OCR Condition Setting for Scan Image File Naming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for setting file names using character recognition from scanned documents often result in errors due to inappropriate OCR processing conditions, leading to decreased accuracy and user inconvenience.
Innovation Solution
An apparatus that allows users to select character areas in a scan image and set conditions for OCR processing based on the selection order and format, performing OCR processing with determined conditions to enhance character recognition accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If OCR processing is performed with predetermined condition settings on every character area, then the process is automated and efficient, but the character recognition accuracy decreases due to inappropriate conditions for specific areas
Solution Approach 1:
The patent divides the document into multiple character areas and applies different OCR processing conditions to each area based on its characteristics. This segmentation allows the system to maintain automated processing while improving accuracy by tailoring conditions to specific regions, such as distinguishing between alphabetic text and numeric fields.
Solution Approach 2:
The patent implements local quality by setting different OCR conditions for different character areas within the same document. Each area receives customized processing parameters appropriate to its content type, thereby resolving the contradiction between automated efficiency and localized accuracy requirements.
2Reliability
If users manually edit character strings obtained by OCR processing, then error corrections can be made, but the operation becomes more complex and time-consuming
Solution Approach 1:
The patent performs preliminary OCR processing with area-specific conditions before user interaction, pre-processing the text to minimize errors. This preliminary action reduces the need for subsequent manual editing while maintaining high accuracy, thereby resolving the contradiction between reliability and ease of operation.
3Measurement precision
If OCR processing conditions are customized for each character area, then recognition accuracy improves, but the device complexity and setup time increase
Solution Approach 1:
The patent implements self-service by enabling the system to automatically determine and apply appropriate OCR conditions for each character area without requiring complex manual configuration. The system autonomously analyzes document structure and selects processing parameters, reducing setup complexity while maintaining high recognition accuracy.
Data Source
AI summary
In a situation of setting a file name and the like by using a character string obtained by performing OCR processing to a scan image, appropriate conditions can be set according to a character string to be scanned so as to increase a character recognition rate. There is provided an apparatus for performing a predetermined process to a scan image obtained by scanning a document, including: a display control unit configured to display a UI screen for performing the predetermined process, the UI screen displaying a character area assumed to be one continuous character string in the scan image in a selectable manner to a user; and a setting unit configured to determine a condition for OCR processing based on selection order of a character area selected by a user via the UI screen and a format of supplementary information for the predetermined process, perform OCR processing by using the determined condition for OCR processing to the selected character area, and set supplementary information for the predetermined process by using a character string extracted in the OCR processing.