A method and system for electronically capturing illegal parking based on big data vehicle trajectories

Abstract

This application belongs to the field of vehicle illegal parking capture technology. A method for electronically capturing illegal parking based on big data vehicle trajectories includes: acquiring a first set of images of a vehicle entering a no-parking zone at a preset first interval; extracting features from the first set of images to obtain a first set of feature information; comparing the first set of feature information with violation image information stored in a database; if a match is not found, generating a message to persuade the vehicle to leave the no-parking zone and sending it to the vehicle owner, while storing the first set of feature information in the database, thus obtaining the first set of violation image information; after a preset second interval, acquiring four more images of the vehicle in the no-parking zone to obtain a second set of violation images; extracting features from the second set of images to obtain a second set of feature information; comparing the second set of feature information with the violation image information stored in the database to generate violation information. This method features strong real-time performance and more accurate violation judgment.

Description

Technical Field

[0001] This application belongs to the field of vehicle illegal parking detection technology, and in particular relates to an electronic detection method and system for illegal parking based on big data vehicle trajectories. Background Technology

[0002] With the acceleration of urbanization and the rapid increase in vehicle ownership, urban road traffic management is facing unprecedented challenges, with illegal parking being a particularly prominent issue. Traditional manual patrol methods are no longer adequate for current management needs due to their inefficiency and limited coverage.

[0003] Furthermore, existing parking enforcement systems have significant shortcomings when processing large amounts of vehicle data. They often fail to analyze and process data in a timely and effective manner, resulting in the inability to monitor and manage illegal parking in a timely manner. More seriously, existing systems also suffer from the problem of duplicate penalties, which not only increases the burden on car owners but also reduces the fairness and efficiency of traffic management.

[0004] Therefore, modern urban traffic management urgently needs a new technological solution to improve the efficiency of monitoring illegal parking, reduce human intervention, achieve rapid identification and processing of violations, and avoid repeated penalties, in order to meet the growing traffic management needs. Summary of the Invention

[0005] Therefore, it is necessary to provide a method and system for electronically capturing illegal parking based on big data vehicle trajectories to solve the aforementioned technical problems.

[0006] Firstly, this application provides a method for electronically capturing illegally parked vehicles based on big data vehicle trajectories, the method including:

[0007] According to a preset first interval time, the first set of violation images of a vehicle entering a no-parking section is acquired, wherein the first set of violation images includes four images taken sequentially at the first interval time.

[0008] Feature extraction is performed on the first set of violation images to obtain the first set of feature information;

[0009] The first set of feature information is compared with the violation image information stored in the database. If the comparison fails, a warning to leave the no-parking section is generated and sent to the vehicle owner. The first set of feature information is then stored in the database to obtain the first violation image information.

[0010] After a preset second interval, four more images of the vehicle in the no-parking zone are acquired to obtain a second set of violation images, wherein the second set of violation images includes four images taken sequentially at the first interval.

[0011] Feature extraction is performed on the second set of violation images to obtain the second set of feature information;

[0012] The second set of feature information is compared with the violation image information stored in the database. When the second set of feature information is successfully compared with the first violation image information, violation information is generated.

[0013] In some feasible methods, the step of acquiring the first violation image of a vehicle entering a no-parking zone according to a preset first interval includes:

[0014] Obtain the regulations regarding the duration of illegal parking in areas where parking is prohibited;

[0015] According to the aforementioned regulations, the preset first interval time is determined;

[0016] The system acquires the first violation image taken when the vehicle enters a no-parking zone, and the second, third, and fourth violation images taken sequentially at a first time interval.

[0017] The first set of violation images is obtained by combining the first violation image, the second violation image, the third violation image, and the fourth violation image.

[0018] In some feasible methods, the step of extracting features from the first set of violation images to obtain the first set of feature information includes:

[0019] For the first set of violation images, feature extraction is performed to obtain the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time;

[0020] Obtain the device code information, device coordinate information, road location information, and road code information for capturing the first set of violation images;

[0021] The first set of feature information is obtained by combining the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time, as well as the device code information, device coordinate information, road location information, and road code information.

[0022] In some feasible methods, the step of comparing the first set of feature information with the violation image information stored in the database, generating a warning to leave the no-parking zone if the comparison fails, sending the warning to the vehicle owner, and storing the first set of feature information in the database to obtain the first violation image information includes:

[0023] The database stores the image information of traffic violations for the day;

[0024] The first set of feature information and the violation image information are compared to obtain the comparison result;

[0025] If the comparison is successful, illegal information will be generated;

[0026] If the comparison fails, a warning to leave the no-parking zone will be generated and sent to the vehicle owner. The first set of feature information will be stored in the database to obtain the first violation image information.

[0027] In some feasible methods, the step of acquiring four more images of the vehicle in the no-parking zone after a preset second interval to obtain a second set of violation images includes:

[0028] Construct the preset second interval time;

[0029] After the last image of the first set of violation images is formed, timing is performed according to the second interval time. After the second interval time, four images of the vehicle in the no-parking section are obtained again in the order of the first interval time to obtain the second set of violation images.

[0030] In some feasible methods, the step of extracting features from the second set of violation images to obtain the second set of feature information includes:

[0031] For the second set of violation images, feature extraction is performed to obtain the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time;

[0032] Obtain the device code information, device coordinate information, road location information, and road code information for capturing the second set of violation images;

[0033] The first set of feature information is obtained by combining the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time, as well as the device code information, device coordinate information, road location information, and road code information.

