CN114757506A - Motorized multi-sensor combined perimeter monitoring method and device - Google Patents
Motorized multi-sensor combined perimeter monitoring method and device Download PDFInfo
- Publication number
- CN114757506A CN114757506A CN202210319762.2A CN202210319762A CN114757506A CN 114757506 A CN114757506 A CN 114757506A CN 202210319762 A CN202210319762 A CN 202210319762A CN 114757506 A CN114757506 A CN 114757506A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- sensor
- perimeter
- task
- information
- 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 519
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000007726 management method Methods 0.000 claims abstract description 35
- 230000005540 biological transmission Effects 0.000 claims abstract description 23
- 238000005516 engineering process Methods 0.000 claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 238000012800 visualization Methods 0.000 claims abstract description 9
- 238000012423 maintenance Methods 0.000 claims description 18
- 238000009826 distribution Methods 0.000 claims description 14
- 241001465754 Metazoa Species 0.000 claims description 12
- 238000012806 monitoring device Methods 0.000 claims description 10
- 239000013307 optical fiber Substances 0.000 claims description 4
- 230000010287 polarization Effects 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 3
- 239000003999 initiator Substances 0.000 claims description 2
- 238000010223 real-time analysis Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 description 8
- 239000013598 vector Substances 0.000 description 6
- 230000000007 visual effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013499 data model Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Evolutionary Biology (AREA)
- Biodiversity & Conservation Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Alarm Systems (AREA)
Abstract
A motorized multi-sensor combined perimeter monitoring method and a device belong to the field of security systems, and are characterized in that: establishing a device management library based on the spatial information; establishing a monitoring sample library based on the spatial position and the monitoring attribute; flexibly calling a monitoring sensor in the equipment management library according to the monitoring task; the perimeter monitoring early warning module is used for monitoring and analyzing the monitoring data according to the acquired monitoring data and the information of the calling monitoring sample library; outputting a monitoring analysis result by a perimeter monitoring early warning module by adopting a GIS visualization technology; the maneuvering calling adopts a monitoring sensor scheduling algorithm based on the conflict probability. The combined study and judgment of the monitoring results of the monitoring samples of various monitoring sensors are comprehensively utilized, and an effective technical means is provided for perimeter monitoring; the GIS visualization technology is adopted to realize dynamic display of monitoring alarm, the perimeter monitoring dynamic early warning is realized by combining the monitoring studying and judging result, and the integrated technical support from information collection, transmission, studying and judging to early warning is provided for perimeter monitoring.
Description
Technical Field
The invention belongs to the field of security systems, and particularly relates to a motorized multi-sensor combined perimeter monitoring method and device.
Background
In recent years, with the rapid development of the internet of things technology and monitoring equipment, the technology of perimeter monitoring is continuously improved, and meanwhile, the perimeter has higher safety protection requirements in a plurality of important public places such as airports, stations, tracks and the like and transportation hubs. Compared with the traditional 24-hour patrol, the periphery anti-monitoring system adopts more Internet of things systems, generally adopts various sensors such as a camera, vibration and infrared to monitor and analyze the periphery intrusion behavior, and can give an alarm in time when the intrusion behavior is detected. However, in the current perimeter monitoring means, a method of densely arranging sensors is mostly adopted to realize full coverage monitoring of a specific area, on one hand, the sensors are inevitably arranged to have blind areas, cannot completely cover a monitoring range, especially has large-range and long-distance monitoring requirements, has higher requirements on cost and maintenance, and is very important how to complete a monitoring task by using limited budget; on the other hand, after the monitoring facility is established, after the intruder is gradually familiar with the sensor arrangement scheme along with the time, a corresponding method is adopted to avoid monitoring, so that the effectiveness of the arranged sensor fails.
Disclosure of Invention
The present invention is directed to solving the above problems, and provides a perimeter monitoring method and apparatus of mobile multi-sensor combination, which can flexibly move multiple monitoring sensors.
The invention relates to a motorized multi-sensor combined perimeter monitoring method, which comprises the following processes: establishing an equipment management library based on spatial information for all monitoring sensors;
establishing a monitoring sample library based on the spatial position and the monitoring attribute;
flexibly calling a monitoring sensor in an equipment management library according to the monitoring task;
the monitoring sensor transmits the acquired monitoring data to the perimeter monitoring and early warning module through a transmission network;
the perimeter monitoring early warning module is used for monitoring and analyzing the monitoring data according to the acquired monitoring data and the information of the calling monitoring sample library;
outputting a monitoring analysis result by a perimeter monitoring and early warning module by adopting a GIS visualization technology;
the maneuvering calling adopts a monitoring sensor scheduling algorithm based on the conflict probability, calculates the probability of the monitoring sensor for executing the current monitoring task according to the monitoring area and the monitoring requirement of the current monitoring task and the service condition of the existing sensor, and allocates the monitoring sensor with the maximum execution probability to the current monitoring task. The motor dispatching of the monitoring sensor is realized through motor dispatching, and the optimal monitoring of the current monitoring task by utilizing the existing monitoring sensor is ensured.
The motorized multi-sensor combined perimeter monitoring method provided by the invention can be used for flexibly laying the existing monitoring sensors and establishing the monitoring sample library, so that the efficiency of the existing monitoring sensors is furthest exerted, and technical support is provided for security protection and monitoring in various fields. A monitoring equipment management library based on spatial information is established, a foundation is laid for the mobile layout of the sensors, the sensors and the monitoring tasks are managed in a unified mode, the information in the equipment management library is updated synchronously after the layout positions of the sensors are adjusted, and the mobile scheduling of the monitoring sensors based on the monitoring tasks is supported; different types of monitoring terminal equipment are connected through a special network, so that real-time transmission and dynamic analysis of monitoring information are realized, and support is provided for fusion use of monitoring results; establishing a monitoring sample library, storing intrusion monitoring samples of different monitoring equipment for pedestrians, vehicles, animals and the like, comprehensively utilizing various monitoring sensor monitoring samples to jointly study and judge monitoring results, and providing an effective technical means for perimeter monitoring; the GIS visualization technology is adopted to realize dynamic display of monitoring alarm, the perimeter monitoring dynamic early warning is realized by combining the monitoring studying and judging result, and the integrated technical support from information collection, transmission, studying and judging to early warning is provided for perimeter monitoring.
