Feature Point Matching Using Local and Global Image Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Feature point matching in images is affected by repetitive patterns, lighting environments, and weather conditions, leading to unstable accuracy in local feature-based matching techniques.
Innovation Solution
An information processing device and method that performs feature point matching using both local features generated from pixels around a feature point and global features derived from entire image pixels, combining these to improve matching accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature point matching is performed based on local features only, then the matching process is simple and fast, but the accuracy is unstable due to repetitive patterns and environmental factors
Solution Approach 1:
The patent combines local features (extracted from regions around feature points) and global features (extracted from the entire image) into a unified matching approach. The matching integration unit processes both feature types and generates matching results based on their combination, thereby improving accuracy while maintaining reasonable computational complexity through efficient feature integration.
2Reliability
If only local features are used for matching, then computational time is reduced, but reliability decreases due to sensitivity to repetitive patterns and lighting conditions
Solution Approach 1:
The system merges local and global feature extraction into a coordinated process where local features provide detailed information about specific regions and global features provide contextual information about the entire image. This combination increases reliability by cross-validating matches against both local and global characteristics, while the processing time remains manageable through optimized feature integration.
3Measurement precision
If local features are extracted from pixels around feature points, then the matching process is efficient, but the accuracy is compromised by repetitive patterns and environmental factors
Solution Approach 1:
The patent introduces global features as an intermediary element that mediates between local features and the final matching decision. The matching integration unit uses global features to disambiguate cases where local features are misleading due to repetitive patterns or environmental factors, thereby improving accuracy without completely discarding the efficient local feature extraction process.
Data Source
AI summary
Accuracy in feature point matching is further improved. There is provided an information processing device including a matching integration unit that performs matching of a feature point detected from an input image, in which the matching integration unit performs the matching of the feature point on the basis of a local feature generated on the basis of a pixel of the feature point and pixels around the feature point and a global feature generated on the basis of entire pixels of the input image.