Parking lot vehicle information acquisition system

Abstract

The invention provides a parking lot vehicle information acquisition system, which comprises a high-definition camera arranged at the entrance and exit of a parking lot and used for shooting vehicle images; the vehicle information acquisition device is connected with the high-definition camera; the server data center library is used for receiving and storing the vehicle information and is linked with a police alarm system and a traffic guidance system; the communication module serves as a standby interface to take over data transmission when network communication is interrupted, and it is ensured that the data are continuously uploaded to a server data center library. According to the method, the multi-spectral imaging technology and the self-adaptive AI algorithm are combined, the license plate recognition rate can still be kept in severe weather, vehicle three-dimensional feature analysis is fused by adopting the deep learning model, vehicle features can be accurate at the same time, the information collection dimension can be improved, and recognition can still be completed through local feature matching when 60% of the license plate is shielded by accumulated snow and dirt.

Description

Technical Field

[0001] The present invention relates to the field of vehicle information collection, and particularly to a vehicle information collection system for a parking lot. Background Art

[0002] With the rapid growth of the urban motor vehicle ownership, the traditional parking lot management system faces many technical bottlenecks: First, the existing systems mostly adopt a single license plate recognition technology, and the recognition rate drops significantly under complex lighting and bad weather conditions, and it is impossible to extract multi-dimensional features of vehicles. Secondly, the information of each parking lot is isolated, lacking a regional collaborative scheduling mechanism, resulting in more than 30% of car owners having to detour to find parking spaces, exacerbating the congestion of the surrounding roads. In terms of safety management, the linkage delay between the traditional system and the database is relatively high, making it difficult to meet the real-time supervision requirements. More prominently, the existing technology lacks a high-precision parking space guiding interface for autonomous driving vehicles, and the electric vehicle charging parking spaces are often occupied by fuel vehicles.

[0003] Therefore, in view of the above, a vehicle information collection system for a parking lot is now developed. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, the present invention provides a vehicle information collection system for a parking lot.

[0005] The technical solution of the present invention is as follows:

[0006] A vehicle information collection system for a parking lot, comprising:

[0007] A high-definition camera, deployed at the entrance and exit of the parking lot, for taking vehicle images;

[0008] A vehicle information collection device, connected to the high-definition camera, including: a camera access and parsing module, for receiving and parsing the license plate information and images transmitted by cameras of different models; a 4G communication module, for uploading the parsed data to the server data center library; a data storage module, for temporarily storing vehicle information for at least 7 days when the network is interrupted;

[0009] A server data center library, for receiving and storing the vehicle information, and being linked with a security alarm system and a traffic guidance system;

[0010] A communication module, as a backup interface, takes over data transmission when the network communication is interrupted, ensuring continuous upload of data to the server data center library;

[0011] A self-recovery module, detects the operating state of the device through a process nesting monitoring mechanism, automatically restarts and restores to the data state before the failure when the system crashes, and the recovery time is less than 10 seconds.

[0012] Preferably, the camera access parsing module supports the compatibility parsing of multiple video signal formats and can perform AI-enhanced recognition on license plate information under low-light and occlusion conditions.

[0013] Preferably, it includes a method for collecting vehicle information in a parking lot, and the steps are as follows:

[0014] Step 1: Deploy high-definition cameras at the entrance and exit of the parking lot to take images of vehicles.

[0015] Step 2: Parse the vehicle images through the AI image recognition module to extract license plate numbers, vehicle brands, models, and color information.

[0016] Step 3: Upload the parsed vehicle information to the server data center library in real time through the 5G communication module.

[0017] Step 4: The server data center library stores and analyzes the data, and is linked with the security alarm system to match the vehicle information in real time.

[0018] Preferably, it includes an intelligent parking guidance unit for autonomous driving vehicles, which specifically includes:

[0019] A communication module for real-time data interaction with the in-vehicle system of the autonomous driving vehicle.

