point to each moving point is generated by adopting a curve fitting mode according to a coordinate point time identifier corresponding to each coordinate. Finally, each invasive subject fits an invasive route representing the movement track of the invasive subject, and the invasive route set of all the invasive subjects forms a first invasive route set. The generation of the intrusion route is dynamic synchronous generation, namely, the first subarea abnormal video stream is synchronously transmitted to a first processing center for image frame extraction while the first subarea abnormal video stream is recorded, meanwhile, the image frames are input into a target intrusion monitoring model for identification, a first tracking monitoring result is obtained, and then curve fitting is carried out according to a moving coordinate point in the first tracking monitoring result, so that real-time response to intrusion is realized.
Further, the embodiment of the application further includes:
the first terminal monitoring node further comprises a first infrared detection unit;
when the state result of the first sub-area is abnormal and the real-time illumination intensity of the first target sub-area is lower than the preset illumination intensity, a first infrared video recording unit in the first terminal monitoring node is activated to record the first target sub-area, and a first sub-area abnormal video stream is obtained.
In a preferred embodiment, to enhance the monitoring capability of the system at night or under poor lighting conditions, the first terminal monitoring node further includes a first infrared detection unit in an infrared band in addition to the video recording unit in a visible light band. The infrared detection unit consists of equipment such as an infrared camera and the like, and can sense and collect image and video information in an infrared band.
When the state result of the first sub-area detects that the first target sub-area has abnormal conditions, if the real-time illumination intensity of the target sub-area is lower than the preset illumination intensity, the condition that the environment light of the target sub-area is dark and weak belongs to night or dark scenes is indicated, and the video recording unit which is not suitable for continuously using visible light is used for collecting information. In this case, a first infrared detection unit in the first terminal monitoring node is started, and video recording under infrared spectrum is performed on the first target sub-area. The first infrared detection unit can acquire thermal imaging information and video data in the target subarea under the condition of weak light or no light, acquire the performance of abnormal conditions under the infrared band, generate a first infrared abnormal video stream as the first subarea abnormal video stream, improve the capability of the system for processing complex illumination conditions, and enhance the all-weather monitoring effect.
In summary, the high-speed railway perimeter intrusion risk monitoring method based on multi-target recognition provided by the embodiment of the application has the following technical effects:
distributing a distributed high-speed rail perimeter monitoring network in the target high-speed rail perimeter area, setting an initial monitoring state of the target high-speed rail perimeter area for the distributed high-speed rail perimeter monitoring network, and realizing comprehensive monitoring coverage of the whole perimeter area. Shooting a first target subarea in a target high-speed rail perimeter area through a first photographing unit in a first terminal monitoring node in the distributed high-speed rail perimeter monitoring network, acquiring a real-time state of the first subarea, and providing a reference for subsequent abnormality judgment. Based on the real-time state of the first sub-region, a first initial monitoring state of the first target sub-region is called in the initial monitoring state; and comparing the real-time state of the first subarea with the first initial monitoring state, acquiring a state result of the first subarea, and judging whether the current condition is abnormal or not. When the state result of the first subarea is abnormal, a first video recording unit in a first terminal monitoring node is activated, video recording is carried out on the first target subarea, abnormal video streams of the first subarea are obtained, and data support for analyzing the abnormality is provided. Transmitting the first sub-region state result and the first sub-region abnormal video stream to a first processing center in a first terminal monitoring node, analyzing the first sub-region state result and the first sub-region abnormal video stream based on the first processing center, acquiring a first terminal monitoring result, and realizing intelligent monitoring; the first terminal monitoring result is transmitted to the railway control center in real time through the wireless ad hoc network, and the railway control center performs visual display on the first terminal monitoring result, so that comprehensive, real-time and accurate intelligent monitoring of the invasion risk of the perimeter region of the high-speed rail is realized.
