Logo Detection System for Image Search Engines

Solution Overview

Problem

Image search engines face challenges in efficiently detecting logos within images, leading to high computational resource usage and false positive matches, as logos often contain distinct characteristics that are not effectively differentiated from other image types.

Innovation Solution

A logo detection system evaluates query images based on characteristics such as gray-level entropy, edge crossings, and gradient magnitudes to determine if they are logos, filtering them out before further image search processing to conserve resources and reduce false positives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If image search engines perform image search on all uploaded images including logos, then comprehensive image matching is achieved, but computational and memory resources are wasted and false positive matches increase

Engineering Contradiction:

Improveimage matching accuracyVSAvoidcomputational resource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary detection to identify logos before the image search process. By detecting logos in advance using specific algorithms and filtering them out, the system avoids wasting computational resources on images that would generate false positive matches, thus resolving the contradiction between comprehensive searching and resource efficiency

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system extracts and separates logo images from the general image collection. By identifying logos as a distinct category and removing them from the search index, the system maintains comprehensive search capability for non-logo images while eliminating the harmful effect of logo-related false positives and resource waste

Inventive Principle:

Principle #2Taking out (Extraction)

2Reliability

If image search engines process all images including logos, then complete image catalog maintenance is achieved, but false positive matches occur since logos may be present as elements rather than focus

Engineering Contradiction:

Improvematch accuracyVSAvoidfalse positive matches

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The system performs preliminary detection to identify logos before the image search process. By detecting logos in advance using specific algorithms and filtering them out, the system avoids wasting computational resources on images that would generate false positive matches, thus resolving the contradiction between comprehensive searching and resource efficiency

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system extracts and separates logo images from the general image collection. By identifying logos as a distinct category and removing them from the search index, the system maintains comprehensive search capability for non-logo images while eliminating the harmful effect of logo-related false positives and resource waste

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS10803116B2Logo detection system for automatic image search engines

Publication Date: 2020.10.13 PICSCOUT ISRAEL

AI summary

A system and method that distinguishes between logos and other categories of images, such as natural images, cartoon images, and computer-processed or generated images (“concept images”). The system receives a query image, which may, for example, be intended to be evaluated by an image search and comparison engine to identify matches to a catalog of images. The system evaluates characteristics of the query image, such as the gray-levels in the image, the edge crossings in the image, and the gradient magnitudes in the image, to identify whether the image is a logo. Based on identifying the image is a logo, the logo query image may be excluded from being evaluated by the image search and comparison engine.