Document Search Engine Image Text Term Integration

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current search engines are not effective in accurately searching for images and related text in documents, as the text near the images may not be related to the search keywords, leading to reduced accuracy.

Innovation Solution

A method and apparatus that identify text and images in documents, extract features from images, and generate image terms, which are then combined with text terms to search a collection of documents, allowing for a more structured analysis of the relationship between images and surrounding text.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If image search is performed separately from text search, then image retrieval capability is improved, but search accuracy deteriorates because text near images may not be related to search keywords

Engineering Contradiction:

Improveimage search capabilityVSAvoidsearch accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent combines image search and text search into a unified search framework. Image features are extracted and converted into image terms that are integrated with text terms from the document. This merging allows the search system to simultaneously consider both image content and surrounding text context, resolving the contradiction between maintaining separate image search capability and improving overall search accuracy.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces image terms as an intermediary representation that bridges image content and text-based search. By extracting features from images and converting them into searchable terms that can be combined with text terms, the system creates a mediator that enables accurate joint searching without requiring separate search processes.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If text-based search techniques are used for images, then search process simplicity is improved, but image understanding capability deteriorates

Engineering Contradiction:

Improvesearch process simplicityVSAvoidimage understanding accuracy

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent replaces direct image analysis with a substitution approach where image features are extracted and transformed into image terms. This substitution allows text-based search techniques to be applied to image content without losing the essential image understanding capability, as the extracted features preserve the semantic information needed for accurate search.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the representation parameters of images by extracting features and converting them into terms that match the text search parameter space. This parameter transformation enables images to be searched using text-based techniques while maintaining the ability to accurately represent and understand image content through the extracted features.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10331732B1Information searching system

Publication Date: 2019.06.25 NATIONAL TECHNOLOGY & ENGINEERING SOLUTIONS OF SANDIA LLC

AI summary

A method and computer system for searching a collection of documents. The method comprises first identifying text in the document and then identifying an image in the document. Then the features of the image are extracted from the document and image terms are identified from the features. Finally, the collection of documents is searched using the image terms in conjunction with text terms.