Example two
Based on the same inventive concept as the high-speed railway perimeter intrusion risk monitoring method based on multi-objective recognition in the foregoing embodiments, as shown in fig. 3, the present embodiment provides a high-speed railway perimeter intrusion risk monitoring system based on multi-objective recognition, where the system includes a distributed high-speed railway perimeter monitoring network, a wireless ad hoc network, and a railway control center, the distributed high-speed railway perimeter monitoring network includes a plurality of terminal monitoring nodes, the terminal monitoring nodes include a camera unit, a video recording unit, and a processing center, and the plurality of terminal monitoring nodes are communicatively connected with the railway control center through the wireless ad hoc network, and the method includes:
the monitoring network layout module 11 is used for layout the distributed high-speed rail perimeter monitoring network in a target high-speed rail perimeter area, and setting an initial monitoring state of the target high-speed rail perimeter area for the distributed high-speed rail perimeter monitoring network;
the real-time state acquisition module 12 is configured to acquire a real-time state of a first sub-region by shooting the first target sub-region in the target Gao Tiezhou boundary region through a first camera unit in a first terminal monitoring node in the distributed high-speed rail perimeter monitoring network;
an initial state retrieving module 13, configured to retrieve, in the initial monitoring state, a first initial monitoring state of the first target sub-region based on the first sub-region real-time state;
The area state result module 14 is configured to compare the real-time state of the first sub-area with the first initial monitoring state, and obtain a first sub-area state result;
the abnormal video stream module 15 is configured to activate a first video recording unit in the first terminal monitoring node when the status result of the first sub-area is abnormal, record the first target sub-area, and obtain a first sub-area abnormal video stream;
the terminal monitoring result module 16 is configured to transmit the first sub-region state result and the first sub-region abnormal video stream to a first processing center in the first terminal monitoring node, and analyze the first sub-region state result and the first sub-region abnormal video stream based on the first processing center to obtain a first terminal monitoring result, where the first terminal monitoring result includes a first intrusion main body set, a first intrusion route set, and a first intrusion risk level set;
the monitoring visualization module 17 is configured to transmit the first terminal monitoring result to the railway control center in real time through the wireless ad hoc network, and the railway control center performs visual display on the first terminal monitoring result.
Further, the embodiment of the application further includes a risk classification module, where the risk classification module includes the following execution steps:
defining a perimeter anti-invasion area for the target high-speed rail perimeter area, defining invasion risk levels based on the perimeter anti-invasion area, and acquiring an area invasion risk distribution map;
dividing the regional invasion risk distribution map based on the monitoring regions of the plurality of terminal monitoring nodes to obtain a subarea invasion risk distribution map corresponding to the plurality of terminal monitoring nodes;
acquiring a first subarea intrusion risk distribution map based on the first terminal monitoring node;
and determining the invasion risk level of the invasion main body according to the first subarea invasion risk distribution map.
Further, the terminal monitoring result module 16 includes the following steps:
collecting high-speed rail perimeter historical image data, marking an invasive subject in the high-speed rail perimeter historical image, and obtaining a perimeter invasive image data set;
based on a YOLOv3 code frame, loading pre-training weights, and training a model by using a perimeter intrusion image data set to obtain an initial intrusion monitoring model;
an initial intrusion monitoring model is exported and converted into a node deployment format, and the initial intrusion monitoring model is optimized based on the node deployment format to obtain a target intrusion monitoring model;
And deploying the target intrusion monitoring model in the processing centers of the plurality of terminal monitoring nodes, and establishing wired communication connection with the photographing unit and the video recording unit.
Further, the terminal monitoring result module 16 further includes the following steps:
after the first sub-region state result and the first sub-region abnormal video stream are transmitted to a first processing center in the first terminal monitoring node, a first sub-region real-time image corresponding to the first sub-region real-time state is extracted from the first sub-region state result;
inputting the first sub-region real-time image into the target intrusion monitoring model to obtain a first intrusion main body set, wherein different intrusion main bodies in the first intrusion main body set have main body unique identifiers and initial coordinate points, and the initial coordinate points have coordinate point time identifiers;
and adding the first invasion main body set to the first terminal monitoring result.
