National and provincial road truck illegal parking evidence obtaining method and system based on eagle eye machine recognition

Abstract

The invention relates to the technical field of illegal parking evidence obtaining, and particularly discloses a national and provincial road truck illegal parking evidence obtaining method and system based on eagle eye machine recognition, and the method comprises the steps: obtaining a national and provincial road pavement monitoring video; when a truck exists in the illegal parking area in the national and provincial road surface monitoring video and the static duration of the truck reaches a first preset time threshold value, acquiring a first-stage snapshot image of the truck; after the first-stage snapshot is completed, acquiring a monitoring video of the illegal parking area; when the truck in the illegal parking area monitoring video is still in a static state within a second preset time threshold value, acquiring a second-stage snapshot image of the truck; and generating an illegal parking evidence obtaining image according to the first-stage snapshot image and the second-stage snapshot image of the truck. According to the invention, illegal parking behaviors of the national and provincial lanes can be identified and obtained, off-site punishment and management and control of the illegal parking behaviors of the national and provincial lanes are realized, traffic potential safety hazards caused by illegal parking of the trucks are restrained, and the normal traffic order of the national and provincial lanes is maintained.

Description

Technical Field

[0001] This invention relates to the field of illegal evidence collection technology, and more specifically, to a method and system for collecting evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition. Background Technology

[0002] To save money or evade inspections, truck drivers tend to use free national and provincial highways whenever possible and avoid entering officially charged parking lots, opting instead for unattended areas, including the roadside. Illegal parking on the roadside disrupts traffic flow, easily leading to congestion and accidents, and is a major and persistent problem affecting traffic order on national and provincial highways, seriously undermining traffic safety. However, there is a shortage of police officers to manage traffic order on national and provincial highways, and it is impractical to install numerous non-on-site enforcement devices for detecting illegal parking on the road.

[0003] Therefore, the management of illegally parked heavy trucks on national and provincial highways urgently requires a low-cost, wide-coverage, and highly reliable method. Firstly, cost reduction is crucial. The biggest cost of illegal evidence collection equipment is the front-end construction cost, especially the cost of erecting the poles. Reusing existing road monitoring equipment is an effective method; high-point monitoring PTZ cameras (eagle-eye PTZ cameras) are common road monitoring equipment on national and provincial highways. Secondly, a large coverage area is essential. National and provincial highways are long, while the total number of illegal parking violations by heavy trucks is limited, and their distribution range is unknown. This necessitates a method with high coverage. Deploying a national and provincial highway heavy truck illegal parking evidence collection system in a central computer room allows for the reuse of existing front-end high-point monitoring PTZ cameras, significantly reducing costs. Furthermore, the central system can poll or control the high-point monitoring PTZ cameras across the entire area in batches, providing comprehensive coverage.

[0004] For the reasons mentioned above, there is an urgent need for a device and system for detecting illegally parked trucks on national and provincial highways based on eagle-eye PTZ cameras. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for obtaining evidence of illegal parking of heavy trucks on national and provincial highways based on eagle-eye camera recognition, which can improve the accuracy and reliability of obtaining evidence of illegal parking of heavy trucks on national and provincial highways.

[0006] As a first aspect of the present invention, a method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition is provided, comprising the following steps: Step S1: Monitor the national and provincial roads using a PTZ camera in a pre-set polling position to obtain video footage of the national and provincial roads; wherein the PTZ camera is positioned on the side of the national and provincial roads. Step S2: Determine whether there is a target truck in the target illegal parking area in the national and provincial road surface monitoring video; when there is a target truck in the target illegal parking area and the stationary time of the target truck reaches the first preset time threshold, the target truck is captured in the first stage by the Eagle Eye PTZ camera at the capture preset position to obtain the first stage illegal parking capture image of the target truck. Step S3: After the first stage of capture is completed, the target illegal parking area is monitored by the Eagle Eye PTZ camera in the polling preset position to obtain the monitoring video of the target illegal parking area; Step S4: When the target truck in the surveillance video of the target illegal parking area is still stationary within the second preset time threshold, it is confirmed that the target truck has committed illegal parking. The target truck is then captured in the second stage by the Eagle Eye PTZ camera at the preset capture position to obtain the second stage illegal parking capture image of the target truck. Step S5: Generate an evidence image of the target truck's illegal parking based on the first-stage illegal parking capture image and the second-stage illegal parking capture image.

