New Points of Fusion Application of Bayonet and Big Data

<div class="iteye-blog-content-contain" style="font-size: 14px;">
With the continuous construction and investment of intelligent transportation information system projects in various places, the transportation facilities and equipment have increased greatly, especially the card The number of bayonet systems has grown rapidly, which has led to a rapid expansion of the passing data generated by bayonet systems. Due to the lack of proper means to utilize these massive data, the phenomenon of "heavy construction and neglect application" has been caused. Combined with practical work experience, the author of this paper proposes several applications based on the bayonet system, which helps to enrich the application forms of the bayonet system. 
Text/Wu Mingyuan
The big data behind the bayonet
Although the bayonet system includes many complex technologies, such as image recognition, the final result generated by the bayonet system is The passing information is very simple, nothing more than the license plate number, model, color, driving speed, passing time, bayonet number, lane number, etc. It is the aggregation of these simple information that forms a huge number of passing records, and finally produces a number of applications based on the bayonet system. 
 Assuming that an average of 3,000 vehicles pass through each checkpoint per hour, and every day is counted as 10 hours (the number of cars passing at night is very small), then there are 30,000 pieces of passing data per day at each checkpoint. At present, the number of checkpoints in first-tier cities is about 500. Based on this calculation, the daily record of passing vehicles will be as high as 15 million, and it will exceed 5 billion records in one year. 
Inspection prevention and control
As the most direct application of the bayonet system, inspection prevention and control is the most widely used, such as remediation of vehicles overdue for annual inspection. The user specifies the license plate number that needs to be controlled, and the bayonet system will scan the vehicle information passing through all bayonet in real time. Once the controlled vehicle is found, it will promptly send an alarm on various terminals in various ways so that relevant personnel can intercept it. 
Although there are many applications of audit prevention and control, there are still areas worth improving. When the control vehicle passes through a certain bayonet, the interceptor is usually not on the scene. When the interceptor arrives at the scene, the control vehicle has already fled, which loses the meaning of prevention and control. If combined with GIS technology, this problem can be well solved. When the bayonet system detects that the vehicle under control passes through a certain bayonet, with the help of GIS technology, it can calculate one or more bayonets that the control vehicle will pass through next, and estimate the possible passing time of each bayonet. This makes it convenient for the command center to send personnel to the relevant bayonet to intercept in time, which effectively improves the efficiency of inspection and prevention. 
In the near future, it is expected to unify all the passing information of the national bayonet system to realize a nationwide inspection and prevention network, thereby greatly improving the investigation and punishment of illegal vehicles. . 
Investigation of vehicles involved in the case
With the rapid development of the city, the city is getting bigger and bigger, and the road network is becoming more and more complicated. In order to quickly escape the arrest of the public security organs, more illegal Criminals use vehicles to commit crimes. In fact, the vast majority of criminals will take vehicles to the scene to step on the spot, repeatedly inspect the scene, wait for the right time to commit illegal and criminal acts, and finally escape in a vehicle. With the help of the massive passing information recorded in the bayonet system, combined with massive data analysis technology, it is entirely possible to identify the vehicles that criminals are riding in from the vast sea of vehicles. 
In addition to this post-event investigation technology for vehicles involved in the case, using the same technology, the system can also take the initiative to predict the illegal behavior that will occur in time, so as to prevent problems before they occur, and further improve the application scope of the system . 
Companion Vehicle Analysis
 Due to the continuous improvement of public security awareness, the success rate of criminals to independently carry out criminal acts has been greatly reduced. Therefore, criminal acts in the new era begin to manifest as group warfare and form criminal gangs. Criminal gangs often use multiple vehicles to increase their success rate when stepping and committing crimes. From the point of view of the bayonet system, the specific manifestation of group crime is that multiple vehicles appear in the coverage area of a specific bayonet at the same time. Using this feature, from the massive passing data, it is possible to extract vehicles that meet specific conditions and improve the detection efficiency. This is the companion vehicle analysis. 
 In the application, the user selects the travel route of the vehicle, sets the interval time between vehicles, the number of vehicles to follow, and the start and end range of the analysis time. All vehicles. Usually, the number of vehicles that accompany the vehicle analysis is not too many. From this small collection, users can quickly locate the vehicles used by criminals from this small collection, combined with the high-definition photos taken by the bayonet system, and then follow the clues and quickly catch the criminals. 
The rational use of accompanying vehicle analysis can not only play a role in quickly locating suspected vehicles after the incident, but also provide early warning information before the incident, so as to prevent problems before they occur and reduce the risk of people's lives and property. loss. 
Analysis of vehicles entering and leaving the crime scene
As mentioned above, before committing a crime, criminals usually continue to step on the spot, investigate the scene many times, and analyze the appropriate time for committing the crime until they succeed. committing a crime, and then not appearing at the crime scene for a long time. From the point of view of the bayonet system, before the time of committing the crime, there will be vehicles frequently entering and leaving the bayonet at the crime scene; after the crime, these vehicles all disappear at these bayonet bays. Using this feature, it is completely possible to find all the vehicles that meet the characteristics of the crime, and then combined with other information, you can quickly find the suspect vehicle. This is the analysis of vehicles entering and leaving the crime scene, which is used to analyze the entry before the crime and the stop during the crime. , The vehicle that left after the incident. 
 It can be seen from the above application scenarios that this analysis method can be effectively used to provide sufficient comparative analysis data only after the case has occurred for a period of time.