Further, according to the perimeter monitoring method of the motorized multi-sensor combination, the coincidence degree of the execution time window of the monitoring task and the available time windows of the monitoring sensors on other tasks is used as a conflict probability measurement index by the monitoring sensor scheduling algorithm, the size of the conflict probability is calculated, and the equipment with the minimum conflict probability is allocated to the current task. In specific application, on the basis of an available time window, additional indexes such as the distance between a task area and a sensor, the task monitoring type, the sensor monitoring performance and the like are added, and an improved scheduling algorithm can be formed.
Further, the invention relates to a motorized multisensor joint perimeter monitoring method, wherein the equipment management base comprises a monitoring sensor unit, a sensor parameter unit, a maintenance information unit and a monitoring task unit; the monitoring sensor unit is used for recording management information including sensor types, working years, working modes and membership; the sensor parameter unit is used for recording parameter names, parameter units and parameter values; the maintenance unit is used for recording the maintenance type, the maintenance time and the maintenance unit of the sensor; the monitoring task unit is used for recording task names, task areas, task outlines, initiators, starting time and ending time; the monitoring sensor unit is connected with the monitoring task unit through a layout relation and is used for recording the layout position and layout time information of the sensors. The information of each unit is maintained and updated uniformly by setting the equipment management library, so that the positions and the state information of the sensors are consistent with the monitoring task, and the maintenance efficiency is improved. Meanwhile, in a conventional situation, the monitoring sensor is arranged at a certain fixed position and is gradually familiar to an intruder along with the time, so that monitoring is intentionally avoided, and the monitoring equipment cannot play a role in actual situation. The method is characterized in that a mode of flexibly dispatching the sensors is adopted, limited sensors are flexibly distributed in a monitoring area, a monitoring task, the distribution positions of the monitoring sensors, affiliated maintenance information, configuration parameters and maintenance information are combined, an equipment management library based on spatial information is established, the management requirement of flexibly and flexibly distributing the sensors is met, the monitoring sensors are dynamically dispatched based on the monitoring task, and therefore the limited monitoring sensors are utilized to exert the maximum monitoring efficiency.
Further, according to the motorized multi-sensor combined perimeter monitoring method, information collected during the establishment of the monitoring sample library comprises spatial positions and monitoring attributes; the monitoring attributes are classified into vibration types, linkage radar types and image types according to the types of the monitoring sensors; wherein the monitored attributes of the vibration-type sensor sample include intensity, wavelength, frequency, phase, and polarization; the monitoring attributes of the linkage radar sample comprise moving speed, track and image; the monitoring attribute of the image type sample is an image. By establishing a uniform monitoring sample library, monitoring information of different monitoring sensors is used as a monitoring sample for uniform storage management, and a sample studying and judging algorithm is used for quickly judging newly received monitoring information, so that the efficiency and reliability of monitoring judgment are improved. Along with the increase of the monitoring time, the collected monitoring samples are gradually improved, the research and judgment algorithm is continuously improved, and the monitoring efficiency and reliability are further improved.
Further, according to the mobile multi-sensor combined perimeter monitoring method, the spatial index based on the spatial position and the hierarchical attribute index based on the monitoring attribute classification are established in the monitoring sample library, so that the matching speed of monitoring information and monitoring samples is increased, and the monitoring and studying efficiency is improved.
Further, according to the mobile multi-sensor combined perimeter monitoring method, samples are classified and stored in a classified and graded mode according to the types of the monitoring sensors by the monitoring sample library; firstly, dividing a sample into four categories of a vibration optical fiber, a vibration sensor, a radar and a remote camera according to the type of a monitoring sensor; secondly, classifying the samples into three primary classes of human, vehicle and animal according to the monitoring result; and finally, further subdividing the primary class of the monitoring result into five secondary classes of human, large vehicle, small vehicle, large animal and small animal.
Further, according to the perimeter monitoring method combining the motorized multisensor, the perimeter monitoring early warning module displays historical monitoring events, real-time analysis and display of a large amount of monitoring data are achieved based on a real-time GIS technology, monitoring analysis results are displayed in a map window according to preset alarm types and levels, meanwhile, an alarm is sent out, a user is helped to quickly position alarm types and positions, and the user is helped to finish treatment decisions. The perimeter monitoring and early warning module can count the monitoring information in a period of time according to the alarm category and level and output the monitoring information in the form of a statistical chart or a table; and (3) outputting the distribution special topic of the intrusion event on the basis of the distribution position of the intrusion event, and assisting a monitoring manager to analyze the type and the distribution condition of the current intrusion event to formulate a coping scheme.
Further, according to the motorized multi-sensor combined perimeter monitoring method, the monitoring sensors adopt an adapter mode to realize unification of sensor analysis protocol interfaces. The realization process is that the interface unification is realized by defining the interface of the adapter, then defining different adapters aiming at each existing interface and through the corresponding relationship between the interface of the adapter and the adapter. The problem that upper layer service logic cannot be uniformly executed due to incompatibility of the two interfaces is solved, for a monitoring platform, the interfaces for analyzing data of different monitoring sensors are compatible, uniformly managed and cooperatively operated, the data analysis types with different protocols and incompatible interfaces are realized to work together, and the requirement for uniformly receiving and analyzing monitoring information of different monitoring sensors is met.
