Document Image Extraction via Additional Object Registration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current image processing systems lack an efficient method to specify and extract specific areas from documents for various business document formats, limiting their ability to accurately extract and process information.

Innovation Solution

An image processing apparatus with an additional-object registration unit and a read-image processing unit that allows users to specify additional objects on a document to define extraction areas and associate them with processing tasks, enabling the system to search for and perform processing on these areas within the document images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional image processing systems are used to extract information from documents, then the system structure remains simple, but the system lacks the capability to efficiently specify and extract specific areas from various document formats

Engineering Contradiction:
Improvecapability to extract specific areas from various document formatsVSAvoidsystem structure
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system segments the document processing task into distinct components: additional objects define extract areas, the additional-object registration unit registers these objects with their corresponding processing operations, and the read-image processing unit executes the processing. This segmentation enables versatile document format handling while maintaining manageable system complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If manual specification of extract areas is implemented, then extraction precision improves, but the operation time and complexity increase

Engineering Contradiction:
Improveextraction precisionVSAvoidoperation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The additional-object registration unit performs preliminary registration of additional objects and their associated processing operations before actual document processing. This pre-registration stores the mapping between additional objects and processing operations, so that during actual processing, the system can quickly retrieve and execute the appropriate processing without time-consuming manual specification, thus maintaining high extraction precision while reducing operation time.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If multiple processing operations are supported for different document formats, then system versatility improves, but the control complexity increases

Engineering Contradiction:
Improveprocessing operations supportVSAvoidcontrol complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The additional-object registration unit acts as an intermediary between the user-specified additional objects and the read-image processing unit. It registers and manages the associations between additional objects and processing operations, serving as a control layer that simplifies the management of multiple processing operations. This intermediary structure allows the system to support diverse processing operations for different document formats while maintaining simple control through a unified registration and retrieval mechanism.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2890100B1Image processing apparatus
Publication Date: 2020.12.16 KYOCERA DOCUMENT SOLUTIONS INC
  • EP2890100B1 patent drawingFigure 1
  • EP2890100B1 patent drawingFigure 2
  • EP2890100B1 patent drawingFigure 3

AI summary

This image processing apparatus (1) includes an additional-object registration unit (21) and a read-image processing unit (22). A setting form contains: (a) an additional-object specification field (41) used to present an additional object (51 - 55) that is placed onto a document in order to specify an extract area to be extracted from an image read from the document; and (b) a processing specification field (42) used to select processing to be performed on information obtained from the extract area. The additional-object registration unit (21) identifies an image of the additional object (51 - 55) presented in the additional-object specification field (41) and the processing selected in the processing specification field (42), and registers the image and the process associated therewith. The read-image processing unit (22) searches the read image of the document for the image of the additional object (51 - 55) and performs the processing associated with the additional object image on the information obtained from the extract area specified by the additional object image.