Further, the terminal monitoring result module 16 further includes the following steps:
after the first invasion main body set is obtained, carrying out image frame uniform interval extraction on the first subarea abnormal video stream to obtain a first abnormal image frame set, wherein each abnormal image frame in the first abnormal image frame set is provided with an image frame time mark;
Sequentially inputting the abnormal image frames in the first abnormal image frame set into the target intrusion monitoring model according to a time sequence based on the image frame time identification, and sequentially outputting a first tracking monitoring result;
the first tracking monitoring result comprises moving coordinate points of different invasion bodies, wherein the moving coordinate points are provided with coordinate point time marks, and the coordinate point time marks are consistent with the image frame time marks.
Further, the terminal monitoring result module 16 further includes the following steps:
after a first tracking monitoring result is obtained, matching the first invasion main body set with the same invasion main body in the first tracking monitoring result based on the main body unique identifier, and obtaining an invasion coordinate point set corresponding to the same invasion main body;
based on the intrusion coordinate point set, taking the same intrusion main body as a fitting main body, taking the initial coordinate point as an intrusion starting point, taking the moving coordinate point as an intrusion moving point, and performing curve fitting according to the coordinate point time mark to obtain a first intrusion route set;
and adding the first intrusion route set to the first terminal monitoring result.
Further, the embodiment of the application further comprises an infrared detection module, and the module comprises the following implementation steps:
The first terminal monitoring node further comprises a first infrared detection unit;
when the state result of the first sub-area is abnormal and the real-time illumination intensity of the first target sub-area is lower than the preset illumination intensity, a first infrared video recording unit in the first terminal monitoring node is activated to record the first target sub-area, and a first sub-area abnormal video stream is obtained.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any of the methods to implement embodiments of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (8)
1. The method is applied to a high-speed railway perimeter intrusion risk monitoring system based on multi-target recognition, the system comprises a distributed high-speed railway perimeter monitoring network, a wireless ad hoc network and a railway control center, the distributed high-speed railway perimeter monitoring network comprises a plurality of terminal monitoring nodes, the terminal monitoring nodes comprise a photographing unit, a video recording unit and a processing center, and the terminal monitoring nodes are in communication connection with the railway control center through the wireless ad hoc network, and the method comprises the following steps:
the distributed high-speed rail perimeter monitoring network is distributed in a target high-speed rail perimeter area, and an initial monitoring state of the target high-speed rail perimeter area is set for the distributed high-speed rail perimeter monitoring network;
shooting a first target subarea in the target Gao Tiezhou boundary area through a first photographing unit in a first terminal monitoring node in the distributed high-speed rail perimeter monitoring network to obtain a first subarea real-time state;
based on the real-time state of the first sub-region, a first initial monitoring state of the first target sub-region is called in the initial monitoring state;
Comparing the real-time state of the first subarea with the first initial monitoring state to obtain a first subarea state result;
when the state result of the first sub-region is abnormal, a first video recording unit in the first terminal monitoring node is activated, video recording is carried out on the first target sub-region, and a first sub-region abnormal video stream is obtained;
transmitting the first sub-region state result and the first sub-region abnormal video stream to a first processing center in the first terminal monitoring node, and analyzing the first sub-region state result and the first sub-region abnormal video stream based on the first processing center to obtain a first terminal monitoring result, wherein the first terminal monitoring result comprises a first invasion main body set, a first invasion route set and a first invasion risk level set;
and transmitting the first terminal monitoring result to the railway control center in real time through the wireless ad hoc network, and visually displaying the first terminal monitoring result by the railway control center.
2. The method according to claim 1, wherein the method further comprises:
defining a perimeter anti-invasion area for the target high-speed rail perimeter area, defining invasion risk levels based on the perimeter anti-invasion area, and acquiring an area invasion risk distribution map;
Dividing the regional invasion risk distribution map based on the monitoring regions of the plurality of terminal monitoring nodes to obtain a subarea invasion risk distribution map corresponding to the plurality of terminal monitoring nodes;
acquiring a first subarea intrusion risk distribution map based on the first terminal monitoring node;
and determining the invasion risk level of the invasion main body according to the first subarea invasion risk distribution map.