[0007] Furthermore, prior to step S1, the procedure also includes: (1) Set the eagle eye camera to multiple polling preset positions for national and provincial road surface scenes, and draw the illegal parking area on the annotation page of the national and provincial road surface scenes; wherein, when the eagle eye camera rotates to different polling preset positions, the national and provincial road surface scenes will be reconstructed. (2) Configure the polling strategy of the Eagle Eye PTZ camera at multiple polling preset positions and start the polling task of the Eagle Eye PTZ camera; wherein, the Eagle Eye PTZ camera is controlled to poll at multiple polling preset positions at a certain time interval to obtain the national and provincial road surface monitoring videos captured by the Eagle Eye PTZ camera at different polling preset positions. The order of the polling preset positions is arranged according to the principle of prioritizing areas with high incidence of illegal parking, and the trigger condition for starting the polling task is a system timed or manual command. The eagle-eye camera can rotate from the polling preset position to the capture preset position to capture images. After capturing an image, the eagle-eye camera can rotate from the capture preset position to the polling preset position to continue polling. (3) Input the authentication information of the Eagle Eye PTZ camera and obtain all the preset position information of the Eagle Eye PTZ camera.

[0008] Furthermore, the determination of whether a target large truck exists within the target illegally parked area in the national and provincial highway surface surveillance video also includes: Image recognition technology is used to identify target illegally parked areas from the road surface monitoring videos of the national and provincial highways, and a multi-scale detection framework with improved YOLO algorithm is used to detect objects within the target illegally parked areas; When the target truck is detected as an object in the target illegal parking area, it is determined whether the stationary duration of the target truck has reached the first preset time threshold; when the target truck is not detected as an object in the target illegal parking area, the detection of objects in the target illegal parking area continues until the target truck is detected.

[0009] Furthermore, when a target truck exists within the target illegal parking area and the target truck remains stationary for a duration reaching a first preset time threshold, the eagle-eye camera performs a first-stage capture of the target truck at a preset capture position to obtain the first-stage illegal parking capture image of the target truck, which further includes: The eagle-eye camera performs a first-stage large-scene capture of the target truck at a preset capture position to obtain a first large-scene capture image of the target truck; wherein, the first large-scene capture image can display the overall picture of the target truck in the target illegal parking area; The eagle-eye camera performs a first-stage close-up capture of the target truck at a preset capture position to obtain a first close-up capture image of the target truck; wherein, the first close-up capture image can display the license plate number information of the target truck.

[0010] Furthermore, step S3 also includes: Determine whether the target truck in the surveillance video of the target illegal parking area is still stationary within a second preset time threshold; wherein, after the first stage of capture is completed, the displacement status of the target truck in the surveillance video of the target illegal parking area is continuously monitored; If the target truck in the surveillance video of the target illegal parking area moves within the second preset time threshold, it is confirmed that the target truck has not committed illegal parking; otherwise, step S4 is executed.

[0011] Further, the second-stage capture of the target truck at a preset capture position using the eagle-eye camera to obtain a second-stage image of the target truck illegally parked includes: The eagle-eye camera performs a second-stage large-scene capture of the target truck at a preset capture position to obtain a second large-scene capture image of the target truck; wherein, the second large-scene capture image can display the overall picture of the target truck in the target illegal parking area; The eagle-eye camera performs a second-stage close-up capture of the target truck at a preset capture position to obtain a second close-up capture image of the target truck; wherein, the second close-up capture image can display the license plate number information of the target truck.

[0012] Further, the step of generating the illegal parking evidence image of the target truck based on the first-stage illegal parking capture image and the second-stage illegal parking capture image includes: An evidence image of the target truck's illegal parking is generated based on the first large-scene captured image, the first close-up captured image, the second large-scene captured image, and the second close-up captured image. The evidence image of the target truck's illegal parking includes at least the time of violation, the location of violation, the license plate number, the violation code, the violation behavior, the equipment code, and the anti-counterfeiting code information.

[0013] As a second aspect of the present invention, a device for collecting evidence of illegal parking of heavy trucks on national and provincial highways based on eagle-eye camera recognition is provided, the device comprising: The first acquisition module is used to monitor the surface of national and provincial roads by using a PTZ camera to poll preset positions in order to obtain monitoring videos of the national and provincial roads; wherein, the PTZ camera is arranged on the side of the national and provincial roads; The first judgment module is used to determine whether there is a target truck in the target illegal parking area in the national and provincial road surface monitoring video; when there is a target truck in the target illegal parking area and the stationary time of the target truck reaches the first preset time threshold, the target truck is captured in the first stage by the eagle eye ball camera at the capture preset position to obtain the first stage illegal parking capture image of the target truck. The second acquisition module is used to monitor the target illegal parking area through the Eagle Eye PTZ camera in a polling preset position after the first stage of capture is completed, so as to obtain the monitoring video of the target illegal parking area. The second judgment module is used to determine that the target truck has committed illegal parking behavior when the target truck in the surveillance video of the target illegal parking area is still stationary within a second preset time threshold, and to perform a second-stage capture of the target truck at the capture preset position through the eagle eye ball camera to obtain the second-stage illegal parking capture image of the target truck. The generation module is used to generate an evidence image of the target truck's illegal parking based on the first-stage illegal parking capture image and the second-stage illegal parking capture image.