[0020] A path planning module for generating the optimal parking path based on the real-time parking space information in the parking lot.

[0021] A settlement module for automatically calculating the fees when the vehicle leaves and deducting the fees through the electronic payment interface.

[0022] Preferably, it further includes a shared parking service unit, and the specific method is as follows:

[0023] Step S1: The server data center library obtains the available parking space information of multiple parking lots in real time.

[0024] Step S2: The user terminal queries and reserves a parking space in a specified parking lot through a mobile application.

[0025] Step S3: After the reservation is successful, the vehicle information collection device automatically verifies the license plate information of the reserved vehicle and guides it to enter the parking lot.

[0026] Preferably, the data storage module adopts a dual-backup mechanism of solid-state drive (SSD) and cloud storage, supports encrypted data storage, and the local cache capacity can be expanded to more than 30 days.

[0027] Preferably, when the server data center library is linked with the security alarm system, it can compare vehicle information with the suspected vehicle database in real time. After finding a matching target, it will trigger an audible and visual alarm and push the location information to the security management terminal. By analyzing the parking lot saturation data, it will generate traffic induction signals and synchronize them to the electronic display screens on the surrounding roads to guide vehicles to be diverted to vacant parking lots.

[0028] Preferably, the system is linked with the charging pile management system, the server data center library obtains the charging pile status information in real time, and preferentially allocates parking spaces near idle charging piles for electric vehicles, and pushes charging guidance instructions through the user terminal.

[0029] By adopting the above technical solution, the present invention has the following advantages:

[0030] 1. The present invention combines multispectral imaging technology with an adaptive AI algorithm to maintain the license plate recognition rate in severe weather conditions. It uses a deep learning model to integrate vehicle three-dimensional feature analysis, which can accurately identify vehicle features and improve the dimension of information collection. When 60% of the license plate is blocked by snow or dirt, it can still be recognized through local feature matching.

[0031] 2. When a suspicious vehicle is detected, the system can simultaneously complete the linkage operations of gate locking, monitoring and tracking, and security dispatch. It is deeply integrated with the traffic guidance system, can predict future traffic trends based on real-time parking space data, and dynamically adjust the guidance plans of surrounding intersections to improve regional traffic efficiency. The intelligent allocation algorithm of charging piles reduces the average waiting time for electric vehicles to charge. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a structural schematic diagram of the present invention.

[0033] Figure 2 The present invention is a flowchart of the steps of the method for collecting parking lot vehicle information.

[0034] Figure 3 This is a flowchart of specific implementation steps of the shared parking service unit of the present invention.

[0035] Figure 4 It is a structural schematic diagram of the intelligent parking guidance unit of the present invention. DETAILED DESCRIPTION

[0036] Reference to an embodiment herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0037] A vehicle information acquisition system for a parking lot, as Figure 1 shown, includes:

[0038] High-definition cameras, deployed at the entrances and exits of the parking lot and key areas, support 4K resolution, HDR imaging, wide dynamic range (120dB), are equipped with infrared supplementary lighting and anti-shake functions, adapt to the environment from -30°C to 60°C, transmit video streams (RTSP protocol) in real time through PoE power supply and gigabit network interfaces, attach a time stamp and a location tag to each frame of the image. Specifically, the inductive loop or radar detects that the vehicle enters the shooting range, continuously shoots 30 frames per second, automatically focuses on the license plate area, crops the image to the license plate area to reduce the transmission bandwidth. The camera access and parsing module supports the compatibility parsing of multiple video signal formats and can perform AI-enhanced recognition on license plate information under low-light and occlusion conditions;

[0039] A vehicle information acquisition device, connected to the high-definition camera, includes:

[0040] A camera access and parsing module, used to receive and parse the license plate information and images transmitted by cameras of different models;

[0041] A 4G communication module, used to upload the parsed data to the server data center library;

