Document Thumbnail Classification for Rapid Page Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods for generating thumbnail images of documents are inefficient, requiring users to manually search through numerous images to find a specific page, as they lack effective classification and navigation features, leading to increased time spent in specifying desired pages.
Innovation Solution
An image processing apparatus and method that extracts and classifies items such as titles, headings, figures, and keywords from scanned documents, allowing users to easily identify and navigate through thumbnail images by grouping and highlighting relevant items, and generating bookmarks for quick access to specific sections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If thumbnail images of scanned documents are displayed to allow users to specify desired pages, then users can visually identify document content, but the number of thumbnail images increases making it difficult to specify a desired image quickly
Solution Approach 1:
The patent divides the document into multiple pages and extracts representative items (titles, headings, figures, tables, keywords) from each page to create structured thumbnail information. This segmentation allows users to quickly identify specific pages without viewing entire documents, resolving the contradiction between content visibility and specification time.
Solution Approach 2:
The patent introduces an intermediary classification system that categorizes extracted items into groups (titles, headings, figures, tables, keywords) and displays them with visual indicators. This intermediary layer between the full document and thumbnail provides structured navigation, enabling users to locate desired pages efficiently while maintaining content visibility.
2Ease of operation
If extracted items from scanned documents are classified into groups and displayed differently for each group, then users can easily identify and navigate to desired sections, but the device complexity increases due to classification and grouping functions
Solution Approach 1:
The patent implements automatic classification of extracted items based on their inherent properties (titles, headings, figures, tables, keywords) without requiring manual user intervention. The system self-organizes the information into structured groups and applies appropriate display indicators, providing ease of navigation while minimizing the operational complexity the user experiences.
Solution Approach 2:
The patent changes the display parameters of extracted items based on their classification groups, applying different visual indicators (such as icons, colors, or positioning) for different item types. This parameter-based differentiation enhances navigability while maintaining a relatively simple system structure by using standardized display rules for each category.
Data Source
AI summary
An image processing system includes a client apparatus and a server apparatus. The server apparatus includes an item extraction unit, an item classification unit, and an image selection processing unit. The item extraction unit extracts a prescribed item from a document. The item classification unit classifies the extracted item into any of a plurality of groups. The image selection processing unit generates a display screen displaying each item included in read data in a manner different for each group. A display of the client apparatus displays the generated display screen.