[0014] As a third aspect of the present invention, a system for obtaining evidence of illegal parking of heavy trucks on national and provincial highways based on eagle-eye PTZ camera recognition is provided, comprising road surface equipment and a central computer room. The road surface equipment includes an eagle-eye PTZ camera, and the central computer room includes the aforementioned device for obtaining evidence of illegal parking of heavy trucks on national and provincial highways based on eagle-eye PTZ camera recognition. The eagle-eye PTZ camera is used to monitor the road surface of national and provincial highways at a polling preset position and to capture images of target heavy trucks at a capture preset position.

[0015] The present invention provides a method for obtaining evidence of illegal parking of heavy trucks on national and provincial highways based on eagle-eye PTZ camera recognition, which has the following advantages: controlling the front-end eagle-eye PTZ camera to access the front-end video stream, identifying and obtaining evidence of illegal parking of heavy trucks on national and provincial highways, realizing non-on-site punishment and control of illegal parking of heavy trucks on national and provincial highways, curbing traffic safety hazards caused by illegal parking of heavy trucks, and maintaining normal traffic order on national and provincial highways. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the following detailed description to explain the invention, but do not constitute a limitation thereof.

[0017] Figure 1 The flowchart illustrates a method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition, as provided by this invention.

[0018] Figure 2 The flowchart illustrates a specific implementation method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition, as provided by the present invention.

[0019] Figure 3 This invention provides a structural diagram of a system for detecting illegally parked trucks on national and provincial highways based on eagle-eye PTZ camera recognition. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for obtaining evidence of illegally parked heavy trucks on national and provincial highways based on eagle-eye camera identification, according to the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of the invention described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] This embodiment provides a method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition, such as... Figure 1 As shown, the method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition includes: Step S1: Monitor the national and provincial roads using the Eagle Eye PTZ camera in the polling preset position to obtain the monitoring video of the national and provincial roads; wherein, the Eagle Eye PTZ camera, which is installed at a height of more than 6 meters on the side of the national and provincial roads, can monitor a wide range of road scenes through its built-in pan-tilt unit and zoom. It should be noted that during the data collection process, the access of road surface monitoring video for national and provincial highways can adopt either the RTSP or GB28181 protocol. The former has lower latency and better real-time performance, while the latter has stronger compatibility and higher security. The access method should be selected appropriately based on the actual situation. To ensure the stability of the video stream, the system adopts a reconnection mechanism after a video stream interruption, automatically attempting to reconnect when the video stream is interrupted; it also supports adaptive bitrate, dynamically adjusting the video bitrate according to network conditions to ensure clear and smooth video.

[0024] Preferred, such as Figure 2 As shown, before step S1, the procedure further includes: (1) Set the eagle eye camera to multiple polling preset positions on national and provincial roads, and draw the illegal parking area on the annotation page of the national and provincial roads; wherein, when the eagle eye camera rotates to different polling preset positions, the national and provincial roads will be reconstructed; wherein, the settings of different scene types are different, and the vehicle driving trajectory and field of vision need to be considered at the curve, and the traffic lights and lane distribution need to be combined at the intersection.

[0025] It should be noted that, firstly, the images from the new perspective of the Eagle Eye PTZ camera are preprocessed to remove noise and interference; then, key feature points are extracted and matched with previously stored scene information; finally, a new scene model is reconstructed based on the matching results.

[0026] (2) Configure the polling strategy of the Eagle Eye PTZ camera at multiple polling preset positions and start the polling task of the Eagle Eye PTZ camera; wherein, the Eagle Eye PTZ camera is controlled to poll at multiple polling preset positions at a certain time interval to obtain the national and provincial road surface monitoring videos captured by the Eagle Eye PTZ camera at different polling preset positions. The polling preset positions are arranged according to the principle of prioritizing areas with high incidence of illegal parking, and the priority rules are determined based on the importance and risk level of the area. The polling task is triggered by a system timer or manual command. When network failures or equipment malfunctions occur, the system has an abnormal recovery mechanism that automatically reconnects the equipment and adjusts the polling order to ensure the task continues.

