Document Layout Automation Using Learned Image Regions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional document processing techniques require operators to explicitly specify layout locations for images, making it difficult to achieve an optimal layout for each image in a document.

Innovation Solution

A document processing apparatus and method that automatically separates image information into regions, associates this information with layout data, and uses learning information to lay out content without operator intervention, selecting suitable templates and adjusting layout based on user actions and frequency data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If operators explicitly specify layout locations for images, then layout control is precise, but operation complexity increases and automation is reduced

Engineering Contradiction:
Improveautomatic layout determinationVSAvoidoperator input requirement
Core Design Contradiction:
Extent of automationVSEase of operation

Solution Approach 1:

The system performs automatic layout determination using learning information stored in the holding unit, enabling the apparatus to serve itself without operator intervention. The layout unit automatically selects optimal positions for document components based on pre-learned layout patterns, eliminating the need for operators to manually specify layout locations while maintaining high-quality layout results

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

Layout information is learned and stored in advance in the holding unit during the learning phase. This preliminary action allows the system to have layout knowledge ready before actual document processing occurs, enabling automatic layout determination without requiring operators to provide input during the actual layout process

Inventive Principle:
Principle #10Preliminary action

2Productivity

If manual layout specification is required, then layout precision is controllable, but productivity decreases

Engineering Contradiction:
Improvedocument composition efficiencyVSAvoidoperator time for layout specification
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The layout unit automatically determines optimal positions for document components using learning information from the holding unit, enabling the system to perform layout tasks without operator intervention. This self-service capability significantly improves document composition efficiency by eliminating the time operators would spend manually specifying layout locations for each component

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The learning information is accumulated and refined continuously through the learning unit, which updates the holding unit with improved layout patterns over time. This continuous improvement ensures that the automatic layout determination becomes increasingly accurate and efficient, maintaining high productivity without sacrificing layout quality

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentUS8910038B2Apparatus and method for image separation, layout, and templating
Publication Date: 2014.12.09 CANON KK
  • US8910038B2 patent drawing
  • US8910038B2 patent drawing
  • US8910038B2 patent drawing

AI summary

A document processing apparatus accepts input of image information which represents a document, separates the image information as content of the document into regions, associates the image information separated as content into regions with information used to lay out the content, and holds the associated information as learning information. Subsequently, the document processing apparatus lays out the content of the inputted document based on the held learning information and outputs a document in which the content is laid out.