Character Image Extraction Using Overlap Merging and Size Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional document filing apparatuses inaccurately extract images of characters from character string regions, including punctuation marks and noise parts, which are not necessary for searching.
Innovation Solution
A character image extracting method and apparatus that involves extracting linked components from character string regions, altering section elements by combining overlapping parts, and selecting elements based on predetermined size criteria to accurately isolate character images while excluding punctuation marks and noise parts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional extraction methods are used to obtain all elements from character string regions, then the extraction process is simple, but punctuation marks and noise parts are incorrectly included along with character images
Solution Approach 1:
The extraction process is divided into multiple sequential steps: initial extraction of all elements, alteration by combining overlapping elements, and selection based on size criteria. This segmentation allows each step to focus on a specific aspect of the problem, improving overall accuracy while maintaining manageable complexity.
Solution Approach 2:
Before final selection, the method performs preliminary alteration by combining overlapping section elements. This preliminary action ensures that fragmented character parts are merged into complete character images before the selection step, preventing incomplete extractions.
2Measurement precision
If all extracted elements are processed without filtering, then the processing speed is fast, but unnecessary elements such as punctuation marks and noise parts are included in the results
Solution Approach 1:
The method extracts only the necessary information (character images) by applying size-based filtering criteria. Elements that do not meet the predetermined size threshold (such as punctuation marks and noise) are automatically excluded, achieving both accuracy and efficiency.
Solution Approach 2:
The method uses size parameters as a filtering criterion to distinguish character images from other elements. By setting a predetermined reference size, the system automatically differentiates between valid characters and unwanted elements like punctuation and noise, improving both accuracy and processing efficiency.
3Measurement precision
If section elements are combined without overlap detection, then the processing is simpler, but fragmented character parts remain separated instead of being merged into complete character images
Solution Approach 1:
The method merges overlapping section elements to form complete character images. By detecting spatial overlap between extracted elements, the system combines fragmented parts into unified character representations, significantly improving extraction accuracy.
Solution Approach 2:
The method creates a consolidated representation of character images by copying and merging overlapping elements. This approach ensures that all parts of a character are captured in a single unified element, preventing fragmentation while maintaining processing efficiency.
Data Source
AI summary
In an extracting step, the extracting portion obtains a linked component composed of a plurality of mutually linking pixels from a character string region composed of a plurality of characters, and extracts section elements from the character string region, the section elements each being surrounded by a circumscribing figure circumscribing to the linked component. In the first altering step, the first altering portion combines section elements at least having a mutually overlapping part among the extracted section elements so as to prepare a new section element. In the first selecting step, the first selecting portion determines a reference size in advance and selects section elements having a size greater than the reference size, from among the section elements altered in the first altering step.