3. The method according to claim 1, characterized in that the method comprises:
collecting high-speed rail perimeter historical image data, marking an invasive subject in the high-speed rail perimeter historical image, and obtaining a perimeter invasive image data set;
based on a YOLOv3 code frame, loading pre-training weights, and training a model by using a perimeter intrusion image data set to obtain an initial intrusion monitoring model;
an initial intrusion monitoring model is exported and converted into a node deployment format, and the initial intrusion monitoring model is optimized based on the node deployment format to obtain a target intrusion monitoring model;
and deploying the target intrusion monitoring model in the processing centers of the plurality of terminal monitoring nodes, and establishing wired communication connection with the photographing unit and the video recording unit.
4. A method according to claim 3, wherein after transmitting the first sub-region status result and the first sub-region abnormal video stream to a first processing center in the first terminal monitoring node, the method comprises:
extracting a first sub-region real-time image corresponding to the first sub-region real-time state from the first sub-region state result;
inputting the first sub-region real-time image into the target intrusion monitoring model to obtain a first intrusion main body set, wherein different intrusion main bodies in the first intrusion main body set have main body unique identifiers and initial coordinate points, and the initial coordinate points have coordinate point time identifiers;
and adding the first invasion main body set to the first terminal monitoring result.
5. The method of claim 4, wherein after acquiring the first set of intrusion bodies, the method comprises:
extracting the abnormal video stream of the first subarea at uniform intervals of image frames to obtain a first abnormal image frame set, wherein each abnormal image frame in the first abnormal image frame set is provided with an image frame time mark;
sequentially inputting the abnormal image frames in the first abnormal image frame set into the target intrusion monitoring model according to a time sequence based on the image frame time identification, and sequentially outputting a first tracking monitoring result;
The first tracking monitoring result comprises moving coordinate points of different invasion bodies, wherein the moving coordinate points are provided with coordinate point time marks, and the coordinate point time marks are consistent with the image frame time marks.
6. The method of claim 5, wherein after obtaining the first tracking monitoring result, the method comprises:
matching the first invasive main body set with the same invasive main body in the first tracking monitoring result based on the main body unique identifier to obtain an invasive coordinate point set corresponding to the same invasive main body;
based on the intrusion coordinate point set, taking the same intrusion main body as a fitting main body, taking the initial coordinate point as an intrusion starting point, taking the moving coordinate point as an intrusion moving point, and performing curve fitting according to the coordinate point time mark to obtain a first intrusion route set;
and adding the first intrusion route set to the first terminal monitoring result.
7. The method according to claim 1, wherein the method further comprises:
the first terminal monitoring node further comprises a first infrared detection unit;
when the state result of the first sub-area is abnormal and the real-time illumination intensity of the first target sub-area is lower than the preset illumination intensity, a first infrared video recording unit in the first terminal monitoring node is activated to record the first target sub-area, and a first sub-area abnormal video stream is obtained.