The invention relates to a motorized multi-sensor combined perimeter monitoring device, which comprises a monitoring sensor, a perimeter monitoring early warning module and a database module which are sequentially connected; the monitoring sensor transmits the acquired monitoring data to the perimeter monitoring and early warning module through a transmission network; the perimeter monitoring and early warning module is provided with a processor module;
the database module is used for storing a device management library established based on the spatial information and a monitoring sample library established based on the spatial position and the monitoring attribute;
the monitoring sensor is used for acquiring monitoring data;
the perimeter monitoring early warning module is used for monitoring and analyzing the monitoring data according to the acquired monitoring data and the information of the calling monitoring sample library; outputting a monitoring analysis result by adopting a GIS visualization technology;
the processor module is used for flexibly calling a monitoring sensor in the equipment management library to start working according to the monitoring task; the maneuvering calling adopts a monitoring sensor scheduling algorithm based on the conflict probability, calculates the probability of the monitoring sensor for executing the current monitoring task according to the monitoring area and the monitoring requirement of the current monitoring task and the service condition of the existing sensor, and allocates the monitoring sensor with the maximum execution probability to the current monitoring task.
Further, the invention relates to the motorized multisensor combined perimeter monitoring device, wherein the transmission network comprises a wireless network and a private network; a front-mounted switch is arranged between the monitoring sensor and the perimeter monitoring and early warning module; the monitoring sensor is connected with the front-end exchanger through a wireless network; the perimeter monitoring and early warning module is connected with the front-end switch through a private network.
The motorized multi-sensor combined perimeter monitoring method and the motorized multi-sensor combined perimeter monitoring device have the following technical effects:
(1) a monitoring equipment management base based on spatial information is established, monitoring tasks, the arrangement condition and the membership, the maintenance and the functional attribute of the monitoring sensors are managed in a unified way, historical monitoring information is combined with the equipment distribution condition and the monitoring range, a monitoring equipment conflict probability algorithm is established, the monitoring sensors are scheduled by a monitoring sensor scheduling algorithm, the maneuvering and combined perimeter monitoring is realized, and the use efficiency of the monitoring equipment is maximized;
(2) the method comprises the steps of establishing a monitoring sample library to store the spatial positions and monitoring attributes of monitoring samples of various sensors in a classified and classified manner, establishing a monitoring sample space and attribute index, realizing the rapid retrieval of samples, respectively adopting a KNN algorithm for vibration sensors and radar sensors, and quickly judging the monitoring result by adopting an SIFT algorithm for image sensors, so that the combined monitoring and research and judgment of the sample library and a research and judgment algorithm are realized;
(3) the perimeter monitoring and early warning module realizes the visual expression of real-time monitoring information and historical monitoring information by taking a map as a window, receives and analyzes the monitoring information in real time, visually displays monitoring alarm information, counts historical monitoring results, outputs special subjects and assists managers to make monitoring decisions.
Drawings
Fig. 1 is a schematic diagram of probability conflict of monitoring tasks according to an embodiment of the present invention;
fig. 2 is a flow chart of monitoring information forwarding according to the embodiment of the present invention;
fig. 3 is a schematic diagram of an adaptive structure for analyzing monitoring data according to the embodiment of the present invention;
fig. 4 is a schematic diagram of a monitoring information transmission network according to an embodiment of the present invention.
Detailed Description
The motorized multisensor combination perimeter monitoring method and apparatus according to the present invention are described in detail below with reference to the accompanying drawings and embodiments.
Example one
The embodiment discloses a motorized multi-sensor combined perimeter monitoring method, which comprises the following processes: establishing an equipment management library based on spatial information for all monitoring sensors; establishing a monitoring sample library based on the spatial position and the monitoring attribute;
flexibly calling a monitoring sensor in the equipment management library according to the monitoring task;
the monitoring sensor transmits the acquired monitoring data to the perimeter monitoring and early warning module through a transmission network;
the perimeter monitoring early warning module is used for monitoring and analyzing the monitoring data according to the acquired monitoring data and the information of the calling monitoring sample library;
outputting a monitoring analysis result by a perimeter monitoring early warning module by adopting a GIS visualization technology;
the maneuvering calling adopts a monitoring sensor scheduling algorithm based on the conflict probability, calculates the probability of the monitoring sensor for executing the current monitoring task according to the monitoring area and the monitoring requirement of the current monitoring task and the service condition of the existing sensor, and allocates the monitoring sensor with the maximum execution probability to the current monitoring task.
Establishing a monitoring equipment management library; in the embodiment of the present disclosure, postgresql9.6 is selected and postgis2.5.3 is adopted to perform space expansion on the monitoring device management library to form a relational space database. The Postgresql database adopts a spatial data model which accords with OGC (open gas grid) specifications, has higher universality, is widely applied to an open source relational spatial database at present, and can meet the requirements of monitoring sensor management and maneuvering scheduling by using Postgresql + PostGIS combination to form the spatial database for storing monitoring sensor information.
The whole equipment management library is divided into four parts, namely a monitoring sensor unit, a sensor parameter unit, a maintenance unit and a monitoring task unit, in the embodiment of the invention, spatial data are adopted for storage, and a database spatial reference system adopts a CGCS2000 national geodetic coordinate system and a 1985 national elevation standard. The monitoring sensor table stores the membership and management information of the sensors and identifies whether the equipment is used or not by setting an equipment state field; the monitoring task table is a space table, the element types are surfaces, the area range, the starting and stopping time, the support personnel and the used monitoring equipment of the primary monitoring task are stored, the equipment layout relation is used as a sub-table of the monitoring task table, the element types are point space tables, all sensors used in the monitoring task are stored, the equipment numbers are associated with the sensors, and the deployment positions of the sensors are stored through the layout of the point positions. The information of the monitoring equipment in use and the information of the monitoring equipment not in use are obtained through the sensor table, the corresponding arrangement position of the monitoring sensor in use and the position of a management unit where the monitoring sensor is not in use are obtained through the incidence relation, and the uniform management of the monitoring equipment based on the space position is realized. And timely updating of equipment information is realized by recording the maintenance record and the task information of the sensor, so that preparation is made for maneuvering scheduling of the monitoring equipment. In the implementation process, a spatial index is established for the monitoring task area range field and the monitoring equipment layout position field, so that the data access efficiency is improved.