[0042] A data storage module, used to temporarily store vehicle information for at least 7 days when the network is interrupted. The data storage module adopts a dual-backup mechanism of solid-state drive (SSD) and cloud storage, supports encrypted data storage, and the local cache capacity can be expanded to more than 30 days;

[0043] The server data center library adopts a sub-library strategy of a distributed microservice architecture, which is divided into a real-time library, a historical library, and a file library. It receives and stores the vehicle information and is linked with the security alarm system and the traffic guidance system. Specifically, for the security alarm interface, it docks with the vehicle library information through the national security boundary platform and compares the license plate hash values in real time. For traffic guidance, based on the parking saturation (parking space occupancy rate > 90%), it generates guidance instructions and issues them to the roadside LED screen through the Modbus TCP protocol;

[0044] A communication module, as a backup interface, takes over data transmission when the network communication is interrupted to ensure continuous data upload to the server data center library;

[0045] A self-recovery module, detects the device operation status through a process nested monitoring mechanism, automatically restarts when the system crashes and restores to the data state before the fault, and the recovery time is less than 10 seconds.

[0046] It should be noted that the system is linked with the charging pile management system. The server data center library can obtain the charging pile status information in real time, and preferentially allocate parking spaces of nearby idle charging piles for electric vehicles. At the same time, charging guidance instructions are pushed through the user terminal.

[0047] As Figure 2 shown, the method for collecting vehicle information in the parking lot is as follows:

[0048] Step 1: Installed directly above the entrance and exit lanes (at a height of 2.5 - 3 meters) with a downward angle of 15° to completely cover the license plate area. When a vehicle enters the recognition area, the camera is triggered by a ground loop or radar and continuously takes 3 high-definition photos at a rate of 25 frames per second to ensure that at least one image completely contains the license plate area. At the same time, information such as the shooting time, location, and device number is recorded.

[0049] Step 2: The vehicle image is transmitted in real time to the edge computing device or cloud server and processed by the built-in AI image recognition module. The image is denoised and enhanced, and then the license plate position is detected and the characters are segmented through deep learning algorithms to accurately identify the license plate number, provincial abbreviation, and alphanumeric combination. At the same time, the vehicle appearance features are analyzed, and the pre-trained vehicle type database is matched to output the vehicle brand (such as "Volkswagen"), specific model (such as "Magotan 330TSI"), and body color (such as "dark blue"), and all recognition results are saved in a structured format.

[0050] Step 3: The parsed vehicle information (including license plate number, vehicle type, color, and timestamp) is packaged into an encrypted data packet through the built-in 5G communication module and uploaded to the remote server data center library in real time via the operator network. If the network signal is unstable, the system will temporarily cache the data in the local storage module and automatically re-upload it after the network is restored to ensure that the data is not lost. During the upload process, the system will generate a unique check code for each piece of data to prevent transmission errors or tampering.

[0051] Step 4: After receiving the vehicle information, the server data center library classifies and stores it in the real-time database and historical archive. Subsequently, the data analysis process is started: the license plate number is compared one by one with the suspected vehicle database provided by the security alarm system. If a completely matching record is found, the alarm protocol will be immediately triggered, and the vehicle location, captured image, and real-time video stream will be sent to the jurisdiction security management terminal, and the parking lot barrier will be controlled to close synchronously to block the suspicious vehicle. At the same time, the system will record the whole process of this alarm event for subsequent verification and statistics.

[0052] As Figure 3 shown, the intelligent parking guidance unit for autonomous vehicles specifically includes:

[0053] Communication module. The communication module adopts V2X (Vehicle-to-Everything) technology, supports dual-mode transmission of 5G and DSRC (Dedicated Short Range Communication), and ensures low latency (<50ms) and high reliability. When an autonomous vehicle approaches a parking lot, this module will actively establish a connection with the vehicle's on-board system (such as the autonomous driving control unit) and exchange data in real time, including: vehicle information, namely license plate number, vehicle model size (length / width / height), current battery level (for electric vehicles); parking lot information, namely the distribution of available parking spaces, the status of charging piles, and the recommended optimal entrance; dynamic instructions, namely route adjustment, temporary obstacle reminder (such as construction area).

