Image Matching Device Candidate Region Homography Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image matching devices require numerous complex operations, particularly when processing multiple camera moving images in real time, leading to increased processing time due to the large number of similarity operations needed for feature point matching.
Innovation Solution
The solution involves setting a matching candidate region in the target image based on a reference point transformed using a homography matrix, reducing the number of similarity operations by focusing on a specific area proportional to camera viewpoint changes, and generating a homography matrix to perform warping and correction operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature point matching is performed across the entire target image, then matching accuracy is improved, but processing time increases significantly
Solution Approach 1:
The patent divides the target image into multiple candidate regions based on feature point locations, rather than performing matching across the entire image. This segmentation reduces the search space from the full image area to specific local regions, significantly decreasing the number of similarity operations required while maintaining matching accuracy within each region.
Solution Approach 2:
The patent applies different processing strategies to different regions of the image. Candidate regions are identified and processed with focused attention, while other regions are excluded from exhaustive search. This local quality approach concentrates computational resources on areas most likely to contain matches, improving efficiency without sacrificing accuracy.
2Reliability
If similarity operations are performed between feature points from reference image and all feature points in target image, then comprehensive matching is achieved, but the number of operations increases
Solution Approach 1:
The patent performs preliminary identification of candidate regions before conducting similarity operations. By pre-defining where matches are likely to occur based on feature point distribution and image characteristics, the system eliminates the need to perform exhaustive comparisons across the entire image, reducing operational complexity while maintaining matching completeness.
Solution Approach 2:
The patent performs similarity operations only on partial sets of feature points - specifically, only those within identified candidate regions. This partial action approach is sufficient to achieve reliable matching results without the excessive computational burden of comparing all possible feature point pairs across the entire image.
Data Source
AI summary
Provided is an image matching method of matching at least two images. The image matching method extracts feature points of a reference image and feature points of a target image, changes a feature point, selected from among the feature points of the reference image, to a reference point in the target image, sets a matching candidate region on the basis of the reference point, in the target image, and performs a similarity operation between the selected feature point in the reference image and a plurality of feature points included in the matching candidate region among the feature points of the target image. The image matching method decreases the number of similarity operations performed in the image matching operation, thereby guaranteeing a high-speed operation.