According to the monitoring sensor scheduling method based on the conflict probability, the distribution probability when a plurality of tasks simultaneously need a certain device is calculated according to the distribution condition of the available time windows of different types of monitoring sensors by calculating the potential conflict coefficient between the available time windows of the monitoring sensors and the time windows of the monitoring tasks, and a sensor scheduling distribution scheme is designed, so that the implementation of the whole monitoring task is guaranteed.
When more than two monitoring tasks simultaneously need to use one device and the monitoring time of the device is overlapped, the device is considered to have use conflict, the conflict probability sum of all devices of each monitoring task is respectively calculated, and finally the monitoring device is distributed to the monitoring task with the minimum conflict probability, and the monitoring task time with the larger conflict probability is adjusted; the algorithm assumes that monitoring devices used by the same monitoring task are mutually independent and equal in weight, and the probability of conflict occurring in a certain task by a certain device is equal; the same equipment conflicts among a plurality of monitoring tasks, different monitoring tasks are mutually independent, and the conflict probability of each equipment is calculated according to the proportion of the overlapping time in the execution time of each task.
The monitoring sensor scheduling algorithm based on the conflict probability in the embodiment of the disclosure has the specific flow:
1) monitoring task conflict probability calculation
The collision probability of the monitoring device is derived from monitoring tasks which collide with each other, and in the embodiment of the present disclosure, the collision probability of the monitoring tasks is calculated by taking the occurrence of collision of two monitoring tasks as an example.
If shown in FIG. 1, the execution time of the monitoring task 1 starts from ST1 to ET1, and the execution time of the monitoring task 2 starts from ST2 to ET2, wherein ET1-ST2 are two conflict times of the monitoring tasks. According to the above assumptions, the collision probability calculation formula of the single monitoring task in the collision is as follows:
where t denotes the time of the monitoring task conflict and is represented in the diagram by the red part, ETi-STiIn order to monitor the execution time of the task, the execution time is equal for the same monitoring task, so the above formula can be simplified as follows:
2) monitoring sensor collision probability calculation
In a monitoring task, it is possible to use multiple monitoring sensors at the same time, and different sensors are considered as equal probability events because they operate independently, so the collision probability of the same monitoring sensor in a certain collision monitoring task is the average of the collision probability of the monitoring task according to the number of the sensors, and the calculation formula is as follows:
in the formula PwiIndicating the probability of conflict of the monitoring task associated with the current sensor, and m indicating the number of used monitoring sensors in the current monitoring task.
The collision probability of the monitoring sensor is derived from the collision probability of the monitoring tasks, the same monitoring sensor is possibly used in the multiple monitoring tasks, and the collision probability of the same monitoring sensor is the sum of the collision probabilities in the multiple monitoring tasks because the probabilities of different monitoring tasks are independent from each other, and the specific calculation formula is as follows:
in the formula PsjiIndicating the probability of collision of sensors in a monitoring task, PsIndicating the probability of collision of the monitoring sensors in all monitoring tasks.
3) Monitoring task collision probability summation calculation
All sensors in the monitoring task independently run in the process of executing the task and are used as independent individuals for maintenance and management, so that the sensors can be considered to be independent from each other, and the monitoring task conflict probability sum is the conflict probability sum of the sensors used by the monitoring task to execute the task, and the calculation formula is as follows:
in the formula PsjThe collision probability of the monitoring equipment in one monitoring task is shown, and m represents the number of the monitoring sensors used in one monitoring task.
4) Monitoring sensor maneuvering dispatch
In the embodiment of the disclosure, the monitoring sensor scheduling is divided into an emergency task guarantee mode and a conventional task mode, and in the emergency task guarantee mode, a sensor with the minimum conflict probability is preferentially allocated to an emergency monitoring task; and in a conventional task mode, scheduling the monitoring sensor according to the principle of preferentially allocating the monitoring equipment with the minimum collision probability for the monitoring task with the minimum collision probability.
In the algorithm implementation process, firstly, according to the starting and stopping time of monitoring task execution, equipment which is relatively close to the task time is searched from an equipment library, and two conditions of conflict and non-conflict are distinguished; then, under the condition of conflict, monitoring equipment conflict probability, taking 0.5 as a threshold value of the monitoring equipment conflict probability value in the embodiment of the disclosure, screening equipment meeting the conditions, and sequencing according to the conflict probability from small to large; and finally, circularly matching all types of sensors for the current monitoring task.
The monitoring transmission network comprises two parts, wherein various sensors are connected in a monitoring task area in a mode of establishing a wireless network, the monitoring task area is connected to a monitoring center through a special line, and the wireless network is accessed to the special network through a switch in the task area. In the monitoring task area, different types of sensors are connected with a monitoring center data receiving server side in a point-to-point mode through different server sides.
In order to realize the purpose of reading the integrated monitoring information under the private network environment under the intranet environment, the private network is accessed into the intranet through the gatekeeper in the embodiment of the disclosure, and the gatekeeper controls the unidirectional transmission of the monitoring information to the intranet.
Because the conventional monitoring sensor only supports point-to-point transmission, namely one monitoring information sending end corresponds to one information receiving end, and accessing the intranet through a gatekeeper needs to realize that one monitoring receiving end simultaneously sends information to two receiving ends. The modified data transmission analysis flow is shown in fig. 2.
In the embodiment of the disclosure, a Socket protocol is adopted in the forwarding process to realize transmission of monitoring information from a monitoring sensor to a data forwarding node and an intranet data receiving node, the data forwarding node serves as a Socket server and a client at the same time, serves as the server to receive the monitoring information from the sensor client, and serves as the client to send the received monitoring information to the intranet data receiving node.
