Vehicle Image Collection System with Character Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of image data collected by vehicle cameras struggle to effectively utilize the information provided, as they cannot immediately understand the relevance of signs, banners, or other information captured in images, leading to potential misinterpretation of locations or events.
Innovation Solution
An information collection system where vehicles equipped with cameras and sensors transmit images and position data to an information center, which recognizes character information, determines points of interest or areas, and generates association information linking the two, enabling users to understand the relevance of the information directly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If images are stored and provided to vehicles without additional processing, then the quantity of image data is maintained, but the usability and understandability of the information deteriorates
Solution Approach 1:
The system performs preliminary actions by automatically recognizing character information in images, determining relevant points of interest or areas, and creating association information before the images are provided to vehicles. This preprocessing enables users to immediately understand the relevance of captured information without manual analysis, thereby improving usability while managing complexity through automated processing.
Solution Approach 2:
The patent introduces an intermediary processing layer that acts as a mediator between the raw image data and the end user. This intermediary system automatically extracts character information from images, determines associated points of interest or areas, and creates association information that bridges the gap between unprocessed images and meaningful information for users, thereby enhancing usability without requiring users to directly analyze raw images.
2Measurement precision
If character recognition and association determination are performed on all received images, then the accuracy of information association is improved, but the processing time and computational load increases
Solution Approach 1:
The system applies local quality by focusing character recognition and association determination only on specific regions or images that contain relevant information. Rather than uniformly processing all images, the system identifies and processes only those images where character recognition is likely to yield meaningful results, thereby maintaining high accuracy of information association while reducing overall processing time and computational resources required.
3Loss of information
If association information is created for every recognized character, then the completeness of information is improved, but the data volume and storage requirements increase
Solution Approach 1:
The system extracts only the essential and relevant association information from recognized characters, rather than creating complete association data for every single character detected. By selectively extracting and storing only the most pertinent associations between character information and points of interest or areas, the system maintains information completeness for useful data while significantly reducing the overall data volume and storage requirements.
Data Source
AI summary
Each of vehicles acquires its own position, and transmits an image of a nearby outside taken by an own camera and a piece of information of the own position. An information center communicatable wirelessly with the vehicles receives the images and the pieces of information of the vehicles' positions transmitted by the vehicles, and recognizes pieces of first character information appearing in the received images, respectively. Based on the received pieces of information of the vehicles' positions as well as at least either points of interest or areas stored as collection target information, the information center determines at least either points of interest or areas for the recognized pieces of first character information, respectively, and stores association information where the pieces of first character information are associated with the determined at least either points of interest or areas.