License Plate-Based Vehicle Geometry Estimation Under Occlusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle sensing technologies face challenges in accurately estimating the geometry of surrounding vehicles due to limited sensor views and potential occlusions, leading to distorted or incomplete geometry estimations.
Innovation Solution
A method and system that utilize a vehicle's database to retrieve vehicle geometry parameters linked to alphanumeric license plate codes, combined with on-board sensor data, to provide a robust and accurate estimation of surrounding vehicle geometry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If vehicle geometry is estimated using only on-board sensor data, then the system remains independent and real-time, but the estimation accuracy deteriorates due to occlusions and limited field of view
Solution Approach 1:
The patent combines on-board sensor data with external database information to estimate vehicle geometry. The system merges real-time sensor measurements with pre-stored vehicle geometry parameters from a database, achieving both system independence and high estimation accuracy by compensating for sensor limitations through additional data sources.
Solution Approach 2:
The patent introduces a database as an intermediary between the sensor system and the geometry estimation process. The database stores vehicle geometry parameters linked to license plate codes, acting as a mediator that provides supplementary information to enhance estimation accuracy without compromising real-time operation.
2Device complexity
If vehicle geometry is estimated using only on-board sensor data, then the system structure remains simple, but the estimation becomes distorted due to occlusions and relative position effects
Solution Approach 1:
The patent merges simple on-board sensor systems with external database queries to achieve accurate geometry estimation. The system maintains structural simplicity by using existing sensors while enhancing accuracy through intelligent data fusion with database information, avoiding complex multi-sensor arrangements.
3Measurement precision
If complete vehicle geometry information is required, then estimation accuracy improves, but data completeness deteriorates when objects are occluded in sensor field of view
Solution Approach 1:
The patent performs preliminary action by pre-storing complete vehicle geometry parameters in a database before they are needed for estimation. When a vehicle is detected, the system queries the database using the license plate code to retrieve complete geometry information, eliminating the need to capture all geometric details in real-time sensor data.
Solution Approach 2:
The database acts as an intermediary that stores and provides complete vehicle geometry information, compensating for information loss due to occlusions. The system queries this intermediary data source using license plate identification, ensuring complete geometry data is obtained regardless of sensor field of view limitations.
Data Source
Figure 1~2
Figure 3
AI summary
The invention relates to a method for providing an estimation of a vehicle geometry of a surrounding vehicle (40). In a first step of the method, an image (I) of at least a portion of the surrounding vehicle (40) is received, the portion comprising a license plate (42) affixed to the surrounding vehicle (40). Next, the license plate (42) is detected within the received image (I) and the alphanumeric identification code of the license plate (42) is extracted. Subsequently, a vehicle geometry parameter (P) being linked to the alphanumeric identification code is requested from a database (36) via a wireless data connection (38). Thereafter, the vehicle geometry parameter (P) is received from the database (36) via the wireless data connection (38), and the vehicle geometry is estimated based on the received vehicle geometry parameter (P). Moreover, a system (12) for providing an estimation of a vehicle geometry of the surrounding vehicle (40) and a vehicle (10) comprising such as system are described. Furthermore, a computer-readable medium is presented.