Document Separation Using Calibration Marks and Image Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing electronic document management (EDM) solutions are inefficient and require specific devices or complex processes to accurately separate multiple documents from a batch during digitization, often involving errors and increased costs due to the need for specific equipment, manual interaction, and complex recognition algorithms.

Innovation Solution

A method that uses a calibration page with user-chosen reference marks, allowing users to easily apply similar separation marks on document ends, enabling automated detection and separation using standard devices without requiring specific equipment or extensive computer interaction, utilizing color and location similarities for accurate page separation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If standard scanning devices are used to digitize batches of documents, then productivity is improved by scanning multiple documents in succession, but document separation becomes difficult without specific detection devices or complex algorithms

Engineering Contradiction:
Improvebatch scanning speedVSAvoidseparation detection complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The method applies separation marks to documents before digitization begins. The marking step occurs prior to scanning, preparing the documents with detectable features that enable automatic separation during or after scanning without requiring complex real-time detection hardware

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The invention introduces separation marks as an intermediary element between the physical document and the digital representation. These marks serve as mediators that the recognition algorithm can detect to identify document boundaries, simplifying the separation task compared to analyzing document content directly

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If specific detection devices are used to identify document boundaries, then measurement precision is improved for detecting separation points, but device complexity and cost increase

Engineering Contradiction:
Improvedocument boundary detection accuracyVSAvoiddetection equipment complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The invention replaces mechanical or specialized detection devices with a software-based recognition algorithm that processes standard scanned images. Instead of using physical sensors or specialized hardware to detect document boundaries, the system uses image processing algorithms to identify separation marks in the digitized pages

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The method uses visual copies (separation marks) that can be detected through standard scanning and image recognition processes. Rather than requiring specialized physical detection devices, the system creates detectable visual patterns that can be identified through software analysis of the scanned image data

Inventive Principle:
Principle #26Copying

3Measurement precision

If complex recognition algorithms are used to separate documents, then separation accuracy is improved, but processing time and computational resources increase

Engineering Contradiction:
Improveseparation accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

By pre-marking documents with separation marks before scanning, the system simplifies the subsequent recognition task. The algorithm only needs to detect these predetermined marks rather than analyzing complex document content to identify boundaries, reducing computational complexity and processing time

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The invention changes the detectable parameters in the document images by adding separation marks with specific visual characteristics (color, pattern, position). This transforms the separation problem from analyzing complex document content to detecting simple, distinct visual parameters that can be quickly identified by algorithms

Inventive Principle:
Principle #35Parameter changes

4Reliability

If manual interaction is required for each document, then separation accuracy is maintained, but productivity decreases due to increased human interaction time

Engineering Contradiction:
Improveseparation reliabilityVSAvoidbatch processing efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system enables automatic document separation through recognition algorithms that autonomously identify separation marks and split documents without human intervention. The batch processing capability combined with automatic recognition allows the system to handle multiple documents independently, eliminating the need for manual review of each document while maintaining reliable separation

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3440827B1Method and system for separating documents during batch digitization
Publication Date: 2022.03.30 SAS SAGES INFORMATIQUE
  • EP3440827B1 patent drawingFigure 1
  • EP3440827B1 patent drawingFigure 2a~2b
  • EP3440827B1 patent drawingFigure 3

AI summary

The present invention relates to the processing of a batch of documents comprising the following steps: a) creating (100) by a user a calibration page comprising at least one reference mark having properties chosen by a user; b) applying (110) by a user to an end page of each document to be processed at least one separation mark whose properties are similar to those of one of the reference marks of the calibration page; c) inserting (120) by a user the calibration page created according to step a) at one of the ends of the batch of documents to be processed; d) digitizing (130) the batch of documents thus prepared according to steps a) to c); e) subsequent to the digitization of the first page of the batch of documents to be processed, detecting (140) the presence/absence of the calibration page; and f) in case of presence of the calibration page, recognizing (160) a page comprising at least one separation mark and in case of positive recognition separating (170) the document from the batch as a function of the separation mark thus recognized.