[0034] In some feasible methods, the step of comparing the second set of feature information with the violation image information stored in the database, and generating violation information when the second set of feature information successfully matches the first violation image information, includes:

[0035] The second set of feature information is compared with the violation image information stored in the database on that day to obtain the comparison result;

[0036] When the second set of feature information is successfully compared with the first violation image information, the first two images in the first violation image information and the last two images in the second set of violation images are extracted to obtain four violation images.

[0037] Based on the four images of the traffic violation, generate violation information;

[0038] The second set of violation images is stored in the database to obtain the second set of violation image information.

[0039] In some feasible methods, after the step of storing the second set of violation images in the database to obtain the second set of violation image information, the method includes:

[0040] After obtaining the second set of violation images at a preset second interval, four more images of the vehicle in the no-parking zone are obtained to obtain the third set of violation images.

[0041] Feature extraction is performed on the third set of violation images to obtain the third set of feature information;

[0042] The third set of feature information is compared with the violation image information stored in the database. When the third set of feature information is successfully compared with the first violation image information and the second violation image information, the third set of violation images is discarded.

[0043] Secondly, this application provides an electronic system for capturing illegal parking based on big data vehicle trajectories, applying the aforementioned method for electronically capturing illegal parking based on big data vehicle trajectories. The system includes:

[0044] The image acquisition module is used to acquire the first set of violation images of a vehicle entering a no-parking section according to a preset first interval time. The first set of violation images includes four images taken sequentially at the first interval time.

[0045] The image recognition module is used to extract features from the first set of violation images to obtain the first set of feature information;

[0046] The big data trajectory analysis module is used to compare the first set of feature information with the violation image information stored in the database. If the comparison fails, it generates a warning to leave the no-parking section information and sends it to the vehicle owner. The first set of feature information is stored in the database to obtain the first violation image information.

[0047] The image acquisition module is also used to acquire four more images of the vehicle in the no-parking section after a preset second interval, to obtain a second set of violation images, wherein the second set of violation images includes four images taken sequentially at the first interval.

[0048] The image recognition module is also used to extract features from the second set of violation images to obtain the second set of feature information;

[0049] The violation processing module is used to compare the second set of feature information with the violation image information stored in the database. When the second set of feature information is successfully compared with the first violation image information, violation information is generated.

[0050] Thirdly, this application provides a computer program that, when executed by a processor, implements the steps of the aforementioned method for electronically capturing illegally parked vehicles based on big data vehicle trajectories.

[0051] Beneficial Effects: This application provides a method for electronically capturing illegal parking based on vehicle trajectories using big data. The method includes: acquiring a first set of violation images of a vehicle entering a no-parking zone at a preset first interval, wherein the first set of violation images includes four images taken sequentially at the first interval. Feature extraction is performed on the first set of violation images to obtain a first set of feature information; the first set of feature information is compared with violation image information stored in a database; if the comparison fails, a message to persuade the vehicle to leave the no-parking zone is generated and sent to the vehicle owner, and the first set of feature information is stored in the database, resulting in the first violation image information; after a preset second interval, four more images of the vehicle in the no-parking zone are acquired to obtain a second set of violation images, wherein the second set of violation images includes four images taken sequentially at the first interval. Feature extraction is performed on the second set of violation images to obtain a second set of feature information; the second set of feature information is compared with violation image information stored in the database; if the second set of feature information successfully matches the first violation image information, violation information is generated. The above method uses road cameras to capture data on illegally parked vehicles, and determines whether illegal parking exists based on the parking time. The violation processing procedure determines whether this is the vehicle's first illegal parking. If it is, a message is generated to remind the vehicle owner, thus promptly preventing further violations and avoiding unnecessary fines. This improves user experience and alleviates traffic hazards caused by indiscriminate parking. Simultaneously, the violation processing procedure uses information about the road segment where the device is located, the vehicle's historical illegal parking data, and the vehicle's travel trajectory to determine whether the vehicle has been parked in the same location for a certain period. If it has already been penalized, it will not be fined again, avoiding unnecessary fines for the user. Compared to existing methods, the electronic illegal parking capture method and system proposed in this application have lower latency, stronger real-time performance, and more accurate violation judgment when processing massive amounts of traffic violation data. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 This is a flowchart of an electronic method for capturing illegal parking based on big data vehicle trajectories in one embodiment;

[0054] Figure 2 This is a schematic diagram of the electronic capture structure of an electronic capture system for illegal parking based on big data vehicle trajectories in one embodiment.

[0055] Figure 3 This is a schematic diagram of the electronic illegal parking capture procedure of an electronic illegal parking capture system based on big data vehicle trajectory in one embodiment. Detailed Implementation

[0056] To facilitate understanding of this application, a more complete description will be provided below with reference to the accompanying drawings, which illustrate embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of this application will be thorough and complete.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The term "and / or" as used herein includes any and all couplings of one or more of the associated listed items.

[0058] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish one element from another.

[0059] The following explanations of the terms used in this application are provided to aid in understanding the proposed solution:

[0060] A histogram is a commonly used statistical tool in image processing and computer vision, used to describe the distribution of pixel values ​​in an image.

[0061] The rapid pace of urbanization has led to a surge in population and vehicle numbers, posing a severe challenge to urban road traffic management. Among numerous traffic problems, illegal parking has become a focal point due to its prevalence and serious impact on urban traffic flow. Illegal parking not only obstructs traffic flow and increases congestion but can also cause traffic accidents, negatively affecting urban order and residents' quality of life.

[0062] Traditional management of illegal parking relies primarily on manual patrols, a method with significant limitations. Patrol personnel need to walk the streets, checking for illegally parked vehicles, recording the violations, and issuing penalties. This method is not only resource-intensive but also inefficient and unable to cover all areas requiring monitoring. Furthermore, the randomness and subjectivity of manual patrols can lead to unfair enforcement and regulatory loopholes.

