Document Distribution System with OCR-Based Enlargement Area Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing document distribution systems face challenges in accurately classifying scanned images, requiring users to manually specify enlargement areas for various document types, which is labor-intensive and prone to errors, especially when display screens are small.
Innovation Solution
A document distribution system that includes a first obtaining unit for sample images and folder names, an analysis unit for character recognition, a determination unit to identify matching character strings, a registration unit to associate enlargement areas with sample images, and a display control unit to enlarge and display images based on registered enlargement areas, reducing user burden by automating the process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If users manually specify enlargement areas for each document type, then display accuracy is improved, but user workload increases
Solution Approach 1:
The system performs automatic character recognition on sample images to extract text and determine enlargement areas without requiring manual user input. The OCR engine automatically identifies character regions and sets enlargement areas based on the recognized text layout, allowing the system to serve itself rather than requiring user intervention for each document type.
Solution Approach 2:
The system performs character recognition and enlargement area determination in advance during the sample image registration phase. By pre-processing the sample images and storing the determined enlargement areas in the database, the system prepares all necessary information beforehand, eliminating the need for users to perform these operations when actual document classification is needed.
2Adaptability or versatility
If display screen size is small, then device portability is improved, but image visibility deteriorates
Solution Approach 1:
The system transitions from displaying the entire document image to displaying an enlarged view of specific character regions. By extracting and enlarging only the relevant text areas identified through OCR, the system provides detailed visibility on small screens without requiring the entire document to be visible, effectively using a different dimensional approach to information presentation.
Solution Approach 2:
The system applies different display qualities to different parts of the document by enlarging specific character regions while keeping other areas at normal or reduced size. This local enhancement allows critical information to be clearly visible on small screens without proportionally increasing the overall display area requirement, maintaining portability while improving local visibility.
3Ease of operation
If automatic character recognition is implemented, then user input is reduced, but processing time increases
Solution Approach 1:
The system performs character recognition during the sample image registration phase rather than during actual document processing. By pre-processing the sample images and storing the recognized character information and enlargement areas in the database, the system shifts the processing time to when it is less critical, enabling fast retrieval and classification during actual use.
Data Source
AI summary
A sample image and a folder name of a selected saving destination folder are obtained, and character recognition processing is performed on the sample image to obtain character strings. When it is determined that the same character string as the folder name of the saving destination folder exists among the obtained character strings, an area where the same character string is detected as an enlargement area, and the identified enlargement area, the sample image, and the folder name are registered while being associated with one another. When a target image on which distribution processing is performed is obtained, a sample image similar to the obtained target image is identified, and an enlargement area registered while being associated with the identified sample image is identified. An area of the target image corresponding to the enlargement area identified by the identification unit is enlarged and displayed on a folder specification screen.