[0027] The eagle-eye camera can rotate from the polling preset position to the capture preset position to capture images. After capturing an image, the eagle-eye camera can rotate from the capture preset position to the polling preset position to continue polling. It should be noted that the Eagle Eye PTZ camera is set with polling preset positions and capture preset positions. When an illegally parked vehicle is detected, the Eagle Eye PTZ camera quickly rotates to the corresponding capture preset position to capture the image. After the Eagle Eye PTZ camera completes the capture of the illegally parked vehicle, the system will automatically send a return command, causing the Eagle Eye PTZ camera to quickly return to the polling preset position so that regular monitoring can continue.

[0028] It should be noted that the polling order and time interval of the preset polling position are automatically adjusted according to road traffic and illegal parking conditions to improve monitoring efficiency.

[0029] Specifically, the polling strategy design is key to ensuring the efficient operation of the system. The time interval can be flexibly set according to different scenarios; it can be shortened to 5-10 minutes for busy road sections and extended to 15-30 minutes for remote road sections.

[0030] (3) Input the authentication information of the PTZ camera and obtain all the preset position information of the PTZ camera. The specific process is as follows: (1) Preparation: Collect the IP address, username, password and other information of the PTZ camera to ensure that the network connection is normal. (2) SDK login: Use the login interface provided by the SDK to input the authentication information of the PTZ camera and adopt a secure authentication method, such as username + password + verification code. (3) API call: Call the preset point query API and set the interface parameters, such as query range, return format, etc. (4) Device compatibility check: Perform adaptation tests on different brands of PTZ cameras, check the compatibility of SDK and API, and if incompatibility occurs, update the SDK version or adjust the interface parameters in time. (5) Result verification: Verify the accuracy of the query results and ensure that all preset position information is obtained.

[0031] Step S2: Determine whether there is a target truck in the target illegal parking area in the national and provincial road surface monitoring video; when there is a target truck in the target illegal parking area and the stationary time of the target truck reaches the first preset time threshold, the target truck is captured in the first stage by the Eagle Eye PTZ camera at the capture preset position to obtain the first stage illegal parking capture image of the target truck. Preferred, such as Figure 2 As shown, the step of determining whether a target large truck exists within the target illegally parked area in the national and provincial highway surface surveillance video also includes: Image recognition technology is used to identify target illegally parked areas from the road surface monitoring videos of the national and provincial highways, and a multi-scale detection framework with improved YOLO algorithm is used to detect objects within the target illegally parked areas; It should be noted that advanced image processing and machine learning algorithms were employed for the real-time identification of road elements such as lane lines and guide lines in the aforementioned national and provincial highway surface monitoring videos. By performing edge detection, feature extraction, and classification on the images, different types of road elements can be accurately identified. For example, Hough transform is used to detect straight lane lines, and color and texture features are used to identify guide lines; then, by modeling the spatial relationship between lane lines and illegal parking areas, it is possible to accurately determine whether a vehicle has deviated from its lane and parked in an illegal parking area.

[0032] Specifically, firstly, image recognition technology is used to identify lane lines and illegally parked areas in the surveillance video of the national and provincial roads; then, the relative positional relationship between the vehicle and the lane lines and illegally parked areas is calculated. When a vehicle deviates from its lane and parks within an illegally parked area, it is determined that the vehicle is suspected of illegally parking. This spatial relationship modeling method improves the accuracy and reliability of illegally parked vehicle determination.

[0033] It should be noted that vehicle target detection and classification models play a crucial role in detecting illegally parked heavy trucks. This invention employs a multi-scale detection framework based on an improved YOLO algorithm to enhance detection accuracy and efficiency. The construction of a heavy truck feature library is a vital foundation for the classification model. By collecting a large number of images of heavy trucks of different brands, models, and colors, and through annotation and feature extraction, a rich heavy truck feature library was constructed. These features include vehicle contours, textures, colors, and other information, used for subsequent vehicle type and model classification.

[0034] Specifically, the multi-scale detection framework of the improved YOLO algorithm can efficiently extract features of large trucks, such as vehicle body outline and front shape. YOLOv8 is based on a deep learning architecture and learns the feature patterns of large trucks through a large amount of training data, enabling it to quickly and accurately identify large truck targets in complex national and provincial highway scenarios.

[0035] When the target truck is detected as an object in the target illegal parking area, it is determined whether the stationary duration of the target truck has reached the first preset time threshold (e.g., 30 seconds); when the target truck is not detected as an object in the target illegal parking area, the detection of objects in the target illegal parking area continues until the target truck is detected.