[0063] With technological advancements, some cities have begun using electronic monitoring systems to assist in managing illegal parking. However, these systems have also revealed numerous problems in practical application. First, existing electronic monitoring systems often struggle to process large volumes of vehicle data, failing to analyze and process the data accurately and in real time. This results in delays in identifying illegal parking, hindering timely penalties and reducing management efficiency. Second, existing systems have deficiencies in data management and analysis, frequently leading to duplicate penalties, causing unnecessary inconvenience and financial losses for vehicle owners. Furthermore, these systems often lack the ability to deeply analyze historical vehicle violation data, making it difficult to effectively identify and prevent repeat violations.

[0064] With the increasing number of vehicles, the problem of illegal parking has become more complex. Urban traffic management departments need more intelligent and efficient technological means to address this challenge. They need systems capable of processing and analyzing massive amounts of data to achieve rapid identification and accurate management of illegal parking. At the same time, the system also needs to be highly automated and intelligent, reducing human intervention and improving the fairness and consistency of law enforcement. In short, the field of urban traffic management urgently needs innovative technological solutions to address the increasingly serious problem of illegal parking.

[0065] In summary, this application provides an electronic method for capturing illegal parking based on big data vehicle trajectories. It combines advanced technologies such as big data processing, cloud computing, the Internet of Things, machine vision, and image processing. Through image capture, image recognition, violation determination, and real-time analysis of vehicle trajectories, it achieves automatic capture, identification, and processing of illegal parking behavior. This system can utilize big data platforms and image processing technology to rapidly process and deeply mine massive amounts of vehicle data, extracting key information such as vehicle characteristics, location, and driving trajectory, and intelligently processing illegal parking behavior based on this information. Simultaneously, the system can upload captured evidence of violations to a data center in real time, providing traffic management departments with timely and accurate enforcement evidence.

[0066] like Figure 1 As shown, in a first aspect, this application provides a method for electronically capturing illegally parked vehicles based on big data vehicle trajectories, the method comprising:

[0067] S100, according to the preset first interval time, acquires the first set of violation images of the vehicle entering the no-parking section.

[0068] The first set of violation images includes four images taken sequentially at a first time interval.

[0069] Specifically, acquiring images of vehicles illegally entering no-parking zones at preset time intervals may include the following steps:

[0070] S101, Obtain the regulations regarding the duration of illegal parking in the area where no-parking zones are located. The system needs to query and obtain the regulations for specific no-parking zones to understand the specific duration of illegal parking in that area; that is, how long an illegal parking session must exceed to be considered a violation. These regulations will determine the length of the preset time intervals in subsequent steps.

[0071] S102, according to the aforementioned regulations, a preset first interval time is determined. Specifically, based on the illegal parking duration regulations obtained in step S101, the system will determine a preset time interval, i.e., the first time interval, for example, 1 minute. This time interval will be used for subsequent capture of violation image sets to ensure that vehicle violations in prohibited parking areas can be captured.

[0072] S103, acquire the first violation image taken when the vehicle enters the no-parking zone, and the second, third, and fourth violation images taken sequentially at a first time interval. When the vehicle enters the no-parking zone, the system will start taking the first violation image. Then, the system will take the second, third, and fourth violation images sequentially at a preset first time interval. These images will be recorded as violations.

[0073] S104, the first violation image, the second violation image, the third violation image, and the fourth violation image are combined to obtain the first set of violation images. Finally, the system combines these four violation images into a complete set of violation images, namely the first set of violation images, for subsequent analysis and processing of illegal parking behavior.

[0074] By capturing four images consecutively using the above method, the system can provide a complete record of the violation, enhancing the integrity and reliability of the evidence. It reduces misjudgments; capturing multiple images helps distinguish between temporary parking and illegal parking, reducing incorrect penalties due to misjudgment. Furthermore, the automated image acquisition and combination process reduces human intervention, improving enforcement efficiency and response speed.

[0075] For example, suppose the permitted parking time in a no-parking zone is 3 minutes. The system will set the preset first interval to 3 minutes.

[0076] S101: The system found that the permitted duration for illegal parking on this section of road is 3 minutes.

[0077] S102: The system determines the preset first interval time to be 3 minutes.

[0078] S103: A car enters a no-parking zone at 9:00 AM, and the system takes the first image at 9:00 AM. Then, the system takes the second, third, and fourth images at 9:01 AM, 9:02 AM, and 9:03 AM, respectively.

[0079] S104: The system combines these four images into a violation image group, recording the vehicle's illegal parking behavior in the no-parking zone from 9:00 to 9:03.

[0080] S200, perform feature extraction on the first group of violation images to obtain the first group of feature information.

[0081] Specifically, obtaining the first set of feature information may include the following steps:

[0082] S201, for the first set of traffic violation images, feature extraction is performed to obtain the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time. Specifically, the system uses image recognition technology to extract key vehicle feature information from the four traffic violation images. This includes the license plate number (used for vehicle identification and subsequent violation processing), the vehicle logo and vehicle type (used to further confirm vehicle identity), the vehicle color (as an additional identification feature), and the shooting time (recording the specific moment the violation occurred).

[0083] S202, acquire the device code information, device coordinate information, road location information, and road code information of the device that captured the first set of violation images. Further, collect relevant information about the device that captured the violation images, including the device code (uniquely identifying each camera or monitoring device), device coordinates (specific geographical location, such as latitude and longitude), road location (name of the street or road segment where the violation occurred), and road code (potentially used for internal identification in the traffic management system).