In order to improve the data analysis expansibility, an adapter mode as shown in fig. 3 is adopted, an adapter interface is defined firstly, then different adapters are defined for each existing interface, and the interfaces are unified through the interface corresponding relation between the adapter interface and the adapter. The problem that the upper layer service logic cannot be uniformly executed due to incompatibility of the interfaces of the monitoring platform and the monitoring platform is solved, and the interfaces for analyzing data of different monitoring sensors are compatible, uniformly managed and cooperatively worked.
In the embodiment of the disclosure, a monitoring sample library is composed of sample entity data, sample attribute information and a field photo, wherein the sample entity data is divided into a vibration type, a radar type and a remote camera type according to the type of a monitoring sensor, and is divided into a relation type and a file type according to the type of a monitoring result, and the vibration type sensor, such as a vibration optical fiber and a vibration sensor, mainly stores vibration attributes such as strength, wavelength, frequency, phase, polarization and the like and is stored by adopting a relation database; the radar sensor mainly stores information such as moving speed, track, image and the like of an object, and stores the information in a form of adding files to a relational database; the remote camera class mainly stores an image file obtained by pixel difference of two continuous frames of images, and stores the image file by adopting a relational database and file adding mode.
In the logic design of a monitoring sample library, three tables are respectively adopted to store sample data of vibration, radar and remote camera shooting, wherein the sample data comprises monitoring attribute information and a spatial position of sensors of different types, and the remote camera sensor judges a monitoring object by adopting a frame difference method, so that the sample also comprises an image file obtained by subtracting two continuous frames of images besides an attribute item; and storing a sample image file record by using a table, wherein the sample image file record comprises information such as the name, the storage path, the type, the acquisition time and the like of the file, and the sample and the image file are associated by a sample ID.
The monitoring sample library is built by adopting Postgresql + PostGIS, the monitoring position is recorded in a point form, and the monitoring result is recorded by the target type, such as pedestrians, vehicles, animals and the like. In the process of collecting and warehousing the samples, all attributes of the samples are indispensable items, the images are required to be clear, the monitoring types can be definitely judged, and achievements in the sample library need to be regularly checked to eliminate records with poor quality.
On the basis of a monitoring sample library, according to the difference of data acquired by a monitoring sensor, the monitoring results of the vibration sensor and the radar sensor are judged by adopting a KNN algorithm, and the monitoring results of the image sensor are judged by adopting an SIFT algorithm so as to achieve the optimal monitoring effect.
In this embodiment, the data input into the KNN classifier is a feature vector formed by a set of discrete high-dimensional monitoring parameters, the distance between the sample to be measured and each sample in a given sample data set needs to be calculated, each dimension of the data represents an extracted feature value, and the dimensions are the same. And the Euclidean distance is used for calculating the absolute difference between sample characteristic vectors, and is suitable for analyzing a data model expressing the difference from the numerical value of the same dimension. Therefore, in this embodiment, the euclidean distance is used for the distance calculation. And calculating the distances among the intensity, wavelength, frequency, phase and polarization of the vibration sensor and the distances among the moving speed and acceleration of the radar by using the Euclidean distance.
In the embodiment of the disclosure, the first-class classes of the identification target classes are determined to be three major classes of pedestrians, vehicles and animals, the vehicles are subdivided according to large-scale vehicles and small-scale vehicles in the second-class classes, and the animals are subdivided according to large-scale animals and small-scale animals. Therefore, the K value of the first class is 3, the K value of the second class is 5, and in the process of establishing the algorithm, proper sample data is selected for the K value determined by actual classification to be verified so as to ensure the effectiveness of the algorithm.
In the embodiment of the disclosure, the K values of the vibration sensor and the radar sensor are respectively checked by adopting a cross validation mode, 20 samples are respectively selected from a sample database for each of the primary class and the secondary class to form a validation set, an algorithm is executed by using the sample data, and the judgment result is compared with an actual result, wherein the actual result shows that the error rate of algorithm identification in the 3 primary classes is 3%, and the error rate of algorithm identification in the secondary class is 9%, which indicates that the algorithm is effective in judging the monitoring results of the vibration sensor and the radar monitoring sensor under normal conditions.
In the embodiment, the gray difference of two adjacent frames of images of the remote camera is used as input data to be compared with the samples in the sample library to judge the monitoring result, and an SIFT algorithm is adopted in the judging process to realize the research and judgment of the monitoring result.
The SIFT algorithm specifically comprises four steps of detection of extreme points of a scale space, positioning of feature points, direction distribution of the feature points, generation of feature descriptors and feature matching.
And (4) detecting extreme points of the scale space, and performing convolution processing on any image by using a Gaussian kernel function to generate a Gaussian scale space.
L(x,y,σ)＝G(x,y,σ)×I(x,y)
And changing the parameter sigma to generate different blurred images, and then carrying out alternate sampling on the blurred images to generate a Gaussian pyramid. The two adjacent layers of pyramids are subjected to subtraction operation to obtain a differential pyramid, namely,
D(x,y,σ)＝L(x,y,kσ)-L(x,y,σ)
and positioning the characteristic points, and calculating to obtain accurate characteristic points according to the extreme points of the discrete space. And taking the extreme point obtained by the comparison detection as an origin, expanding the function Taylor at the nearby sample point, interpolating by virtue of local sample point information to obtain a real extreme point, and removing unstable points by a method of setting a threshold value.
The method comprises the steps of distributing direction of characteristic points and generating a characteristic descriptor, distributing a main direction to each characteristic point according to gradient distribution of characteristic point neighborhoods, enabling the generated characteristic descriptor to have rotation invariance, calculating the gradient and the direction of each pixel in a 3 sigma neighborhood in a scale space where the characteristic points are located, and defining the gradient size m (x, y) and the direction theta (x, y) of the pixel (x, y) in any characteristic point neighborhood as follows
The coordinate axis is rotated to be parallel to the main direction of the feature point, then a 16 × 16 pixel neighborhood taking the feature point as the center is selected as a sampling window, gradient direction histograms of 8 directions are calculated in each 4 × 4 window, the accumulated value of each gradient direction is drawn, so that seed points with 8 data are generated, each feature point is composed of 4 × 4 seed points, each seed point contains 8-dimensional feature vectors, and finally a 128-dimensional feature descriptor is generated.