[0054] Route planning module. The route planning module is based on the real-time updated electronic map of the parking lot (accuracy ±10cm), combines the vehicle's current position and the target parking space, and dynamically calculates the optimal driving route. The specific process is as follows: First, perform parking space matching. The system preferentially allocates the parking space closest to the elevator / charging pile and avoids narrow channels (applicable to large vehicles). Then, consider factors such as turning radius and slope to generate the most time-saving route. Next, use the camera to monitor pedestrians and obstacles in real time. If a sudden situation (such as a temporary stop) is detected, immediately re-plan the route and push it to the vehicle. Finally, the route information is sent to the in-vehicle display screen in the form of a high-precision map, and at the same time, key nodes are prompted by voice (such as "Turn right ahead, target parking space B12").

[0055] Settlement module. The settlement module integrates contactless payment technology and supports mainstream electronic payment platforms (Alipay, WeChat, ETC). The specific process is as follows: The geomagnetic sensor or camera detects that the vehicle has left the parking space, and the system automatically associates the entry time. If the user has bound the license plate for password-free payment, the payment is directly deducted from the pre-authorized account. When not bound, a bill is pushed through text message or APP. After the user scans the code to pay, the barrier gate is released. After the payment is successful, an electronic invoice is automatically generated and sent to the user's reserved email or APP.

[0056] As Figure 4 shown, the specific implementation method of the shared parking service unit is as follows:

[0057] Step S1: The server data center library establishes a dedicated line connection with each parking lot management system through the Internet of Things gateway, polls the parking space status data every 5 seconds, the system integrates the status information from geomagnetic sensors, video recognition, and parking space locks, calculates the real-time number and distribution location of available parking spaces in each parking lot (including ordinary parking spaces, charging parking spaces, and disabled person's special parking spaces), and at the same time, the system will predict the supply and demand situation of parking spaces in the next 30 minutes (based on historical data and current traffic flow trends), and classify and store this information according to regions, prices, and parking space types to form a dynamically updated parking space resource pool;

[0058] Step S2: After the user opens the mobile APP, the system will automatically recommend available parking lots within 3 kilometers around based on GPS positioning. The APP interface will clearly display: the real-time remaining parking spaces in each parking lot (such as "15 / 200 remaining"); the charging standard (time-based price, capped fee); special services (charging piles, car wash services, elevator access, etc.); user reviews and recommendation indexes. The user can set filtering conditions (such as "only display charging spaces", "price from low to high"). After selecting the target parking lot, the user can view the parking space floor plan and manually select a specific parking space (such as "No. 12 in Area B"). When confirming the reservation, the user needs to select the estimated arrival time (supporting reservations within 15 minutes - 2 hours) and pay a reservation deposit (usually 50% of the first hour's fee, which will be deducted from the parking fee when leaving).

[0059] Step S3: When the reserved vehicle arrives at the parking lot entrance, the high-definition camera captures the license plate information and compares it with the reservation database in real time. The system verifies three matching elements: license plate number, reservation time period, and paid deposit status. After successful verification, the barrier gate automatically lifts, and at the same time, the entrance display screen shows a welcome message and navigation information (such as "Welcome, please go to Parking Space No. 05 in Area A"). The APP pushes the parking space navigation map (supporting AR real-scene guidance). Specifically, for indoor parking lots, the indicator light of the parking space is automatically activated (LED floor lights guide the route). In special cases where license plate recognition fails (such as being damaged or soiled), the user can scan the dynamic QR code generated by the APP (refreshing every 30 seconds). If the vehicle arrives early / late by more than 15 minutes, the system will automatically reassign the nearest available parking space. When a non-reserved vehicle attempts to enter, a voice prompt will say "This parking lot is only open to reserved users".