[0084] S203, the system combines the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time, along with the device code information, device coordinate information, road location information, and road code information, to obtain the first set of feature information. The system then integrates all extracted vehicle and device feature information to form a complete set of feature information. This set of information will be used for recording, analyzing, and processing illegal parking.

[0085] The methods described above allow for detailed recording of vehicle and recording equipment information, ensuring the accuracy and fairness of traffic violation processing. Enhanced data traceability, with complete characteristic information enabling the tracing and verification of every violation, increases data transparency. Improved traffic management, as this information can be used to analyze patterns and trends in violation behavior, thereby optimizing traffic management and enforcement strategies.

[0086] S300: The first set of feature information is compared with the violation image information stored in the database. If the comparison fails, a warning to leave the no-parking section is generated and sent to the vehicle owner. The first set of feature information is stored in the database to obtain the first violation image information.

[0087] Specifically, obtaining the image information of the first traffic violation may include the following steps:

[0088] S301, the system stores the traffic violation image information for the day in the database. Specifically, the system stores the traffic violation image information for the day in a central database. This database contains detailed information on all traffic violations of the day, including vehicle characteristic information and the specific details of the violation.

[0089] S302, the first set of feature information and the violation image information are compared to obtain a comparison result. Specifically, the system compares the extracted first set of feature information with the violation image information stored in the database. The comparison process may involve information such as license plate number, vehicle characteristics, violation time and location, to determine whether the vehicle has a previous violation record.

[0090] If the comparison is successful, a violation notice will be generated. If the comparison shows that the vehicle has a previous traffic violation record, the system will generate a violation notice, meaning the vehicle has violated parking regulations and will be subject to appropriate penalties.

[0091] If the comparison fails, a warning to leave the no-parking zone will be generated and sent to the vehicle owner. The first set of feature information will also be stored in the database, resulting in the first violation image information. Alternatively, if the comparison shows the vehicle has no previous violation records, the system will generate a warning to leave the no-parking zone and send it to the vehicle owner. Simultaneously, the system will store the first set of feature information in the database, marking it as the first violation image information for subsequent comparisons.

[0092] The above methods automate the comparison and information processing workflow, reducing manual operations and improving the efficiency of handling traffic violations. They also reduce misjudgments caused by incorrect information through precise feature comparison. Timely response is crucial; for vehicles committing their first violation, promptly sending a warning message can prevent further violations, reducing traffic congestion and potential safety hazards. Storing violation image information for the current day adheres to the legal requirement of only one penalty per location per day.

[0093] For example, suppose a car is illegally parked in a no-parking zone. The system extracts the first set of feature information and performs the following operations:

[0094] S301: The system stores information on all traffic violations for the day in the database.

[0095] S302: The system compares the first set of extracted feature information with the violation image information in the database and finds no matching record, indicating that this is the vehicle's first violation.

[0096] Match failed: The system generated a warning message, which could read: "Dear vehicle owner, your vehicle is illegally parked in a no-parking zone on a certain street. Please move your vehicle immediately to avoid being penalized." This message was sent to the vehicle owner via SMS.

[0097] Information storage: At the same time, the system stores the first set of feature information of this violation in the database as a record of the first violation image information.

[0098] Through this process, the system not only notifies car owners in a timely manner, but also provides data support for future traffic management.

[0099] S400: After a preset second interval, acquire four more images of the vehicle in the no-parking zone to obtain a second set of violation images.

[0100] The second set of violation images includes four images taken sequentially at the first time interval.

[0101] Specifically, obtaining the second set of violation images may include the following steps:

[0102] S401, Establish a preset second interval time. The system sets a preset second interval time based on traffic regulations or specific operating procedures. This time interval is used to determine how long to wait after the first violation image is captured before taking another image to confirm whether the vehicle is still illegally parked.

[0103] S402, after the last image of the first set of violation images is formed, timing is started according to the second interval time. After the second interval time, four images of the vehicle in the no-parking zone are acquired again in the first interval time sequence to obtain the second set of violation images. Specifically, after the first set of violation images is captured and the preset second interval time has elapsed, the system will restart the image capturing process. The system will capture four new images in the first interval time sequence (e.g., once per minute). These images will constitute the second set of violation images for further confirmation of the vehicle's violation behavior.

[0104] By employing the above methods, the accuracy of traffic violation determination is improved. By capturing images at different times, the system can more accurately determine whether a vehicle is continuously illegally parked. False alarms are reduced; if a vehicle has already left after the second interval, the system will not incorrectly process its violation. Enforcement is strengthened; for vehicles with continuous violations, the second set of images provides additional evidence, allowing law enforcement agencies to take stricter measures.

[0105] For example, suppose a car is illegally parked in a no-parking zone. The system takes the first set of images of the violation between 9:00 AM and 9:03 AM and sends a warning message to the car owner to move the car. However, the car owner does not move the car within the specified time.

[0106] After 9:03 (the time the last image of the first set of violation images was captured), the system waits for 10 minutes, from 9:03 to 9:13. At 9:13, the system begins capturing four more images at a rate of one per minute, specifically at 9:13, 9:14, 9:15, and 9:16. This yields a second set of violation images. The system combines these four newly captured images to confirm the vehicle's continued violation of parking rules in the no-parking zone. Through this process, the system ensures effective monitoring and handling of persistent illegal parking and provides strong evidence for law enforcement.

[0107] S500, extract features from the second group of violation images to obtain the second group of feature information.

[0108] Obtaining the second set of feature information may include the following steps:

[0109] S501, perform feature extraction on the second group of violation images to obtain the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time.

