Key Point Extraction Reliability Scoring for Image Processing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing information processing systems for extracting key points from images lack an effective method to assess the reliability of extracted features, leading to inconsistent accuracy and quality in image processing tasks.

Innovation Solution

An information processing apparatus and method that extracts key points from images and calculates an index value, known as an image score, to evaluate the reliability of these key points, using a combination of image conversion processing and comparison between original and converted images, with optional illumination control and image selection based on calculated scores.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If key point extraction is performed without reliability evaluation, then processing speed is maintained, but accuracy and reliability of extracted features deteriorate

Engineering Contradiction:

Improvereliability of key point extractionVSAvoidprocessing speed

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent applies preliminary action by performing image conversion processing before key point extraction to create multiple versions of the image with different characteristics. This allows the extraction unit to select the most suitable version for reliable key point extraction without significantly increasing processing time, as the conversion is done in advance for various image parameters.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback by calculating an image score that evaluates the reliability of extracted key points and uses this score to determine whether to perform additional image conversion processing. The score feedback mechanism allows the system to adaptively adjust processing based on the quality of extracted features, maintaining both reliability and efficiency.

Inventive Principle:

Principle #23Feedback

2Measurement precision

If multiple image conversion processes are performed to improve key point extraction quality, then accuracy improves, but device complexity increases

Engineering Contradiction:

Improveaccuracy of key point extractionVSAvoidcomplexity of image processing system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies partial action by performing image conversion processing selectively rather than exhaustively. The system determines whether to perform conversion based on the calculated image score and only applies conversion when necessary, avoiding unnecessary processing steps and reducing overall system complexity while maintaining adequate accuracy.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The patent implements multi-functionality by designing the image conversion processing unit to handle multiple types of conversions (brightness, contrast, saturation, etc.) through a unified framework. This allows the system to improve key point extraction accuracy by adjusting various image parameters without proportionally increasing system complexity, as the same structural framework handles different conversion types.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12190571B2Processing apparatus, information processing method and recording medium

Publication Date: 2025.01.07 NEC CORP

AI summary

An information processing apparatus (2) includes: an extracting unit (212) that extracts a key point (KP) of a target object as a target key point (KP1) from an input image (IMG1); and a calculating unit (213) that calculates, as a score (SC) of the input image from which the target key point is extracted, an index value related to a reliability of the target key point based on the target key point.