 Analysis of abnormal vehicles
 During the process of stamping, criminals usually frequently enter and exit specific checkpoints at specific times. Use data warehouse analysis technology to find out the most active vehicles in a specific time period (such as night) and at a specific bayonet (such as a high-incidence area), users can further analyze these vehicles, detect potential criminal behavior in time, and effectively deter criminals , which is the abnormal vehicle analysis. 
In addition, it can also analyze the abnormal traffic patterns of vehicles. For example, it suddenly appears in an area that doesn't appear often, or it should appear frequently in a certain area, but suddenly it doesn't appear for a few days. These abnormal behaviors can be mined from the massive passing data. 
When using abnormal vehicle analysis, you can filter out the types of vehicles such as taxis as needed to reduce the scope of investigation and improve the analysis efficiency. 
Travel Time Calculation
When the vehicle passes through each bayonet, the bayonet system will record the passing time and license plate number, so that the passing time of the vehicle between adjacent bayonets can be calculated. time, this is the travel time calculation. 
With the travel time between adjacent bayonets, combined with GIS, the road length between adjacent bayonets can be automatically measured, and the average driving speed of the vehicle between adjacent bayonets can be calculated. It's the interval speed. 
Due to the extensive use of electronic dogs, many drivers will actively reduce the speed when passing through the bayonet, and once they leave the coverage of the bayonet, they will rapidly increase the speed and drive over speed. The traditional single-point speed measurement This overdrive could not be detected. The use of interval speed measurement can effectively solve this problem. At this time, only need to install bayonet at both ends of the closed road to find the speeding behavior. Using this method can effectively reduce the number of bayonet construction and save construction funds; at the same time, it solves the drawbacks of single-point speed measurement, which can effectively reduce accidents caused by overspeed, strengthen road network traffic standards, and improve road network traffic capacity. 
 To calculate the vehicle travel time and interval travel speed, in terms of technical implementation, the bayonet system is required to find out the last passing record of the same license plate with the closest time interval for each vehicle passing record, so as to complete the calculation. If the bayonet corresponding to the previous traffic record found by chance is far away from the bayonet corresponding to the current record, it is completely impossible for the vehicle to pass through the two bayonets within this time interval, which means that the vehicle cover is found. card behavior. There is no need to go into details about the danger of card deck behavior. Once the card deck behavior is discovered, the bayonet system can notify relevant personnel in various ways to intercept it. 
Traffic flow analysis
Traditionally, when analyzing traffic flow, data from geomagnetic, microwave and other detection sources are often used, but these detection sources have one thing in common, that is, they cannot Check the license plate number. This limits the application scenarios of traditional traffic analysis, which can only analyze a single road section/intersection, and cannot form a global traffic analysis. 
The bayonet system has a natural advantage over geomagnetic, microwave and other detection equipment because it records the vehicle number plate. Based on the traffic analysis of the bayonet system, the distribution of the occurrence and attraction of motor vehicles in each district of the city can be calculated, and various analysis applications such as travel destination analysis and travel path analysis can be realized, and then the urban hotspot areas can be found out for managers. Provide decision-making reference, better optimize the road network mechanism, and plan more reasonable road network parameters. 
Image Analysis
The passing data recorded by the bayonet system is very simple in structure. From the perspective of the storage space occupied, the part other than the picture is only a few hundred bytes , but each image is hundreds of K in size. With the promotion of high-definition bayonet, it is believed that the picture size will become larger and larger, occupying more and more storage space. 
Traditionally, the pictures taken by the bayonet system are only used for manual review, which is used to confirm the vehicle at the final stage. Considering the huge passing data, I believe that most of the pictures cannot be used effectively, occupying a huge storage space in vain. 
 In fact, there is a lot of information in the pass-through picture of the bayonet, which cannot be effectively identified by the bayonet device itself, such as vehicle brand, vehicle model, whether the annual inspection logo is pasted, and whether the driver wears a seat belt. , whether to make a phone call while driving, etc. 
As image analysis technology becomes more and more intelligent, it becomes more and more important to establish post-image analysis applications on the passing pictures collected by the bayonet system, and use image analysis technology to detect illegal and illegal vehicles in time. It works. For example, by comparing with the vehicle information registered in the vehicle driving management system, it is possible to check whether the license plate is duplicated. 
Using post-image analysis, without changing the existing bayonet equipment, more vehicle features can be excavated, which facilitates the realization of more applications, effectively utilizes the existing bayonet equipment, reduces the Unnecessary bayonet rebuilding inputs. 
The application of bayonet big data is more abundant
Using advanced massive data processing technology, combined with the current urban development needs, fully tap the value of the data collected by the bayonet system, for the construction of a safe city Laying a solid foundation, which is the current range of applications for bayonet-based systems. 
In addition to traditional inspection and prevention applications, based on the massive passing data of the bayonet application, applications such as investigation of involved vehicles, travel time calculation, traffic flow analysis, and image analysis can be realized. The investigation of vehicles involved in the case also includes various application forms such as analysis of accompanying vehicles, analysis of vehicles entering and leaving the crime scene, and analysis of abnormal vehicles; from the calculation of travel time, applications such as interval speed measurement and set detection can be derived; traffic flow analysis is a checkpoint The important application direction of the system can provide many application forms that cannot be supported by traditional detection equipment; image analysis can make full use of the pictures captured by the bayonet equipment to provide more complete vehicle feature information. 
</div>

New Points of Fusion Application of Bayonet and Big Data

Guess you like