[0110] S502, acquire the device code information, device coordinate information, road location information, and road code information for capturing the second set of violation images.

[0111] S503, combine the vehicle's license plate number, vehicle logo, vehicle type, vehicle color, and shooting time, as well as the device code information, device coordinate information, road location information, and road code information to obtain the first set of feature information.

[0112] It should be noted that the steps for obtaining the second set of violation images are the same as those for obtaining the first set of violation images, and will not be repeated here.

[0113] S600, compare the second set of feature information with the violation image information stored in the database. When the second set of feature information is successfully compared with the first violation image information, generate violation information.

[0114] Specifically, generating illegal information may include the following steps:

[0115] S601, the second set of feature information is compared with the traffic violation image information stored in the database for that day to obtain a comparison result. The system compares the feature information of the second set of traffic violation images (such as license plate number, vehicle type, vehicle color, etc.) with the traffic violation image information stored in the database for that day. This process aims to verify whether the vehicle has had a traffic violation record before the same day and to confirm the vehicle's identity.

[0116] When the second set of feature information is successfully compared with the first violation image information, the first two images in the first violation image information and the last two images in the second set of violation images are extracted to obtain four violation images.

[0117] S602, Based on the four violation images, generate violation information. Once a match is successful, the system extracts the first two images from the first set of violation images and the last two images from the second set of violation images, for a total of four violation images. These four images will serve as evidence of the illegal parking. The system will generate violation information based on these images, including detailed information such as vehicle characteristics, violation time, and location.

[0118] S603, the second set of violation images is stored in the database to obtain the second violation image information. Finally, the system stores the second set of violation images in the database and marks it as the second violation image information. This provides a complete record for future queries, analysis, and law enforcement.

[0119] It should be noted that the illegal information generated in step S302 based on the first set of feature information is the same as that in step S602, and will not be described again.

[0120] It should also be noted that after storing the second set of violation images in the database to obtain the second set of violation image information, the process includes:

[0121] S604, after obtaining the second set of violation images at a preset second interval, four more images of the vehicle in the no-parking section are obtained to obtain the third set of violation images.

[0122] S605, extract features from the third set of violation images to obtain the third set of feature information.

[0123] S606, compare the third set of feature information with the violation image information stored in the database. When the third set of feature information is successfully compared with the first violation image information and the second violation image information, discard the third set of violation images.

[0124] It should be noted that after the second set of violation images is acquired and processed, the system will wait for another preset second interval. Afterward, the system will acquire four new images from the no-parking zone, forming the third set of violation images. This step aims to monitor whether the vehicle continues to park illegally. The third set of violation images is processed in the same way as the second set, and will not be repeated here. If the third set of violation images matches the first and second sets, it indicates that the vehicle has committed repeated violations at different times. According to some traffic management regulations, multiple violation records may be allowed within a certain period, but after a certain number, subsequent violations may be considered duplicates. Therefore, the third set of violation images will be discarded without further penalty, avoiding the problem of repeated penalties at the same location.

[0125] For example, suppose a car is illegally parked in a no-parking zone. The system takes the first set of violation images between 9:00 and 9:03 a.m., the second set of violation images between 10:00 and 10:03 a.m., and the third set of violation images between 11:00 and 11:03 a.m.

[0126] S604: The system acquired a third set of violation images after a preset time following the acquisition of the second set of violation images.

[0127] S605: The system extracts features from the third set of violation images and obtains the third set of feature information.

[0128] S606: The system compares the third set of feature information with the first and second violation images in the database. The comparison is successful. According to regulations, the third set of violation images is discarded and the vehicle owner is not penalized again.

[0129] In one embodiment, after the second set of violation images is acquired and processed, the system waits for another preset second interval. Afterward, the system acquires images again in the no-parking zone, including the following steps:

[0130] Obtain the first image after the preset second interval time;

[0131] Feature extraction is performed on the first image, and the extracted features are compared with the second violation image information in the database to obtain the comparison result;

[0132] If the comparison result is the same vehicle, the setting of generating four images to form a violation image group will be discarded, and the image will be acquired again at the next second interval.

[0133] If the comparison result indicates that the vehicle is not the same, then three images are taken sequentially at the first interval to form a violation image group consisting of four images.

[0134] Using the above method, the features of the first image can be used to make a judgment before the third set of violation images are formed. If the features of the first image can be successfully compared with the violation image information in the database, the three images taken in sequence will become meaningless when compared again at the first interval. Therefore, the subsequent program that requires taking three more images is discarded. At the same time, the first image is discarded and does not need to be stored in the database, thus avoiding increasing the burden on the computer room.

[0135] In one embodiment, an electronic method for capturing illegal parking based on big data vehicle trajectories further includes the following steps:

[0136] Based on the first set of feature information, the license plate of the vehicle is marked with features to obtain the first label of the license plate, and stored in the database;

[0137] For the first image in the second group of violation images, license plate features are extracted to obtain the second label of the license plate;

[0138] The first tag in the database is compared with the license plate features corresponding to the second tag to obtain the comparison result;

[0139] If the comparison result shows the same license plate, then no further comparison of the characteristics of other vehicles will be made, and the vehicle will be determined to be the same vehicle.

[0140] The above method effectively reduces the number of features compared, requiring only license plate comparison. In this case, the second, third, and fourth images taken at the first time interval are used as supplementary evidence, again requiring only license plate recognition. That is, after the first image in the second set of violation images is generated, if the vehicle leaves, the first image in the second set of violation images, along with the first two images in the first set of violation images, are used together as evidence of the violation to generate violation information. It can be understood that if the vehicle is photographed again at the second time interval and the same vehicle information is obtained, it indicates that the vehicle was illegally parked. Therefore, even if the second, third, and fourth images in the second set of violation images are missing, it is still determined that the vehicle was illegally parked.