8. A high-speed railway perimeter intrusion risk monitoring system based on multi-target recognition, which is used for implementing the high-speed railway perimeter intrusion risk monitoring method based on multi-target recognition according to any one of claims 1 to 7, wherein the system comprises a distributed high-speed railway perimeter monitoring network, a wireless ad hoc network and a railway control center, the distributed high-speed railway perimeter monitoring network comprises a plurality of terminal monitoring nodes, the terminal monitoring nodes comprise a photographing unit, a video recording unit and a processing center, and the terminal monitoring nodes are in communication connection with the railway control center through the wireless ad hoc network, and the system comprises:
the monitoring network layout module is used for layout the distributed high-speed rail perimeter monitoring network in a target high-speed rail perimeter area, and setting an initial monitoring state of the target high-speed rail perimeter area for the distributed high-speed rail perimeter monitoring network;
the real-time state acquisition module is used for shooting a first target sub-region in the target Gao Tiezhou boundary region through a first camera unit in a first terminal monitoring node in the distributed high-speed rail perimeter monitoring network to acquire the real-time state of the first sub-region;
The initial state calling module is used for calling a first initial monitoring state of the first target subarea in the initial monitoring state based on the real-time state of the first subarea;
the regional state result module is used for comparing the real-time state of the first subregion with the first initial monitoring state to obtain a first subregion state result;
the abnormal video stream module is used for activating a first video recording unit in the first terminal monitoring node when the state result of the first sub-region is abnormal, recording the first target sub-region and obtaining a first sub-region abnormal video stream;
the terminal monitoring result module is used for transmitting the first sub-region state result and the first sub-region abnormal video stream to a first processing center in the first terminal monitoring node, analyzing the first sub-region state result and the first sub-region abnormal video stream based on the first processing center, and acquiring a first terminal monitoring result, wherein the first terminal monitoring result comprises a first invasion main body set, a first invasion route set and a first invasion risk level set;
The monitoring visualization module is used for transmitting the first terminal monitoring result to the railway control center in real time through the wireless ad hoc network, and the railway control center performs visual display on the first terminal monitoring result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311572266.9A CN117768610A (en) | 2023-11-23 | 2023-11-23 | High-speed railway perimeter intrusion risk monitoring method and system based on multi-target recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311572266.9A CN117768610A (en) | 2023-11-23 | 2023-11-23 | High-speed railway perimeter intrusion risk monitoring method and system based on multi-target recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117768610A true CN117768610A (en) | 2024-03-26 |
Family
ID=90317229
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311572266.9A Pending CN117768610A (en) | 2023-11-23 | 2023-11-23 | High-speed railway perimeter intrusion risk monitoring method and system based on multi-target recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117768610A (en) |
-
2023
- 2023-11-23 CN CN202311572266.9A patent/CN117768610A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101942491B1 (en) | Hybrid ai cctv mediation module device consisting of road traffic situation monitoring and real time traffic information analysis | |
| CN111967393B (en) | Safety helmet wearing detection method based on improved YOLOv4 | |
| CN105744232B (en) | A kind of method of the transmission line of electricity video external force damage prevention of Behavior-based control analytical technology | |
| CN103069434B (en) | For the method and system of multi-mode video case index | |
| CN113887412B (en) | Detection method, detection terminal, monitoring system and storage medium for pollution emission | |
| CN111695512B (en) | Unattended cultural relic monitoring method and unattended cultural relic monitoring device | |
| CN111383429A (en) | Method, system, device and storage medium for detecting dress of workers in construction site | |
| JP3466169B2 (en) | Management system for roads and surrounding facilities | |
| CN112084963B (en) | Monitoring early warning method, system and storage medium | |
| CN112671104A (en) | Transformer substation multidimensional scene control platform facing complex scene | |
| CN106339657A (en) | Straw incineration monitoring method and device based on monitoring video | |
| CN112084892B (en) | Road abnormal event detection management device and method thereof | |
| US11398054B2 (en) | Apparatus and method for detecting fog on road | |
| US11580659B2 (en) | Method for size estimation by image recognition of specific target using given scale | |
| CN111680610A (en) | Construction scene abnormity monitoring method and device | |
| CN113313006A (en) | Urban illegal construction supervision method and system based on unmanned aerial vehicle and storage medium | |
| CN210222962U (en) | Intelligent electronic fence system | |
| CN114067396A (en) | Vision learning-based digital management system and method for live-in project field test | |
| CN114494845A (en) | Artificial intelligence hidden danger troubleshooting system and method for construction project site | |
| CN116886874A (en) | Ecological garden security monitoring and early warning data acquisition method and system | |
| CN117423052A (en) | Monitoring equipment adjustment and measurement system and method based on data analysis | |
| KR101542134B1 (en) | The apparatus and method of surveillance a rock fall based on smart video analytic | |
| CN117768610A (en) | High-speed railway perimeter intrusion risk monitoring method and system based on multi-target recognition | |
| KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
| CN113569956A (en) | Mountain fire disaster investigation and identification method based on AI algorithm |
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 |