Image Matching Apparatus Using Local Network Segmentation

Solution Overview

Problem

Existing augmented reality applications face challenges in accurately and efficiently identifying devices through image matching due to high processing times and decreased accuracy when large image databases are used, particularly in augmented reality applications that rely on image comparison without location information or markers.

Innovation Solution

An image matching apparatus that narrows down device comparisons by specifying local networks connected to the device, utilizing a storage unit for image data, an obtaining unit for capturing device images, and an image matching unit that performs comparisons against stored data within the specified local network, employing algorithms like SIFT, SURF, and FAST for feature extraction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If image matching is performed against a large image database, then more devices can be identified, but processing time increases and accuracy degrades

Engineering Contradiction:

Improvedevice identification capabilityVSAvoidprocessing time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent segments the image database by organizing images according to local network information. Instead of a single large database, the system divides images into multiple subsets corresponding to different local networks, allowing the matching process to operate on smaller, more manageable segments rather than the entire database at once.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by making the image matching process adaptive to local network characteristics. The system specifies particular local networks for matching based on the captured image's context, thereby concentrating computational resources on relevant local subsets rather than uniformly processing the entire database, which improves both speed and accuracy.

Inventive Principle:

Principle #3Local quality

2Adaptability or versatility

If image matching is performed against a large image database, then more devices can be identified, but matching accuracy decreases

Engineering Contradiction:

Improvedevice identification capabilityVSAvoidimage matching accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent segments the image database by organizing images according to local network information. Instead of a single large database, the system divides images into multiple subsets corresponding to different local networks, allowing the matching process to operate on smaller, more manageable segments rather than the entire database at once.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by making the image matching process adaptive to local network characteristics. The system specifies particular local networks for matching based on the captured image's context, thereby concentrating computational resources on relevant local subsets rather than uniformly processing the entire database, which improves both speed and accuracy.

Inventive Principle:

Principle #3Local quality

3Adaptability or versatility

If comprehensive image matching is performed across all networks, then device identification coverage is maximized, but resource demands increase

Engineering Contradiction:

Improvedevice identification coverageVSAvoidprocessing resource consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent segments the image database by organizing images according to local network information. Instead of a single large database, the system divides images into multiple subsets corresponding to different local networks, allowing the matching process to operate on smaller, more manageable segments rather than the entire database at once.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by making the image matching process adaptive to local network characteristics. The system specifies particular local networks for matching based on the captured image's context, thereby concentrating computational resources on relevant local subsets rather than uniformly processing the entire database, which improves both speed and accuracy.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS9098948B2Image matching apparatus and image matching method

Publication Date: 2015.08.04 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

An image matching apparatus is provided. The apparatus includes a storage unit, an obtaining unit, a specification unit, and an image matching unit. The storage unit is configured to store image data of one or more devices that are connected to a local network. The obtaining unit is configured to obtain image data of device image obtained by capturing a device. The specification unit is configured to specify one or more local networks to be used for image matching. The image matching unit is configured to perform image matching of the obtained image data against the stored image data of one or more devices that are connected to the specified local network.