[0036] It should be noted that the association rule between the first preset time threshold and the road area is as follows: different first preset time thresholds are set for different road areas. For example, in road sections where parking is prohibited, the first preset time threshold may be set to a shorter time, such as 3 minutes; while in areas where temporary parking is allowed, the first preset time threshold may be appropriately extended. When a vehicle remains stationary in a certain road area for more than the first preset time threshold, it will be determined that the vehicle is suspected of illegal parking.

[0037] Preferred, such as Figure 2 As shown, when a target truck exists within the target illegal parking area and the truck remains stationary for a duration reaching a first preset time threshold, the eagle-eye camera performs a first-stage capture of the target truck at a preset capture position to obtain the first-stage illegal parking capture image of the target truck. This process also includes: The eagle-eye camera performs a first-stage large-scene capture of the target truck at a preset capture position to obtain a first large-scene capture image of the target truck; wherein, the first large-scene capture image can display the overall picture of the target truck in the target illegal parking area, record the relative position of the target truck, and provide scene information for subsequent analysis; The eagle-eye camera performs a first-stage close-up capture of the target truck at a preset capture position to obtain a first close-up capture image of the target truck; wherein, the first close-up capture image can display the license plate number information of the target truck.

[0038] It should be noted that in the first phase of close-up capture, the Eagle Eye PTZ camera quickly zooms in to capture close-up shots of the truck with high-precision focus adjustment, while using advanced license plate recognition technology to identify license plate information to ensure accurate vehicle identification.

[0039] Step S3: After the first stage of capture is completed, the target illegal parking area is monitored by the Eagle Eye PTZ camera in the polling preset position to obtain the monitoring video of the target illegal parking area; Preferred, such as Figure 2 As shown, step S3 further includes: Determine whether the target truck in the surveillance video of the target illegal parking area is still stationary within a second preset time threshold (10 minutes); wherein, after the first stage of capture is completed, the displacement status of the target truck in the surveillance video of the target illegal parking area is continuously monitored; If the target truck in the surveillance video of the target illegal parking area moves within the second preset time threshold, it is confirmed that the target truck has not committed illegal parking; otherwise, step S4 is executed.

[0040] It should be noted that if 10 minutes is set as the second preset time threshold, the system records the start timestamp when it detects the vehicle is stationary, and then continuously updates the timestamp to keep track of the time. The timing termination rule stipulates that the timing will stop when the vehicle moves (regardless of whether it leaves the illegally parked area).

[0041] Specifically, the displacement detection method employs pixel difference and feature point matching. Pixel difference determines whether the vehicle has moved by comparing the differences in pixel values ​​between adjacent frames; feature point matching extracts the vehicle's feature points, such as license plate corner points and vehicle body contour points, and detects displacement by comparing the positional changes of these feature points at different times.

[0042] It's important to note that Kalman filtering is used to predict the displacement of the target truck. In real-world scenarios, the target truck may be affected by factors such as occlusion and changes in lighting, leading to its loss of position. Kalman filtering predicts the target truck's position by establishing a motion model and combining it with historical trajectory information, thus enabling continuous tracking. For example, when the truck is briefly obscured by other vehicles, Kalman filtering can predict its position during the obstruction period based on its previous motion state, and quickly re-lock onto the target truck after the obstruction disappears.

[0043] It should be noted that, to ensure accurate determination of illegal parking, this invention employs a continuous monitoring mechanism for illegal parking status. After determining that a vehicle is suspected of illegal parking, the system continuously monitors the vehicle's status. If the vehicle remains stationary for a period of time, the illegal parking is confirmed; if the vehicle moves during the monitoring process, the illegal parking determination is cancelled.

[0044] Step S4: When the target truck in the surveillance video of the target illegal parking area is still stationary within the second preset time threshold, it is confirmed that the target truck has committed illegal parking. The target truck is then captured in the second stage by the Eagle Eye PTZ camera at the preset capture position to obtain the second stage illegal parking capture image of the target truck. Preferred, such as Figure 2As shown, the second-stage capture of the target truck by the eagle-eye camera at the preset capture position to obtain the second-stage illegal parking capture image of the target truck includes: The eagle-eye camera performs a second-stage large-scene capture of the target truck at a preset capture position to obtain a second large-scene capture image of the target truck; wherein, the second large-scene capture image can display the overall picture of the target truck in the target illegal parking area; the second large-scene capture image is compared with the first large-scene capture image to prove that the vehicle is continuously illegally parked.

[0045] The eagle-eye camera performs a second-stage close-up capture of the target truck at a preset capture position to obtain a second close-up capture image of the target truck; wherein, the second close-up capture image can display the license plate number information of the target truck.

[0046] It should be noted that in the second phase of close-up capture, the cloud-controlled PTZ camera zoomed in again to capture close-up shots, further confirming the license plate number information and ensuring the accuracy and completeness of the evidence.

