Edge Code Histogram Pattern Matching for Fast Image Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods for pattern matching in Factory Automation require repeated model rotation and pattern matching processes across the entire image, leading to time-consuming searches for matching regions in input images.
Innovation Solution
An image processing method that involves registering edge code histograms for models, setting regions of interest, calculating coincidence between edge code histograms, and evaluating candidate points to efficiently search for matching regions by combining rough and detailed search processes, including rotation angle determination and removal of proximate matches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the model rotation process and pattern matching process are repeated over the entire range of the image to search for multiple patterns, then the position and orientation of patterns can be obtained, but the processing time becomes excessively long
Solution Approach 1:
The patent divides the image into multiple regions of interest (ROIs) and performs pattern matching separately in each ROI. By segmenting the search space, the system avoids processing the entire image uniformly, reducing overall processing time while maintaining detection accuracy in each segment. The edge code histograms are calculated and compared within each ROI independently, enabling parallel processing potential.
Solution Approach 2:
The patent pre-calculates and stores edge code histograms for the model patterns before actual pattern matching. This preliminary preparation of histogram data allows the system to quickly compare candidate regions against pre-processed model representations, eliminating the need to recalculate histograms during the search process and significantly reducing processing time.
2Measurement precision
If edge code histograms are calculated and compared sequentially across the entire image to find matching regions, then accurate pattern matching is achieved, but the search speed is reduced
Solution Approach 1:
The patent segments the image into multiple regions of interest and performs histogram comparison operations within each segment rather than across the entire image. This segmentation approach maintains matching accuracy within each region while enabling faster processing through localized operations that can be executed in parallel or sequentially with reduced computational overhead.
Solution Approach 2:
The patent focuses pattern matching efforts on specific regions of interest rather than uniformly processing the entire image. By identifying and prioritizing certain areas where patterns are more likely to occur, the system performs partial action on the most relevant portions of the image, improving search speed without sacrificing accuracy in critical regions.
3Adaptability or versatility
If the entire image is processed to search for multiple patterns with different orientations, then comprehensive pattern detection is achieved, but computational complexity increases
Solution Approach 1:
The patent divides the complex task of multi-pattern detection into smaller sub-tasks by processing different regions of interest separately. Each ROI is analyzed for pattern matches independently, reducing the computational complexity of handling multiple patterns across the entire image. This segmentation allows the system to manage complexity through modular, region-based processing.
Solution Approach 2:
The patent pre-processes model patterns into histogram representations and stores them for quick reference. This preliminary action of preparing model data in advance simplifies the actual pattern matching process, reducing computational complexity during the search phase. The pre-calculated histograms enable efficient comparison operations without requiring complex real-time calculations.
Data Source
AI summary
An edge code histogram of a model generated in a model image is registered. A target region with respect to the input image is set. An edge code histogram for the target region is generated. A relative positional relationship between the edge code histogram of the model and the edge code histogram for the target region is sequentially changed, and a degree of coincidence between the edge code histograms at each relative position is calculated. A possibility that the region that matches the model is contained in the set target region from the sequentially calculated degree of coincidence between the edge code histograms is evaluated. Then a candidate point having a possibility of matching the model in the input image is specified while sequentially changing the position of the target region with respect to the input image and repeating steps above for each target region.