[0141] It should be noted that retrieving a vehicle's license plate involves the following steps:

[0142] Build the initial large model;

[0143] The initial large model is input into already marketed vehicles for training, resulting in the target large model. The purpose of training the initial large model is to be able to specifically identify license plate locations and effectively extract license plate information for different vehicle models. The architecture of the large model is a conventional large model architecture, and this application does not modify the architecture of the large model.

[0144] The first image from the second set of violation images is input into the target large model to obtain the vehicle's license plate information features and form the second label;

[0145] The second label is compared with the first label formed by the license plate corresponding to the first set of feature information to obtain the comparison result.

[0146] Using the method described above, a large model is employed to selectively identify the license plate information features of the first image in the second set of violation images, while discarding other features. This allows for comparison of the license plate information features corresponding to the second label and the first label; comparing only this feature is sufficient to determine if it is the same vehicle. Furthermore, this simplifies the time required for vehicle identification and avoids wasting resources.

[0147] In one embodiment, if a vehicle enters a no-parking zone and its license plate information is obscured by a following vehicle or a preceding vehicle that violated the parking rule, the following steps are included:

[0148] Acquire several frames of images prior to the first set of violation images; wherein, the several frames of images can be acquired at set intervals. For example, the time interval between frames can be set to 2 seconds, 4 seconds, 6 seconds, etc., that is, after acquiring one frame of image, the next frame of image is acquired after a set number of seconds.

[0149] The system performs feature recognition on several frames of images in chronological order to obtain the vehicle's license plate information. The chronological order is derived from the first set of violation images, proceeding sequentially. For example, if the first image in the first set of violation images was created at 9:00:00, then if the license plate information feature is not identified in the frame created at 8:59:58, the next frame created at 8:59:56 will be identified two seconds earlier. Once the license plate information feature is identified at 8:59:56, the video from 8:59:56 to 9:00:00 is used as an additional feature of the first set of violation information. In other words, when storing the data in the database, both the first set of violation images and the video from 8:59:56 to 9:00:00 are stored simultaneously to ensure the integrity of the violation evidence.

[0150] The vehicle's license plate information is added to the first set of feature information formed from the first set of violation images. Because the vehicle is obscured by a following or preceding vehicle, the license plate information is lost when the first set of violation images is formed. Therefore, the license plate information needs to be added to complete the vehicle violation information.

[0151] The first set of feature information is compared with the violation image information stored in the database.

[0152] The above method ensures that even if a vehicle is parked in a no-parking zone and its license plate is obscured by other vehicles or objects, the license plate captured during the vehicle's movement can still be used as a supplementary feature to the first set of feature information and compared with the violation images stored in the database.

[0153] It should be noted that if a vehicle is parked in a no-parking zone and its license plate is obscured, the license plate may still not be able to be obtained when acquiring the second set of violation images at the preset second interval. Therefore, the second set of violation images may include the following steps:

[0154] Perform grayscale conversion on any image in the first group of violation images to obtain the first grayscale image;

[0155] Based on the first grayscale image, histogram calculation is performed to obtain the first histogram;

[0156] The first image in the second set of violation images is converted to grayscale to obtain the second grayscale image;

[0157] Based on the second grayscale image, histogram calculation is performed to obtain the second histogram;

[0158] The similarity between the first histogram and the second histogram is compared to obtain the comparison result; the comparison can be performed using methods such as histogram intersection, chi-square test, and Bach distance.

[0159] If the comparison result is greater than a preset threshold, then the vehicle in the first histogram and the second histogram are determined to be the same vehicle.

[0160] The above method effectively identifies whether a vehicle is the same one even when it is parked in a no-parking zone and its license plate is obscured. Example

[0161] This application provides a method for electronically capturing illegally parked vehicles based on big data vehicle trajectories, the method comprising:

[0162] S1. When a vehicle enters a no-parking zone, the road camera automatically captures a set of images and continuously records data on illegal parking of motor vehicles at set time intervals.

[0163] The road camera captures 4 images per group, with a 10-minute time interval between groups and a 1-minute time interval within each group.

[0164] S2. Input the image data captured in S1 into the image recognition module to extract the vehicle feature information of the violating vehicle.

[0165] S3. Determine whether illegal parking exists based on the vehicle's parking time. If so, generate four images with specified time intervals and send the image data and violation data to the violation processing procedure.

[0166] The illegal data includes equipment information and vehicle information. Equipment information mainly includes equipment name, equipment code, equipment coordinates, road location, and road code; vehicle information mainly includes license plate number, vehicle logo, vehicle type, vehicle color, and time of violation.

[0167] S4. The violation processing procedure determines the vehicle's violation status based on the road section information where the equipment is located, the vehicle's historical illegal parking data, and the vehicle's trajectory.

[0168] S5. When a vehicle is determined to be illegally parked for the first time by the violation processing procedure, an SMS or message reminder will be generated and sent to the communication device of the vehicle owner who is bound to the vehicle, notifying the owner to leave within 10 minutes.

[0169] S6. When a vehicle is determined by the violation processing procedure to be a non-first-time illegal parking, a violation is generated, and the violation data of the illegally parked vehicle is stored in the Oracle database, while the composite image data is stored in the FastDFS file system.

[0170] The composite image consists of four images: the first two images from the first group and the last two images from the second group. These four images represent evidence of the violation: a message was sent to the user upon the initial violation, and if the user did not leave within 10 minutes, the composite image recorded the violation as evidence.

