Vehicular Sign Recognition Using Extracted Character Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sign recognition systems for vehicles face inefficiencies in processing and communicating large amounts of data when analyzing multiple information signs, leading to increased time for data communication and processing, which complicates the identification of signs captured by vehicular cameras.
Innovation Solution
A sign recognition system that extracts and recognizes specific character types, such as numerals, from captured image data and stores this information in a database, reducing the amount of data processed and enabling faster identification by limiting the recognition dictionary and using image processing techniques to extract and arrange characters for attribute information items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all captured sign data is processed and communicated, then complete sign information is obtained, but data transmission time and processing requirements increase significantly
Solution Approach 1:
The patent extracts only specific character types (such as numerals) from sign images rather than processing all visual data. The extraction unit identifies and extracts characters belonging to predetermined types from the sign image, thereby reducing the amount of data that needs to be transmitted and processed while retaining essential identification information.
Solution Approach 2:
The patent creates simplified copies of sign data by converting full sign images into extracted character strings. Instead of transmitting and processing complete images, the system uses extracted text copies (attribute information items) that contain the essential identification data, significantly reducing computational requirements and transmission time.
2Measurement precision
If character recognition is performed on all sign data, then accurate sign identification is achieved, but processing complexity and time increase
Solution Approach 1:
The system extracts only specific character types (e.g., numerals) that are essential for sign identification rather than analyzing all visual features. This selective extraction simplifies the recognition process by focusing on key identifying elements while ignoring redundant information.
Solution Approach 2:
The patent changes the parameter being processed from complete image data to extracted character strings. By transforming the data format from pixels to text characters, the system reduces processing complexity while maintaining identification accuracy, as text data is more efficient to store, transmit, and compare.
3Loss of information
If complete sign images are stored in the database, then full visual information is preserved, but storage requirements and data communication overhead increase
Solution Approach 1:
The patent stores simplified text copies (extracted character strings) in the database instead of complete image files. These text copies serve as efficient representations that preserve essential identification information while occupying minimal storage space and enabling rapid communication between system components.
Solution Approach 2:
The system extracts only the essential identifying elements (specific character types) from complete sign images and stores only these extracted elements in the database. This selective storage approach maintains the ability to identify signs accurately while dramatically reducing the quantity of data that must be stored and managed.
Data Source
AI summary
A sign recognition system according to an exemplary aspect of the present disclosure is configured to, when image data of a sign is captured by a vehicular camera, recognize one or more characters, which are included in the sign and belong to at least one specified character type, from the image data, thus extracting, from the image data, the recognized one or more characters. The sign recognition system is also configured to store a sign dataset in a sign database; the sign dataset includes (i) an installation-position information item indicative of the installation position of the sign, and (ii) an attribute information item about the sign.