[0047] Step S5: Generate an evidence image of the target truck's illegal parking based on the first-stage illegal parking capture image and the second-stage illegal parking capture image.

[0048] Preferred, such as Figure 2 As shown, the step of generating the illegal parking evidence image of the target truck based on the first-stage illegal parking capture image and the second-stage illegal parking capture image includes: An illegal parking evidence image of the target truck is generated based on the first large scene capture image, the first close-up capture image, the second large scene capture image, and the second close-up capture image. The character overlay information of the illegal parking evidence image of the target truck includes at least the time of violation, the location of violation, the license plate number, the violation code, the violation behavior, the equipment code, and the anti-counterfeiting code.

[0049] In this embodiment of the invention, the core elements of the evidence chain are crucial to ensuring the legality and validity of evidence collection for illegal parking. The synthesis of the illegal parking evidence images (four-in-one illegal parking evidence images) of the target truck must follow certain rules. The layout ratio of the large scene image to the close-up image is 2:2 to ensure that both the overall scene of the illegally parked vehicle and the vehicle details are clearly presented.

[0050] In this embodiment of the invention, timestamp synchronization technology is used to ensure the accuracy of time; geographic information overlay is used to clearly mark the location of illegal parking in the evidence; hash verification is used to prevent tampering and ensure the integrity and authenticity of the evidence. During the synthesis process, the requirements of GA / T832 must be strictly followed to ensure the clarity, integrity, and compliance of the images. The resolution, color mode, etc., of the images must meet the standards, and the accuracy and completeness of the metadata are also crucial to ensure the legal validity of the illegal evidence images as valid evidence.

[0051] This invention provides a method for capturing evidence of illegally parked heavy trucks on national and provincial highways based on Eagle Eye PTZ camera recognition. It comprehensively utilizes technologies such as automatic repositioning, scene reconstruction, a heavy truck feature database, and a judgment rule engine to control the front-end Eagle Eye PTZ camera to access the front-end video stream and capture evidence of illegally parked heavy trucks on national and provincial highways. Specifically, the method includes: 1) setting the initial position of the Eagle Eye PTZ camera and automatically triggering a repositioning action after completing a specific task, so as to capture illegally parked heavy trucks without affecting daily monitoring; 2) employing real-time road element recognition and scene reconstruction technologies to achieve multi-preset position polling for evidence of illegally parked heavy trucks; 3) constructing a heavy truck feature database and quickly identifying heavy trucks through vehicle target detection and classification models, reducing the impact of illegal parking detection on daily monitoring; and 4) establishing a heavy truck illegal parking behavior judgment rule engine, improving the accuracy and reliability of illegal parking judgment through techniques such as static time threshold and road area association rules, continuous monitoring mechanism of illegal status, and spatial relationship modeling of lane departure and illegal parking area.

[0052] As another embodiment of the present invention, a device for collecting evidence of illegal parking of heavy trucks on national and provincial highways based on eagle-eye camera recognition is provided. The device includes: The first acquisition module is used to monitor the surface of national and provincial roads by using a PTZ camera to poll preset positions in order to obtain monitoring videos of the national and provincial roads; wherein, the PTZ camera is arranged on the side of the national and provincial roads; The first judgment module is used to determine whether there is a target truck in the target illegal parking area in the national and provincial road surface monitoring video; when there is a target truck in the target illegal parking area and the stationary time of the target truck reaches the first preset time threshold, the target truck is captured in the first stage by the eagle eye ball camera at the capture preset position to obtain the first stage illegal parking capture image of the target truck. The second acquisition module is used to monitor the target illegal parking area through the Eagle Eye PTZ camera in a polling preset position after the first stage of capture is completed, so as to obtain the monitoring video of the target illegal parking area. The second judgment module is used to determine that the target truck has committed illegal parking behavior when the target truck in the surveillance video of the target illegal parking area is still stationary within a second preset time threshold, and to perform a second-stage capture of the target truck at the capture preset position through the eagle eye ball camera to obtain the second-stage illegal parking capture image of the target truck. The generation module is used to generate an evidence image of the target truck's illegal parking based on the first-stage illegal parking capture image and the second-stage illegal parking capture image.

[0053] As another embodiment of the present invention, a system for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition is provided, such as... Figure 3 As shown, the system for detecting illegally parked trucks on national and provincial highways based on eagle-eye camera identification includes road surface equipment and a central computer room. The road surface equipment includes eagle-eye cameras, and the central computer room includes the aforementioned device for detecting illegally parked trucks on national and provincial highways based on eagle-eye camera identification. The eagle-eye cameras are used to monitor the road surface of national and provincial highways at polling preset positions and to capture images of target trucks at capture preset positions.