[0060] The above embodiments are provided for those skilled in the art to implement or use the present invention. Those skilled in the art can make various modifications or changes to the above embodiments without departing from the idea of the present invention. Therefore, the protection scope of the present invention is not limited by the above embodiments, but should be the maximum scope that conforms to the innovative features mentioned in the claims.

Claims

1. A vehicle information collection system for a parking lot, characterized in that, It includes: A high-definition camera, deployed at the parking lot entrance and exit, for capturing vehicle images; A vehicle information collection device, connected to the high-definition camera, including: a camera access parsing module, for receiving and parsing license plate information and images transmitted by cameras of different models; a 4G communication module, for uploading the parsed data to the server data center library; a data storage module, for temporarily storing vehicle information for at least 7 days when the network is interrupted; The server data center library, which receives and stores the vehicle information and is linked with the security alarm system and the traffic guidance system; A communication module, serving as a backup interface to take over data transmission when the network communication is interrupted, ensuring continuous data upload to the server data center library; A self-recovery module, which detects the device operation status through a process nesting monitoring mechanism, automatically restarts when the system crashes and restores to the data state before the fault, and the recovery time is less than 10 seconds.

2. A vehicle information collection system for a parking lot according to claim 1, characterized in that, The camera access parsing module supports compatibility parsing of multiple video signal formats and can perform AI-enhanced recognition on license plate information under low-light and blocked conditions.

3. A vehicle information acquisition system for a parking lot according to claim 1, characterized in that, It includes a method for collecting parking lot vehicle information, and the steps are as follows: Step 1: Deploy a high-definition camera at the parking lot entrance and exit to capture vehicle images; Step 2: Parse the vehicle images through an AI image recognition module to extract license plate numbers, vehicle brands, models, and color information; Step 3: Upload the parsed vehicle information to the server data center library in real time through a 5G communication module; Step 4: The server data center library stores and analyzes the data and is linked with the security alarm system to match the vehicle information in real time.

4. A vehicle information collection system for a parking lot according to claim 1, characterized in that, It includes an intelligent parking guidance unit for autonomous driving vehicles, specifically including: A communication module, for real-time data interaction with the in-vehicle system of the autonomous driving vehicle; A path planning module, for generating the optimal parking path according to the real-time parking space information in the parking lot; A settlement module, for automatically calculating the fees when the vehicle leaves the lot and deducting the fees through an electronic payment interface.

5. A vehicle information collection system for a parking lot according to claim 1, characterized in that, It also includes a shared parking service unit, and the specific method is as follows: Step S1: The server data center library obtains the available parking space information of multiple parking lots in real time; Step S2: The user terminal queries and reserves a parking space in a specified parking lot through a mobile application; Step S3: After the reservation is successful, the vehicle information collection device automatically verifies the license plate information of the reserved vehicle and guides it to enter the lot.

6. A vehicle information collection system for a parking lot according to claim 1, characterized in that The data storage module adopts a dual-backup mechanism of solid-state drive (SSD) and cloud storage, supports encrypted data storage, and the local cache capacity can be extended to more than 30 days.

7. A vehicle information acquisition system for a parking lot according to claim 1, characterized in that, When the server data center library is linked with the security alarm system, it can compare the vehicle information with the suspected vehicle database in real time, trigger an audible and visual alarm after finding a matching target and push the location information to the security management terminal, generate traffic guidance signals through analyzing the parking lot saturation data and synchronize them to the electronic displays on the surrounding roads to guide vehicle diversion to idle parking lots.

8. A vehicle information collection system for a parking lot according to claim 1, characterized in that, The system is linked with the charging pile management system. The server data center library obtains the charging pile status information in real time, preferentially allocates parking spaces near idle charging piles for electric vehicles, and at the same time pushes charging guidance instructions through the user terminal.

Images