And (3) feature matching, wherein SIFT takes Euclidean distance as the similarity measurement of the feature vector, and if the ratio of the nearest neighbor distance to the next nearest neighbor distance of the feature point is less than a certain threshold value, the feature point is judged to be correctly matched with the nearest neighbor point. And finally, rejecting mismatching points through a Random Sample Consensus (RANSAC) algorithm to finish matching of the feature points.
In the embodiment of the disclosure, the monitoring information is collected by the monitoring sensor and transmitted to the private network through the wireless network, after the private network receives the monitoring information, the monitoring result is sent to the private network client for display after being monitored and distinguished, and meanwhile, the monitoring information is forwarded to the intranet data receiving node, received by the intranet data receiving node and sent to the client for display.
In the embodiment of the disclosure, after receiving the monitoring message, the private network or intranet data receiving server actively pushes data to the client by using a WebSocket data transmission technology, so that a large amount of monitoring information is dynamically updated and displayed at the client in real time. After the WebSocket connection is established, subsequent data are transmitted in a frame sequence form. Before the WebSocket connection is disconnected at the client side or the connection is interrupted at the Server side, the client side and the Server side do not need to initiate the connection request again, the consumption of network bandwidth resources is greatly saved, and the sending and receiving messages are initiated on the same persistent connection, so that the method has obvious performance advantages.
In the visual display process of the monitoring alarm information, space basic data such as high-resolution remote sensing data, vector data, place name data and the like are positioning reference information, the updating frequency of the positioning reference information is not very high compared with that of the monitoring alarm information, and the monitoring alarm information does not need to be updated at a client side when updated, so that the alarm state of the client side is updated by a server side by adopting a WebSocket technology, and the requirement of real-time information display can be met.
According to the embodiment of the invention, in the visual display process of the monitoring information, the state change of the alarm information is dynamically displayed, the influence range of the monitoring alarm is subjected to analog display, so that a user can visually see the type of perimeter intrusion, and meanwhile, a disposal method is formulated by means of space basic data and the spatial position of the alarm sensor.
Meanwhile, in the embodiment of the disclosure, the monitoring information is counted according to the alarm categories and levels in monthly and annual time periods, the position of the monitoring area is combined for spatial statistics, different alarm categories are formed into a time statistics histogram and a statistical report according to the time statistics, the spatial statistics is performed according to the spatial distribution of the monitoring alarm information by taking the administrative area where the monitoring area is located as a unit, the categories and the number of various alarm information of each administrative area are taken as thematic data, and the high-resolution remote sensing image data, the vector administrative division and the place name data are taken as spatial basic data to manufacture a monitoring alarm thematic map, so that a monthly and annual monitoring distribution thematic map is formed, and monitoring personnel are assisted in evaluating the monitoring efficiency.
Example two
The embodiment discloses a motorized multi-sensor combined perimeter monitoring device, which comprises a monitoring sensor, a perimeter monitoring early warning module and a database module which are connected in sequence; the monitoring sensor transmits the acquired monitoring data to the perimeter monitoring and early warning module through a transmission network; the perimeter monitoring and early warning module is provided with a processor module; the database module is used for storing a device management library established based on the spatial information and a monitoring sample library established based on the spatial position and the monitoring attribute; the monitoring sensor is used for acquiring monitoring data; the perimeter monitoring early warning module is used for monitoring and analyzing the monitoring data according to the acquired monitoring data and the information of the calling monitoring sample library; and outputting a monitoring analysis result by adopting a GIS visualization technology.
The processor module is used for flexibly calling a monitoring sensor in an equipment management library to start working according to a monitoring task; the maneuvering calling adopts a monitoring sensor scheduling algorithm based on conflict probability, calculates the probability of the monitoring sensor for executing the current monitoring task according to the monitoring area and the monitoring requirement of the current monitoring task and the service condition of the existing sensor, and allocates the monitoring sensor with the maximum execution probability to the current monitoring task. The transmission network realizes real-time transmission of monitoring information from the sensor to the monitoring platform by adopting a mode of combining a wireless network and a private network, and the wireless network is adopted near the sensor to connect the sensor and is converged to the private network to ensure the transmission efficiency of data; the method comprises the following steps that a Socket technology is adopted to realize transmission and reception of monitoring information from a sensor to a monitoring platform, the sensor serves as a Socket client, and the monitoring platform serves as a Socket server to monitor information and analyze the received information in real time; in the embodiment of the present disclosure, as shown in fig. 4, a front-end switch is used to access a monitoring terminal device to a private network, and the security of the whole network data is ensured through a firewall, a core switch and a data center switch. And the real-time transmission of monitoring data is realized on the basis of the support of private network infrastructure. The monitoring sensors connected to the monitoring platform send monitoring information to the monitoring platform in real time through respective matched information forwarding hosts, and the monitoring platform serves as a Socket server, continuously monitors messages from the terminal equipment, and receives and analyzes the monitoring information in real time.
The perimeter monitoring device for mobile multi-sensor combination according to the embodiment of the present disclosure is configured to, when performing mobile multi-sensor combination perimeter monitoring, establish an equipment management library, a monitoring sample library, a monitoring sensor scheduling algorithm based on collision probability, and transmit monitoring information and analyze results, where the specific processes or implementation steps involved in establishing the equipment management library, the monitoring sample library, the monitoring sensor scheduling algorithm based on collision probability are the same as those described in the first embodiment, and are not described herein again.