[0054] In this embodiment of the invention, the video image data collected by the road equipment is processed in two ways: first, it is automatically identified and the identification result is uploaded to the central computer room; second, it is handed over to the national and provincial highway truck illegal parking evidence collection device deployed in the central computer room for identification, and the identification result is uploaded to the command and dispatch center.

[0055] In this embodiment of the invention, based on the existing prevention and control facilities and equipment layout of national and provincial highways, by supplementing and strengthening the hardware and software, the eagle eye cameras on the side of national and provincial highways and the central computer room are combined into a national and provincial highway truck illegal parking evidence collection system, so as to comprehensively realize the evidence collection of national and provincial highway truck illegal parking.

[0056] In this embodiment of the invention, the Eagle Eye PTZ camera is a high-altitude PTZ camera used for monitoring national and provincial highways; the device for collecting evidence of illegal parking of large trucks on national and provincial highways based on the Eagle Eye PTZ camera is deployed in a central computer room. The central computer room uses a 2U rack-mounted server as the core hardware and connects to the Eagle Eye PTZ camera at the front end through a gigabit network to realize the collection of evidence of illegal parking of large trucks on national and provincial highways.

[0057] The present invention provides a system for collecting evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye PTZ cameras. The system controls the front-end eagle-eye PTZ cameras to access the front-end video stream, identifies and collects evidence of illegal parking of large trucks on national and provincial highways, realizes non-on-site punishment and control of illegal parking of large trucks on national and provincial highways, curbs the traffic safety hazards caused by illegal parking of large trucks, and maintains the normal traffic order of national and provincial highways.

[0058] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition, characterized in that, The method for obtaining evidence of illegally parked heavy trucks on national and provincial highways based on Eagle Eye PTZ camera recognition includes the following steps: Step S1: Monitor the national and provincial roads using a PTZ camera in a pre-set polling position to obtain video footage of the national and provincial roads; wherein the PTZ camera is positioned on the side of the national and provincial roads. Step S2: Determine whether there is a target truck in the target illegal parking area in the national and provincial road surface monitoring video; when there is a target truck in the target illegal parking area and the stationary time of the target truck reaches the first preset time threshold, the target truck is captured in the first stage by the Eagle Eye PTZ camera at the capture preset position to obtain the first stage illegal parking capture image of the target truck. Step S3: After the first stage of capture is completed, the target illegal parking area is monitored by the Eagle Eye PTZ camera in the polling preset position to obtain the monitoring video of the target illegal parking area; Step S4: When the target truck in the surveillance video of the target illegal parking area is still stationary within the second preset time threshold, it is confirmed that the target truck has committed illegal parking. The target truck is then captured in the second stage by the Eagle Eye PTZ camera at the preset capture position to obtain the second stage illegal parking capture image of the target truck. Step S5: Generate an evidence image of the target truck's illegal parking based on the first-stage illegal parking capture image and the second-stage illegal parking capture image.

2. The method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition as described in claim 1, characterized in that, Before step S1, the procedure also includes: (1) Set the eagle eye camera to multiple polling preset positions for national and provincial road surface scenes, and draw the illegal parking area on the annotation page of the national and provincial road surface scenes; wherein, when the eagle eye camera rotates to different polling preset positions, the national and provincial road surface scenes will be reconstructed. (2) Configure the polling strategy of the Eagle Eye PTZ camera at multiple polling preset positions and start the polling task of the Eagle Eye PTZ camera; wherein, the Eagle Eye PTZ camera is controlled to poll at multiple polling preset positions at a certain time interval to obtain the national and provincial road surface monitoring videos captured by the Eagle Eye PTZ camera at different polling preset positions. The order of the polling preset positions is arranged according to the principle of prioritizing areas with high incidence of illegal parking, and the trigger condition for starting the polling task is a system timed or manual command. The eagle-eye camera can rotate from the polling preset position to the capture preset position to capture images. After capturing an image, the eagle-eye camera can rotate from the capture preset position to the polling preset position to continue polling. (3) Input the authentication information of the Eagle Eye PTZ camera and obtain all the preset position information of the Eagle Eye PTZ camera.

3. The method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition as described in claim 1, characterized in that, The determination of whether a target large truck exists within the target illegally parked area in the surveillance video of the national and provincial highways also includes: Image recognition technology is used to identify target illegally parked areas from the road surface monitoring videos of the national and provincial highways, and a multi-scale detection framework with improved YOLO algorithm is used to detect objects within the target illegally parked areas; When the target truck is detected as an object in the target illegal parking area, it is determined whether the stationary duration of the target truck has reached the first preset time threshold; when the target truck is not detected as an object in the target illegal parking area, the detection of objects in the target illegal parking area continues until the target truck is detected.

