Image Match-Point Detection Using Pattern Numbers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image match-point detection techniques face challenges in accuracy and efficiency due to the need for direct comparison of pixel values, leading to high operational costs and limited detection capabilities, especially when considering camera movements and noise reduction.
Innovation Solution
An image match-point detection apparatus that uses pattern numbers to represent characteristic pixel blocks, allowing for the extraction and comparison of pattern information instead of direct pixel value comparison, thereby reducing operational complexity and increasing accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct comparison of pixel values is performed to detect match points, then detection accuracy can be maintained, but the amount of operation becomes enormous
Solution Approach 1:
The patent divides the image into multiple blocks and performs match-point detection independently in each block. This segmentation reduces the overall operation amount by breaking down the global comparison task into smaller local tasks, while maintaining detection accuracy through systematic processing of all blocks
Solution Approach 2:
The patent extracts characteristic points from each pixel block and uses only these extracted points for comparison, rather than comparing all pixel values. This extraction process significantly reduces the data volume while preserving the essential information needed for accurate match-point detection
2Productivity
If image resolution is downscaled to reduce data comparison amount, then operation amount decreases, but detection accuracy deteriorates due to detection errors in periodic patterns and noises
Solution Approach 1:
By segmenting the image into blocks and processing each block independently, the patent maintains local detail information even when overall data volume is reduced. This approach avoids the accuracy loss that occurs with global downsampling while still reducing the total operation amount through localized processing
Solution Approach 2:
The patent applies different processing approaches to different regions by working with individual pixel blocks. This allows preservation of local characteristics and patterns in each block without requiring full-resolution processing of the entire image, thereby maintaining accuracy while reducing operations
3Measurement precision
If more match points are detected to accurately identify camera movement and shake, then identification accuracy improves, but the amount of operation and device cost increase
Solution Approach 1:
The patent detects match points in each pixel block independently, allowing multiple match points to be found across different blocks without requiring a single complex system. This distributed approach increases the total number of detectable match points while keeping individual block processing simple and cost-effective
Solution Approach 2:
The patent performs detection in multiple pixel blocks, finding more match points than a single-block approach would provide. This excessive action of processing multiple blocks yields sufficient match points for accurate camera movement identification without requiring proportionally increased device complexity
4Reliability
If RRF filter is used for noise reduction and practical use of match-point detection, then detection reliability improves, but the amount of operation increases further
Solution Approach 1:
The patent applies filtering operations to individual pixel blocks rather than the entire image, significantly reducing the total number of operations required for noise reduction. This segmented approach maintains detection reliability by filtering local regions while keeping computational costs manageable
Data Source
AI summary
An image match-point detection apparatus has a pattern detection part, a pattern storage part, a characteristic-point extraction part and a match-point detection part. The match-point detection part determines whether the pattern numbers of the pattern information extracted as the characteristic-point pattern information for the bitmap images are identical to one another and detects a plurality of the pattern information comprising the identical pattern number, as match-point information.