Document Processing Apparatus Unified Extraction Flow
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information processing technologies are inefficient in reducing the overall processing time for extracting and processing regions with character images from completed documents, as they only facilitate the extraction step without addressing the broader processing flow.
Innovation Solution
An information processing apparatus with an acquirer and an updater that acquires and updates processing procedure information, including the extraction step and other steps, to streamline the processing of completed document images, enabling automated execution of form, test marking, and questionnaire processing without user intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If only the extraction step is facilitated by appending annotations to the template document, then the extraction of character-filled portions is improved, but the overall processing time is not significantly reduced
Solution Approach 1:
The patent applies preliminary action by pre-appending annotations to the template document that describe not only the extraction regions but also the subsequent processing steps and their parameters. This allows the entire processing flow (extraction, recognition, and further processing) to be pre-configured, enabling automated execution without user intervention and significantly reducing overall processing time while maintaining high extraction efficiency
Solution Approach 2:
The patent merges the annotation function with the processing procedure information by combining the description of extraction regions with the description of subsequent processing steps into a unified annotation structure. This integration allows the system to treat the extraction step and subsequent processing as a unified flow, improving overall productivity by eliminating gaps between steps and enabling fully automated processing
2Ease of operation
If annotations are appended to describe extraction regions, then the extracting step is facilitated, but the processing procedure becomes complex and requires multiple separate steps
Solution Approach 1:
The patent merges the annotation function with the processing procedure information by combining the description of extraction regions with the description of subsequent processing steps into a unified annotation structure. This integration allows the system to treat the extraction step and subsequent processing as a unified flow, simplifying the overall procedure while maintaining ease of operation
Solution Approach 2:
The patent creates a universal annotation structure that serves multiple functions: it describes extraction regions, specifies subsequent processing steps, defines processing parameters, and enables automated execution. This multi-functional annotation system eliminates the need for separate configuration files and multiple processing stages, reducing complexity while maintaining operational ease
Data Source
AI summary
An information processing apparatus includes an acquirer and an updater. The acquirer acquires a template document image obtained as a result of reading a template document. The updater updates, based on the template document image, processing procedure information indicating a procedure of processing including an extracting step and another step to processing procedure information indicating a procedure of processing including the extracting step and a step whose content is updated. The processing is processing to be executed based on a completed document image obtained as a result of reading a completed document generated by filling characters into the template document. The extracting step is a step of extracting a region including a character image from the completed document image.