4. The method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition as described in claim 1, characterized in that, When a target truck is present within the target illegal parking area and the truck remains stationary for a duration reaching a first preset time threshold, the eagle-eye camera performs a first-stage capture of the target truck at a preset capture position to obtain the first-stage illegal parking capture image of the target truck. This process also includes: The eagle-eye camera performs a first-stage large-scene capture of the target truck at a preset capture position to obtain a first large-scene capture image of the target truck; wherein, the first large-scene capture image can display the overall picture of the target truck in the target illegal parking area; The eagle-eye camera performs a first-stage close-up capture of the target truck at a preset capture position to obtain a first close-up capture image of the target truck; wherein, the first close-up capture image can display the license plate number information of the target truck.

5. The method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition as described in claim 1, characterized in that, Step S3 further includes: Determine whether the target truck in the surveillance video of the target illegal parking area is still stationary within a second preset time threshold; wherein, after the first stage of capture is completed, the displacement status of the target truck in the surveillance video of the target illegal parking area is continuously monitored; If the target truck in the surveillance video of the target illegal parking area moves within the second preset time threshold, it is confirmed that the target truck has not committed illegal parking; otherwise, step S4 is executed.

6. The method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition as described in claim 4, characterized in that, The second stage of capturing the target truck at a preset capture position using the eagle-eye camera to obtain a second-stage image of the target truck illegally parked includes: The eagle-eye camera performs a second-stage large-scene capture of the target truck at a preset capture position to obtain a second large-scene capture image of the target truck; wherein, the second large-scene capture image can display the overall picture of the target truck in the target illegal parking area; The eagle-eye camera performs a second-stage close-up capture of the target truck at a preset capture position to obtain a second close-up capture image of the target truck; wherein, the second close-up capture image can display the license plate number information of the target truck.

7. The method for obtaining evidence of illegal parking of large trucks on national and provincial highways based on eagle-eye camera recognition as described in claim 6, characterized in that, The process of generating the illegal parking evidence image of the target truck based on the first-stage illegal parking capture image and the second-stage illegal parking capture image includes: An evidence image of the target truck's illegal parking is generated based on the first large-scene captured image, the first close-up captured image, the second large-scene captured image, and the second close-up captured image. The evidence image of the target truck's illegal parking includes at least the time of violation, the location of violation, the license plate number, the violation code, the violation behavior, the equipment code, and the anti-counterfeiting code information.

8. A device for collecting evidence of illegally parked heavy trucks on national and provincial highways based on eagle-eye camera recognition, used to implement the method for collecting evidence of illegally parked heavy trucks on national and provincial highways based on eagle-eye camera recognition as described in any one of claims 1 to 7, characterized in that, The device for detecting illegally parked trucks on national and provincial highways based on eagle-eye camera recognition includes: The first acquisition module is used to monitor the surface of national and provincial roads by using a PTZ camera to poll preset positions in order to obtain monitoring videos of the national and provincial roads; wherein, the PTZ camera is arranged on the side of the national and provincial roads; The first judgment module is used to determine whether there is a target truck in the target illegal parking area in the national and provincial road surface monitoring video; when there is a target truck in the target illegal parking area and the stationary time of the target truck reaches the first preset time threshold, the target truck is captured in the first stage by the eagle eye ball camera at the capture preset position to obtain the first stage illegal parking capture image of the target truck. The second acquisition module is used to monitor the target illegal parking area through the Eagle Eye PTZ camera in a polling preset position after the first stage of capture is completed, so as to obtain the monitoring video of the target illegal parking area. The second judgment module is used to determine that the target truck has committed illegal parking behavior when the target truck in the surveillance video of the target illegal parking area is still stationary within a second preset time threshold, and to perform a second-stage capture of the target truck at the capture preset position through the eagle eye ball camera to obtain the second-stage illegal parking capture image of the target truck. The generation module is used to generate an evidence image of the target truck's illegal parking based on the first-stage illegal parking capture image and the second-stage illegal parking capture image.

9. A system for detecting illegally parked trucks on national and provincial highways based on eagle-eye camera recognition, characterized in that, The system includes road surface equipment and a central control room. The road surface equipment includes an eagle-eye camera, and the central control room includes the device for collecting evidence of illegal parking of large trucks on national and provincial highways based on the eagle-eye camera identification as described in claim 8. The eagle-eye camera is used to monitor the road surface of national and provincial highways at a polling preset position and to capture images of target large trucks at a capture preset position.

Images