Common View Region Scale Alignment for Faster Image Feature Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods for matching feature points in images suffer from low efficiency due to the sequential processing of each feature point, leading to suboptimal processing rates and overall inefficiency.
Innovation Solution
An image processing method that includes acquiring an image pair, extracting image features, identifying common view regions, calculating scale differences, and adjusting image sizes to facilitate efficient feature point extraction and processing based on common view features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single-point matching is performed on feature points step by step, then feature point matching can be achieved, but processing efficiency is low
Solution Approach 1:
The patent segments the image processing task by first identifying common view regions between image pairs, then performing feature extraction and matching only within these segmented regions rather than processing entire images. This segmentation approach reduces the search space and computational burden while maintaining matching accuracy
Solution Approach 2:
The patent performs preliminary actions by first estimating consistent scale regions and identifying common view areas before conducting feature point matching. This preliminary region identification prepares the data structure and reduces the scope of subsequent matching operations, improving overall processing efficiency
2Measurement precision
If feature points are processed sequentially, then accurate matching can be achieved, but processing rate is low
Solution Approach 1:
The patent merges multiple processing operations by combining region estimation, common view identification, and feature matching into an integrated workflow. By merging these operations and processing them in a coordinated manner within common view regions, the system reduces overall processing time while maintaining accuracy
3Productivity
If scale alignment is not performed, then processing speed may be faster, but feature matching accuracy decreases
Solution Approach 1:
The patent applies local quality by performing scale alignment specifically within identified common view regions rather than uniformly across entire images. This localized approach ensures matching precision is maintained where it matters most (in overlapping regions) while avoiding unnecessary processing in non-overlapping areas
Data Source
AI summary
Embodiments of this application disclose an image processing method and apparatus, and a computer-readable storage medium. The method includes acquiring an image pair to be processed, and performing image feature extraction on an image to be processed in the image pair to be processed to obtain an image; extracting an association feature from the image feature, the association feature characterizing mutual information between images in the image pair to be processed; identifying a common view image of a common view region in the image to be processed according to the association feature, and calculating a scale difference between common view images; adjusting a size of the common view image based on the scale difference to obtain an adjusted common view image; and extracting at least one common view feature point from each of the adjusted common view images, and processing the image based on the common view feature point.