Tag Information Extraction and Document Image Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing technologies fail to effectively separate document images from tags attached to them, leading to difficulties in maintaining tag information, requiring manual efforts for classification, and risking document loss or tag detachment during scanning.

Innovation Solution

An image processing device and method that extracts tag information, removes tags from document images, and generates separate images without tags, while associating tag information with the document images, allowing for grouping and outputting of tag information and document images independently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a document with a tag is read by a scanner, then the document can be converted into data, but the tag portion appears dirty and may not be desired on the document image

Engineering Contradiction:

Improvedocument conversion efficiencyVSAvoidtag appearance quality

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent segments the document image processing into two distinct outputs: a clean document image with the tag portion removed, and a separate tag information image containing the extracted tag data. This segmentation allows the document image to be free from tag contamination while preserving tag information separately.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts the tag information from the document image and removes the tag portion from the document image. The tag information is then placed in a separate tag information image, effectively taking out the harmful element (tag appearance) while preserving its informational value.

Inventive Principle:

Principle #2Taking out (Extraction)

2Ease of operation

If manual efforts are used for classification based on tag type, then documents can be sorted, but there is a high risk of losing documents and peeling off tags

Engineering Contradiction:

Improvedocument classification capabilityVSAvoiddocument integrity

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent replaces the manual mechanical classification process with an automated image processing system. The system automatically detects tags in the document images, extracts tag information, and performs classification without human intervention, thereby eliminating the risks of document loss and tag detachment associated with manual handling.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a digital copy of the tag information in the tag information image, preserving the classification data without requiring physical tags to remain attached to documents during handling and sorting processes.

Inventive Principle:

Principle #26Copying

3Object-affected harmful factors

If tag information is removed from the document image, then the document image is cleaned, but it becomes difficult to trace the history of the tag

Engineering Contradiction:

Improvedocument image cleanlinessVSAvoidtag history information

Core Design Contradiction:

Object-affected harmful factorsVSLoss of information

Solution Approach 1:

The patent moves tag information from the spatial dimension (visual appearance in document image) to a separate informational dimension (tag information image). This dimensional separation allows the document image to be clean while the tag history is preserved in a different format and location.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces a tag information image as an intermediary that carries the tag history information. This intermediary preserves the tag information that would otherwise be lost when the tag portion is removed from the document image.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Productivity

If documents are grouped using delimitation by tag, then classification is achieved, but it cannot be employed if documents are not arranged in order

Engineering Contradiction:

Improvedocument grouping efficiencyVSAvoidprocessing flexibility

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent replaces the mechanical arrangement requirement (documents must be physically ordered) with an automated image processing system that can detect and analyze tags regardless of document sequence. The system processes documents in any order and performs grouping based on tag information extraction and analysis.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11805216B2Image processing device and image processing method capable of reading document selectively attached with a tag

Publication Date: 2023.10.31 SHARP KK

AI summary

An image processing device reads a plurality of documents selectively attached with a tag, and processes the read document images with a tag. The image processing device includes a tag information extractor that extracts tag information about a tag from the document images with a tag, and an image processor that separately generates, based on the tag information, document images without a tag in which a tag is removed from the document images with a tag, and a tag information image in which the tag information about the document images without a tag is written.