License Plate State Identification via Multi-OCR Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ALPR systems face challenges in accurately determining the issuing state of a license plate due to variations in font styles across different states, which affects the overall accuracy and efficiency of multi-state ALPR solutions.
Innovation Solution
The approach involves running multiple OCR engines, each tuned to a specific state, and using a Bayesian-based method to merge confidence data and syntax information to estimate the issuing state, leveraging intermediate OCR knowledge to improve identification accuracy without relying on image content outside the plate characters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple OCR engines are deployed, each tuned to a specific state font, then state identification accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the license plate recognition task by deploying multiple specialized OCR engines, each tuned to recognize a specific state's font style. Instead of using a single general-purpose OCR engine, the system divides the recognition task across multiple engines specialized in different state-specific fonts, thereby improving accuracy for each state while managing complexity through modular architecture
Solution Approach 2:
The system changes the parameter of font style recognition by training each OCR engine with state-specific font characteristics. Each engine is optimized with different parameter settings corresponding to its assigned state's license plate font, allowing the system to adapt to varying font styles across different states without requiring a completely different recognition approach
2Measurement precision
If multiple OCR engines are used to improve accuracy, then processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-training each OCR engine with state-specific font characteristics before deployment. The engines are prepared in advance with their specialized knowledge, so during actual license plate recognition, they can quickly and accurately process images without requiring extensive computation or sequential processing of multiple general-purpose engines
Solution Approach 2:
The system creates specialized copies of OCR engines for each state, where each copy is trained on that specific state's license plate font. Rather than using one general engine that attempts to recognize all fonts, the system deploys targeted copies optimized for specific state fonts, improving recognition speed and accuracy for each state's unique characteristics
Data Source
AI summary
Methods and systems for automatically determining the issuing state of a license plate. An image of a license plate acquired by an ALPR engine can be processed via one or more OCR engines such that each OCR engine among the OCR engines is tuned to a particular state. Confidence data output from the OCR engine(s) can be analyzed (among other factors) to estimate the issuing state associated with the license plate. Multiple observations related to the issuing state can be merged to derive an overall conclusion and assign an associated confidence value with respect to the confidence data and determine a likely issuing state associated with the license plate.