Trailer ID Recognition via Notation Format Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently detecting trailer IDs or container IDs from a multitude of character strings on trailers, due to the lack of a unified standard for trailer IDs and varying standards for container IDs.
Innovation Solution
A character recognition device that includes a recognizer to identify character strings from images of trailers and a selector to choose character strings conforming to a specified notation format, with a similarity to master data equal to or larger than a predetermined threshold, as the trailer ID or container ID.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a quadrangle detection method is used to identify license plate regions, then the detection is straightforward for standardized license plates, but it fails to detect trailer IDs or container IDs due to lack of unified standards for position, size, and character arrangement
Solution Approach 1:
The patent changes the detection parameters from fixed geometric constraints (quadrangle position, size, aspect ratio) to flexible character string attributes (notation format, character composition, length). This allows the system to adapt to various ID types including trailer IDs and container IDs that have different positional and size characteristics but share common notation patterns
Solution Approach 2:
The patent creates a universal detection method that can handle multiple types of identification strings (license plates, trailer IDs, container IDs) through a single character string detection approach. The system uses notation format specifications that can be configured for different ID types, making the detection mechanism multi-functional rather than requiring separate specialized detectors for each ID type
2Measurement precision
If character recognition is performed on all detected character strings, then all text is recognized, but it becomes difficult to identify which character string is the actual trailer ID or container ID among multiple character strings with various sizes, colors, and designs
Solution Approach 1:
The patent extracts the essential characteristics of target IDs (notation format, character composition patterns, length constraints) from the multitude of character strings on trailers. By specifying these key attributes in advance, the system can filter and identify the actual trailer ID or container ID from recognized character strings, separating the target information from non-target character strings
Solution Approach 2:
The patent implements a feedback mechanism where recognized character strings are evaluated against pre-specified notation formats and ID characteristics. The system uses this feedback to determine whether a recognized character string matches the expected pattern of a trailer ID or container ID, enabling selective identification among multiple recognized texts
3Adaptability or versatility
If multiple standards for container IDs are accommodated, then more container types can be identified, but the detection complexity increases due to varying standards and formats
Solution Approach 1:
The patent implements a dynamic detection system where notation format specifications can be adjusted and configured for different container ID standards. Rather than having fixed rigid detection rules, the system can adapt its detection criteria based on the type of container being identified, allowing it to handle multiple standards through a single flexible framework
Solution Approach 2:
The patent segments the detection process into modular components: character string detection, notation format validation, and ID type classification. Each component handles a specific aspect of the detection task, making the overall system more manageable and adaptable to different standards without requiring complete redesign for each container type
Data Source
AI summary
A character recognition device includes a recognizer that recognizes at least one character string from a plurality of character strings in an image including a trailer captured by an imaging device, and a selector. The selectors selects one of the at least one character string as the trailer ID/container ID based on the at least one character string having a similarity to a master data of a previously held trailer ID/container ID equal to or larger than a predetermined threshold value.