According to the perimeter monitoring method and device based on the mobile multi-sensor combination, disclosed by the invention, all sensors can be concentrated to ensure the perimeter safety of a certain area through mobile scheduling sensors, the sensors are scheduled to a new monitoring area according to the monitoring effect, and a fixed arrangement mode is replaced by a mobile arrangement mode, so that the sensor efficiency is maximally exerted. Meanwhile, the advantages of various sensors such as the vibration optical fiber, the vibration sensor, the linkage radar and the remote camera are comprehensively utilized, and the maneuvering and combined perimeter monitoring is realized. The sensors are uniformly managed by combining a space positioning technology, and a conflict probability scheduling algorithm is adopted to allocate the monitoring sensors, so that support is provided for the flexible layout of the sensors; monitoring alarm information is transmitted in real time based on a private network, and real-time visual expression is carried out through a GIS technology, so that visual display of a monitoring result is realized; and a joint monitoring studying and judging mechanism based on a sample library is established, so that the monitoring results of various sensors can be quickly judged.
Claims (10)
1. A motorized multi-sensor combined perimeter monitoring method is characterized in that:
establishing an equipment management library based on spatial information for all monitoring sensors;
establishing a monitoring sample library based on the spatial position and the monitoring attribute;
flexibly calling a monitoring sensor in the equipment management library according to the monitoring task;
the monitoring sensor transmits the acquired monitoring data to the perimeter monitoring and early warning module through a transmission network;
the perimeter monitoring early warning module is used for monitoring and analyzing the monitoring data according to the acquired monitoring data and the information of the calling monitoring sample library;
outputting a monitoring analysis result by a perimeter monitoring early warning module by adopting a GIS visualization technology;
the maneuvering calling adopts a monitoring sensor scheduling algorithm based on the conflict probability, calculates the probability of the monitoring sensor for executing the current monitoring task according to the monitoring area and the monitoring requirement of the current monitoring task and the service condition of the existing sensor, and allocates the monitoring sensor with the maximum execution probability to the current monitoring task.
2. The motorized multisensor combination perimeter monitoring method of claim 1, wherein: the monitoring sensor scheduling algorithm takes the coincidence degree of the execution time window of the monitoring task and the available time windows of the monitoring sensors on other tasks as a conflict probability measurement index, calculates the size of the conflict probability, and allocates the equipment with the minimum conflict probability to the current task.
3. The motorized multisensor combination perimeter monitoring method of claim 1, wherein: the equipment management library comprises a monitoring sensor unit, a sensor parameter unit, a maintenance information unit and a monitoring task unit; the monitoring sensor unit is used for recording management information including sensor types, working years, working modes and membership; the sensor parameter unit is used for recording parameter names, parameter units and parameter values; the maintenance unit is used for recording the maintenance type, the maintenance time and the maintenance unit of the sensor; the monitoring task unit is used for recording task names, task areas, task summaries, initiators, starting time and ending time; the monitoring sensor unit is connected with the monitoring task unit through a layout relation and is used for recording the layout position and layout time information of the sensors.
4. The motorized multisensor combination perimeter monitoring method of claim 1, wherein: the information collected when the monitoring sample library is established comprises a spatial position and a monitoring attribute; the monitoring attributes are classified into vibration types, linkage radar types and image types according to the types of the monitoring sensors; wherein the monitored attributes of the vibration-type sensor sample include intensity, wavelength, frequency, phase, and polarization; the monitoring attributes of the linkage radar sample comprise moving speed, track and image; the monitoring attribute of the image type sample is an image.
5. The motorized multisensor combination perimeter monitoring method of claim 4, wherein: the monitoring sample library is established with a spatial index based on spatial position and a hierarchical attribute index based on monitoring attribute classification.
6. Motorized multisensor joint perimeter monitoring method according to claim 4 or 5, wherein: the monitoring sample library is used for classifying, grading and storing samples according to the category of the monitoring sensor; firstly, dividing a sample into four categories, namely a vibration optical fiber, a vibration sensor, a radar and a remote camera according to the type of a monitoring sensor; secondly, classifying the samples into three primary classes of human, vehicle and animal according to the monitoring result; and finally, further subdividing the primary class of the monitoring result into five secondary classes of human, large vehicle, small vehicle, large animal and small animal.
7. The motorized multisensor combination perimeter monitoring method of claim 1, wherein: the perimeter monitoring and early warning module displays historical monitoring events, realizes real-time analysis and display of a large amount of monitoring data based on a real-time GIS technology, displays the monitoring analysis result in a map window according to preset warning categories and levels, and simultaneously sends out a warning;
the perimeter monitoring and early warning module can count the monitoring information in a period of time according to the alarm category and level and output the monitoring information in the form of a statistical chart or a table; and outputting the distribution special topic of the intrusion event based on the distribution position of the intrusion event.
8. The motorized multisensor combination perimeter monitoring method of claim 1, wherein: the monitoring sensor adopts an adapter mode to realize the unification of sensor analysis protocol interfaces.
9. A perimeter monitoring device that motor-driven multisensor is united which characterized in that: the system comprises a monitoring sensor, a perimeter monitoring early warning module and a database module which are connected in sequence; the monitoring sensor transmits the acquired monitoring data to the perimeter monitoring and early warning module through a transmission network; the perimeter monitoring and early warning module is provided with a processor module;
the database module is used for storing a device management library established based on the spatial information and a monitoring sample library established based on the spatial position and the monitoring attribute;
the monitoring sensor is used for acquiring monitoring data;
the perimeter monitoring early warning module is used for monitoring and analyzing the monitoring data according to the acquired monitoring data and the information of the calling monitoring sample library; outputting a monitoring analysis result by adopting a GIS visualization technology;
the processor module is used for flexibly calling a monitoring sensor in the equipment management library to start working according to the monitoring task; the maneuvering calling adopts a monitoring sensor scheduling algorithm based on the conflict probability, calculates the probability of the monitoring sensor for executing the current monitoring task according to the monitoring area and the monitoring requirement of the current monitoring task and the service condition of the existing sensor, and allocates the monitoring sensor with the maximum execution probability to the current monitoring task.
