Document Translation System Preserving Layout via Text Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing translation methods fail to retain the contextual information and structure of documents when translating text from one language to another, leading to incomplete understanding due to loss of layout and formatting.

Innovation Solution

A method and system that segment an input image of a document to extract text and layout information, translate the text while preserving the original layout, and generate an editable translated electronic document with identical structure, using visual and semantic analyses and components like imaging, segmenting, image processing, translation, and output engines.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If text translation is performed without preserving layout information, then translation speed is improved, but contextual information and document structure are lost

Engineering Contradiction:

Improvetranslation speedVSAvoidcontextual information and document structure

Core Design Contradiction:

ProductivityVSLoss of information

Solution Approach 1:

The document is segmented into multiple regions using image processing techniques. Each region contains specific text elements with their layout information preserved. This segmentation allows the system to process and translate text while maintaining the original document structure, resolving the contradiction between translation speed and information preservation by enabling parallel processing of segmented regions.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Layout information serves as an intermediary between the original document image and the translated text output. The system extracts layout information from the input image, uses it to guide text translation, and then applies the layout information to the translated text to reconstruct the document structure. This intermediary mechanism enables both efficient translation and structure preservation.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If identical text strings are translated separately, then translation accuracy is improved, but processing time increases

Engineering Contradiction:

Improvetranslation accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system merges identical text strings from different regions into a single translation unit. By identifying and consolidating duplicate text elements, the system translates each unique string only once while maintaining accurate translations across all occurrences. This merging approach reduces redundant processing time while preserving translation accuracy through consistent handling of identical content.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS11604930B2Generation of translated electronic document from an input image by consolidating each of identical untranslated text strings into a single element for translation

Publication Date: 2023.03.14 KONICA MINOLTA BUSINESS SOLUTIONS USA INC

AI summary

A method of generating an editable translated electronic document from an input image of an original document with a first layout includes: segmenting the input image to generate a first region including first untranslated text; extracting, from the first region, the first untranslated text and a first layout information; generating an editable output data including the first untranslated text and the first layout information; translating the first untranslated text into a translated text; editing the output data to include the translated text; and generating, using the first layout information, the translated electronic document including the translated text and a second layout that is identical to the first layout.