Tolling System Data Management via Onboard Unit Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Infrastructureless traffic tolling systems face challenges in efficiently managing data storage and computational resources due to the need to store and process large amounts of vehicle images and perform OCR on license plate numbers, especially on multilane roads with numerous vehicles.
Innovation Solution
A method that involves onboard units sending tolling messages with lane information and timestamps to a tolling server, allowing for the deletion of corresponding vehicle images from surveillance stations, thereby reducing data storage and computational requirements by correlating pictures with tolling records without the need for exhaustive OCR verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all vehicle pictures are stored in databases for enforcement purposes, then the system can detect vehicles with malfunctioning or non-used onboard units, but the storage requirements and computational resources for OCR processing become enormous
Solution Approach 1:
The patent extracts only the necessary identification information (license plate number) from the vehicle pictures using OCR, and stores only this extracted data along with corresponding metadata in the database. The full high-resolution pictures are not stored, only the essential information needed for enforcement detection.
Solution Approach 2:
Instead of storing the original pictures, the system creates simplified copies in the form of extracted text data (license plate numbers) and metadata. This copying process transforms the data into a compact format that retains the essential information for enforcement while dramatically reducing storage requirements.
2Measurement precision
If OCR processing is performed on all stored pictures to retrieve license plate numbers, then accurate vehicle identification is achieved, but huge computational resources are consumed
Solution Approach 1:
The patent applies OCR processing selectively and partially - only to the extent necessary to extract the license plate number from each picture. The processing is optimized to perform only the minimum necessary operation (text extraction) rather than full picture analysis or storage, reducing computational load while maintaining identification accuracy.
Solution Approach 2:
The system performs OCR processing and extracts license plate information as a preliminary step before storing data in the database. By completing the computationally intensive OCR task before storage, the system avoids repeated processing and reduces overall computational resource requirements during system operation.
3Measurement precision
If high quality pictures are stored for enforcement or reliable OCR readings, then accurate vehicle identification is ensured, but the storage space required increases significantly
Solution Approach 1:
The system extracts only the essential information (license plate number as text) from the pictures rather than storing the full high-quality images. This extraction approach maintains OCR reading reliability by capturing the necessary data in a compact text format, while dramatically reducing the storage space required in the database.
4Reliability
If surveillance stations record and store pictures of all passing vehicles, then complete enforcement data is available, but the system becomes inefficient on multilane roads with numerous vehicles
Solution Approach 1:
The system creates compact text-based copies of the essential enforcement information (license plate numbers and metadata) instead of storing full pictures. This copying approach maintains enforcement data completeness by preserving all necessary identification information while significantly improving system processing efficiency through reduced data volumes.
Data Source
Figure 1
Figure 2
AI summary
The present invention relates to a method for surveilling a tolling system (1), in which vehicles (3) carry onboard units (2) and travel on a multilane road (4) on which a surveillance station (7) is set up, comprising: taking a picture (PIC) of each passing vehicle (3); in a database, storing the pictures (PIC); in each onboard unit (2), when the vehicle (3) is within a predefined area (12) of the surveillance station (12), generating a and sending a tolling message via the mobile network (6) to the tolling server (5); in the database, deleting those pictures whose timestamp (tpic) matches the timestamp (tmsg) of a tolling message if the lane information (Linf) of this tolling message corresponds to the lane identification (Linf) of this picture.