10. The motorized multisensor combination perimeter monitoring apparatus of claim 9, wherein: the transmission network comprises a wireless network and a private network; a front-mounted switch is arranged between the monitoring sensor and the perimeter monitoring and early warning module; the monitoring sensor is connected with the front-end exchanger through a wireless network; the perimeter monitoring and early warning module is connected with the front-end switch through a private network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210319762.2A CN114757506A (en) | 2022-03-29 | 2022-03-29 | Motorized multi-sensor combined perimeter monitoring method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210319762.2A CN114757506A (en) | 2022-03-29 | 2022-03-29 | Motorized multi-sensor combined perimeter monitoring method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114757506A true CN114757506A (en) | 2022-07-15 |
Family
ID=82327832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210319762.2A Pending CN114757506A (en) | 2022-03-29 | 2022-03-29 | Motorized multi-sensor combined perimeter monitoring method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114757506A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8065342B1 (en) * | 2008-02-22 | 2011-11-22 | BorgSolutions, Inc. | Method and system for monitoring a mobile equipment fleet |
| CN106408833A (en) * | 2016-11-02 | 2017-02-15 | 北京弘恒科技有限公司 | Perimeter intrusion detection method and system |
| CN106780256A (en) * | 2016-12-16 | 2017-05-31 | 北京华鼎新铭智能科技发展有限公司 | A kind of comprehensive safeguard information management system |
| US20170248699A1 (en) * | 2016-02-26 | 2017-08-31 | Carrier Corporation | Visualization system for monitored space |
| CN110956769A (en) * | 2019-12-13 | 2020-04-03 | 珠海大横琴科技发展有限公司 | Monitoring method of perimeter anti-intrusion system based on target position |
| KR20210078338A (en) * | 2019-12-18 | 2021-06-28 | 주식회사 포휴먼테크 | Intrusion detection system combining high performance rader and machine learning |
| CN114067522A (en) * | 2021-10-15 | 2022-02-18 | 金茂数字科技有限公司 | Wisdom security protection management system based on wisdom thing allies oneself with |
-
2022
- 2022-03-29 CN CN202210319762.2A patent/CN114757506A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8065342B1 (en) * | 2008-02-22 | 2011-11-22 | BorgSolutions, Inc. | Method and system for monitoring a mobile equipment fleet |
| US20170248699A1 (en) * | 2016-02-26 | 2017-08-31 | Carrier Corporation | Visualization system for monitored space |
| CN106408833A (en) * | 2016-11-02 | 2017-02-15 | 北京弘恒科技有限公司 | Perimeter intrusion detection method and system |
| CN106780256A (en) * | 2016-12-16 | 2017-05-31 | 北京华鼎新铭智能科技发展有限公司 | A kind of comprehensive safeguard information management system |
| CN110956769A (en) * | 2019-12-13 | 2020-04-03 | 珠海大横琴科技发展有限公司 | Monitoring method of perimeter anti-intrusion system based on target position |
| KR20210078338A (en) * | 2019-12-18 | 2021-06-28 | 주식회사 포휴먼테크 | Intrusion detection system combining high performance rader and machine learning |
| CN114067522A (en) * | 2021-10-15 | 2022-02-18 | 金茂数字科技有限公司 | Wisdom security protection management system based on wisdom thing allies oneself with |
Non-Patent Citations (2)
| Title |
|---|
| 朱以洲 等: "GIS技术在路灯监控系统中的应用研究", 城市勘测, no. 1, 29 February 2016 (2016-02-29), pages 81 - 84 * |
| 陈敏: "人工智能通信理论与方法", vol. 1, 31 January 2020, 华中科技大学出版社, pages: 283 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Al-qaness et al. | An improved YOLO-based road traffic monitoring system | |
| AU2014214545B2 (en) | A surveillance system | |
| Xiong et al. | A kind of novel ITS based on space-air-ground big-data | |
| CN102929903B (en) | Rapid video retrieval method based on layered structuralized description of video information | |
| CN104200671A (en) | Method and system for managing virtual gate based on big data platform | |
| CN110825728A (en) | Three-dimensional command center of synthesizing in wisdom city | |
| CN113642403B (en) | Crowd abnormal intelligent safety detection system based on edge calculation | |
| CN108710827A (en) | A kind of micro- police service inspection in community and information automatic analysis system and method | |
| KR102573708B1 (en) | Apparatus and method for controlling situation board of transport information center | |
| Ramesh et al. | Smart traffic congestion model in IoT-A review | |
| CN114783188A (en) | Inspection method and device | |
| CN114841712B (en) | Method and device for determining illegal operation state of network appointment vehicle tour and electronic equipment | |
| CN118038674A (en) | Monitoring system real-time dynamic analysis warning system based on big data | |
| CN114023076A (en) | Specific vehicle tracking method based on multi-source heterogeneous data | |
| Abu-Alsaad | CNN-Based Smart Parking System. | |
| CN114757506A (en) | Motorized multi-sensor combined perimeter monitoring method and device | |
| CN112529466A (en) | Smart city management method and system based on big data | |
| KR20220005353A (en) | Method and apparatus for urban traffic network modeling with multiple cctv videos | |
| KR101686851B1 (en) | Integrated control system using cctv camera | |
| Rashid et al. | Automated traffic measurement system based on FCD and image processing | |
| Dalkic et al. | A self-policing smart parking solution | |
| CN117746630B (en) | Industry safety production supervision and management system based on big data | |
| He et al. | Applications of deep learning techniques for pedestrian detection in smart environments: a comprehensive study | |
| CN118469426B (en) | Data processing method and system based on Internet of things | |
| Gregor et al. | Design and implementation of a counting and differentiation system for vehicles through video processing |
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 |