[0171] Furthermore, the specific steps in step S4 for determining vehicle traffic violations are as follows:

[0172] S4.1 The violation processing procedure queries the Oracle database for the vehicle's illegal parking data for the day based on the equipment information and vehicle information. If the license plate number, vehicle type, road name, and road code are the same, but the equipment name and equipment code are different, then this violation data is filtered to avoid cross-enforcement, and the procedure continues to step S4.2.

[0173] The fields for querying equipment information are equipment name, equipment code, road name, and road code; the fields for querying vehicle information are license plate number and vehicle type.

[0174] S4.2 The violation processing procedure queries the Oracle database for the vehicle's illegal parking data for the day based on the license plate number and vehicle type. If no results are found within the specified time range, proceed to step S6; otherwise, continue to step S4.3.

[0175] S4.3 The violation processing procedure queries the results of step S4.2 based on the road location and road code. If no data results are found, proceed to step S6; otherwise, continue to step S4.4.

[0176] S4.4 The violation processing procedure uses the license plate number and vehicle type as conditions to query the trajectory information of the illegally parked vehicle on the same day in the big data trajectory analysis. If the vehicle trajectory information is found within the specified time range, step S6 is executed; otherwise, the violation penalty will not be repeated, and this violation data will be filtered.

[0177] Among these measures, vehicles will only be penalized once per day at the same location for any traffic violation.

[0178] like Figure 2 and Figure 3 As shown, in a second aspect, this application provides an electronic system for capturing illegal parking based on big data vehicle trajectories, which applies the aforementioned method for capturing illegal parking based on big data vehicle trajectories. The system includes an image acquisition module, an image recognition module, a violation processing program module, a big data trajectory analysis module, and a data and image storage module.

[0179] The image acquisition module is configured to acquire a first set of violation images of a vehicle entering a no-parking zone at a preset first interval, wherein the first set of violation images includes four images taken sequentially at the first interval. The image acquisition module is also configured to acquire four more images of the vehicle in the no-parking zone after a preset second interval, resulting in a second set of violation images, wherein the second set of violation images includes four images taken sequentially at the first interval. In other words, data on illegal parking of motor vehicles entering the no-parking zone is continuously captured at set time intervals.

[0180] The image recognition module is used to extract features from the first group of violation images to obtain a first set of feature information. The image recognition module is also used to extract features from the second group of violation images to obtain a second set of feature information. In other words, it identifies vehicle features (such as license plate number, vehicle logo, vehicle type, vehicle color, etc.) in the images captured by the image acquisition module.

[0181] The violation processing module is used to compare the second set of feature information with the violation image information stored in the database. When the second set of feature information successfully matches the first violation image information, violation information is generated. In other words, the system determines the current vehicle's violation status and generates violation information based on the road segment information where the device is located, the vehicle's historical illegal parking data, and the vehicle's trajectory.

[0182] The big data trajectory analysis module is used to compare the first set of feature information with the violation image information stored in the database. If the comparison fails, it generates a warning to leave the no-parking zone and sends it to the vehicle owner. It also stores the first set of feature information in the database, thus obtaining the first violation image information. In other words, based on license plate number, vehicle type, and time, big data analysis is used to determine if the illegally parked vehicle has a driving trajectory and whether it has been consistently parked in the same location.

[0183] The data and image storage module is used to store the violation image information from the big data trajectory analysis module. Specifically, it stores the violation data of illegally parked vehicles in an Oracle database and the synthesized image data in a FastDFS file system. Historical illegal parking data for vehicles can be retrieved from the Oracle database.

[0184] Thirdly, this application provides a computer program that, when executed by a processor, implements the steps of the aforementioned method for electronically capturing illegally parked vehicles based on big data vehicle trajectories.

[0185] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0186] The various embodiments in this disclosure are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0187] The scope of protection of this disclosure is not limited to the embodiments described above. Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from its scope and spirit. If such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, then the intent of this disclosure also includes such modifications and variations.

