Image Search Apparatus Using Visual Feature Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current trademark and logo searching methods are cumbersome and time-consuming due to reliance on text-based descriptions, leading to difficulties in identifying similar images, especially for abstract logos, as they require sifting through tens of thousands of text descriptors and can result in missed matches due to descriptor inconsistencies.

Innovation Solution

An image searching method and apparatus that utilize electronic processing devices to segment, process, and rank images by identifying similar images through image recognition and metadata tags, allowing for the identification of visually similar images and further refinement using text-based descriptors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If text-based descriptions are used to search trademark images, then the search can be performed using existing metadata, but the search becomes cumbersome and time-consuming

Engineering Contradiction:

Improveease of search operationVSAvoidsearch time

Core Design Contradiction:

Ease of manufactureVSLoss of time

Solution Approach 1:

The patent replaces manual text-based searching with automated image recognition technology. The system automatically analyzes uploaded images, extracts visual features, and performs similarity matching against the trademark database, eliminating the need for manual text description and significantly reducing search time while improving accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service searching where users simply upload an image and the system automatically performs the entire search process including image processing, feature extraction, similarity comparison, and result ranking without requiring user intervention in the complex search parameters or text descriptors

Inventive Principle:

Principle #25Self-service

2Reliability

If text descriptors are used to describe images for searching, then the search can be performed against metadata, but matches may be missed due to descriptor inconsistencies

Engineering Contradiction:

Improvesearch accuracyVSAvoidmissed matches

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent replaces unreliable manual text descriptor matching with automated image recognition and visual feature analysis. The system extracts objective visual features directly from images and performs similarity comparison based on actual visual content rather than subjective text descriptions, ensuring consistent and accurate matching

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the search parameter from text-based descriptors to visual feature-based parameters. By transforming the search methodology from textual metadata comparison to direct image feature analysis, the system achieves more reliable and consistent matching results that are not affected by descriptor inconsistency

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If image recognition is used to identify similar images, then the search becomes more accurate, but the processing complexity increases

Engineering Contradiction:

Improveimage similarity detection accuracyVSAvoidsearch system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the image search process into distinct modular stages: image preprocessing, feature extraction, similarity comparison, and result ranking. Each module handles a specific task independently, making the complex system manageable and maintainable while achieving high accuracy through specialized processing at each stage

Inventive Principle:

Principle #1Segmentation

4Reliability

If tens of thousands of text descriptors are sifted through, then comprehensive coverage is achieved, but the search becomes cumbersome

Engineering Contradiction:

Improvesearch completenessVSAvoidsearch operation simplicity

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The patent replaces the manual process of sifting through tens of thousands of text descriptors with automated image recognition that directly analyzes visual content. The system processes images based on their actual visual features rather than requiring users to navigate through extensive text-based classification systems

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system extracts only the most relevant visual features from images for searching, eliminating the need to process or consider all possible text descriptors. By extracting key visual characteristics directly from the image content, the system achieves comprehensive coverage with simplified operation

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS11853377B2Image searching method and apparatus

Publication Date: 2023.12.26 SEE OUT PTY LTD

AI summary

Apparatus for performing searching of a plurality of reference images, the apparatus including one or more electronic processing devices that search the plurality of reference images to identify first reference images similar to a sample image, identify image tags associated with at least one of the first reference image, search the plurality of reference images to identify second reference images using at least one of the image tags and provide search results including at least some first and second reference images.