Claims

1. A method for electronic capture of illegal parking based on big data vehicle trajectory, characterized in that, The method comprises: According to the preset first interval time, a first group of illegal images of the vehicle entering the no-parking road section is obtained, wherein the first group of illegal images comprises four images sequentially captured at the first interval time; Feature extraction is performed on the first group of illegal images to obtain first group of feature information; The first group of feature information is compared with the illegal image information stored in the database, and if the comparison is not successful, warning information of the no-parking road section is generated and sent to the owner of the vehicle, and the first group of feature information is stored in the database to obtain the first illegal image information; After a preset second interval time, four images of the vehicle in the no-parking road section are obtained again to obtain a second group of illegal images, wherein the second group of illegal images comprises four images sequentially captured at the first interval time; Feature extraction is performed on the second group of illegal images to obtain second group of feature information; The second group of feature information is compared with the illegal image information stored in the database, and when the second group of feature information is successfully compared with the first illegal image information, illegal information is generated; wherein After the second group of illegal images is obtained and processed, another preset second interval time is waited, and images are obtained again in the no-parking road section, comprising: The first image after the preset second interval time is obtained; Feature extraction is performed on the first image, and the extracted features are compared with the second illegal image information in the database to obtain a comparison result; If the comparison result is the same vehicle, the generation of four images to form an illegal image group is abandoned, and the next second interval time is waited to obtain images again; If the comparison result is not the same vehicle, three images are sequentially captured at the first interval time to form an illegal image group consisting of four images; The method further comprises: According to the first group of feature information, feature labeling is performed on the license plate of the vehicle to obtain a first label of the license plate, and the first label is stored in the database; Feature extraction is performed on the first image in the second group of illegal images to obtain a second label of the license plate; The first label in the database is compared with the license plate features corresponding to the second label to obtain a comparison result; If the comparison result is the same license plate, no comparison of features of other vehicles is performed, and the same vehicle is determined. 2.The electronic snapshot illegal parking method based on big data vehicle trajectory according to claim 1, wherein, The step of obtaining the first group of illegal images of the vehicle entering the no-parking road section according to the preset first interval time comprises: Obtaining the illegal parking time limit of the area where the no-parking road section is located; According to the regulation, the preset first interval time is determined; The first illegal image captured when the vehicle enters the no-parking road section, the second illegal image, the third illegal image and the fourth illegal image sequentially captured at the first interval time are obtained; The first illegal image, the second illegal image, the third illegal image and the fourth illegal image are combined to obtain the first group of illegal images. 3.The electronic snapshot illegal parking method based on big data vehicle trajectory according to claim 2, characterized in that, The step of performing feature extraction on the first group of illegal images to obtain the first group of feature information comprises: The first group of illegal image is extracted to obtain the license plate number, logo, vehicle type, vehicle color and shooting time of the vehicle; Obtain the device code information, device coordinate information, road location information and road code information of the device for shooting the first group of illegal images; Combine the license plate number, logo, vehicle type, vehicle color and shooting time of the vehicle, and the device code information, device coordinate information, road location information and road code information to obtain the first group of feature information. 4.The method of claim 1, wherein, The step of comparing the first group of feature information with the illegal image information stored in the database, if the comparison is not successful, generating the information of urging to leave the prohibited parking section and sending it to the owner of the vehicle, and storing the first group of feature information in the database to obtain the first illegal image information, comprises: Storing the illegal image information of the day in the database; Comparing the first group of feature information with the illegal image information to obtain the comparison result; If the comparison result is successful, the illegal information is generated; If the comparison result is not successful, the information of urging to leave the prohibited parking section is generated and sent to the owner of the vehicle, and the first group of feature information is stored in the database to obtain the first illegal image information. 5.The big data vehicle trajectory based electronic snap violation parking method according to claim 1, wherein, The step of obtaining the second group of illegal images after a preset second interval time comprises: Constructing the preset second interval time; After the last image of the first group of illegal images is formed, the second interval time is counted, and after the second interval time, the second group of illegal images is obtained by sequentially obtaining four images of the vehicle in the prohibited parking section at the first interval time. 6.The big data vehicle trajectory based electronic snap violation parking method according to claim 1, wherein, The step of extracting features from the second group of illegal images to obtain the second group of feature information comprises: The second group of illegal images is extracted to obtain the license plate number, logo, vehicle type, vehicle color and shooting time of the vehicle; Obtain the device code information, device coordinate information, road location information and road code information of the device for shooting the second group of illegal images; Combine the license plate number, logo, vehicle type, vehicle color and shooting time of the vehicle, and the device code information, device coordinate information, road location information and road code information to obtain the first group of feature information. 7.The electronic ticketing method based on big data vehicle trajectory for illegal parking according to claim 1, wherein, The step of comparing the second group of feature information with the illegal image information stored in the database, when the second group of feature information is compared successfully with the first illegal image information, generating the illegal information, comprises: Compare the second group of feature information with the illegal image information stored in the database to obtain the comparison result; When the second group of feature information is compared successfully with the first illegal image information, the first two images in the first illegal image information and the last two images in the second group of illegal images are extracted to obtain four illegal images; According to the four illegal images, the illegal information is generated; The second group of illegal images is stored in the database to obtain the second illegal image information. 8.The big data vehicle trajectory based electronic snap violation parking method according to claim 7, wherein, After the step of storing the second set of illegal images in the database to obtain second illegal image information, the method comprises: After a preset second interval time of obtaining the second set of illegal images, four images of the vehicle in the no-parking section are acquired again to obtain a third set of illegal images; Feature extraction is performed on the third set of illegal images to obtain third set of feature information; The third set of feature information is compared with the illegal image information stored in the database, and when the third set of feature information is successfully compared with the first illegal image information and the second illegal image information, the third set of illegal images is discarded.

9. An electronic snap-shot illegal parking system based on big data vehicle trajectory, characterized in that, The system is applied to the electronic illegal parking method based on big data vehicle trajectory according to any one of claims 1-8, and the system comprises: An image acquisition module is configured to acquire a first set of illegal images of a vehicle entering a no-parking section according to a preset first interval time, wherein the first set of illegal images comprises four images sequentially captured at the first interval time; An image recognition module is configured to perform feature extraction on the first set of illegal images to obtain first set of feature information; A big data trajectory analysis module is configured to compare the first set of feature information with illegal image information stored in a database, and if the comparison is unsuccessful, generate information for urging the vehicle to leave the no-parking section and send the information to the owner of the vehicle, and store the first set of feature information in the database to obtain first illegal image information; The image acquisition module is further configured to acquire four images of the vehicle in the no-parking section again after a preset second interval time to obtain a second set of illegal images, wherein the second set of illegal images comprises four images sequentially captured at the first interval time; The image recognition module is further configured to perform feature extraction on the second set of illegal images to obtain second set of feature information; A law violation processing program module is configured to compare the second set of feature information with the illegal image information stored in the database, and when the second set of feature information is successfully compared with the first illegal image information, generate law violation information.

10. A computer program, characterized in that, The computer program is executed by a processor to implement the steps of the electronic illegal parking method based on big data vehicle trajectory according to any one of claims 1-8. The computer program is executed by a processor to implement the steps of the electronic illegal parking method based on big data vehicle trajectory according to any one of claims 1-8.

Images