Marker Detection Using Contour-Based Pattern Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image recognition methods for detecting markers in images suffer from low accuracy and high processing load, especially in varying lighting conditions and partial occlusions, and fail to accurately determine the position of the marker.
Innovation Solution
A system that includes image acquisition, detection, and pattern matching processes to identify contours and vertices in images, with additional determination steps for accurate detection and position correction, using a computer-readable storage medium to store detection position information for improved marker detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional binarization with fixed threshold is used for marker detection, then the processing is simple, but the detection accuracy deteriorates in varying lighting conditions and partial occlusions
Solution Approach 1:
The patent performs contour detection and vertex detection before pattern matching to identify the marker's position and orientation. This preliminary action allows the system to adapt the pattern matching process to the actual marker state, improving detection accuracy under varying lighting conditions and partial occlusions while maintaining processing efficiency.
Solution Approach 2:
The patent dynamically adjusts the detection strategy based on whether contours and vertices are successfully detected. When contours/vertices are detected, pattern matching is performed within the identified area; when not detected, full-image pattern matching is performed. This dynamic adaptation resolves the contradiction between processing simplicity and detection accuracy.
2Reliability
If pattern matching is performed on the entire image when contours cannot be detected, then detection coverage is improved, but processing load increases
Solution Approach 1:
The patent applies partial action by performing pattern matching only within the area surrounded by detected contours when available, rather than scanning the entire image. This reduces processing load while maintaining detection coverage. When contours are not detected, full-image matching is performed to ensure detection coverage, resolving the contradiction between reliability and productivity.
Solution Approach 2:
The contour detection and vertex detection serve as preliminary actions that define the search area for pattern matching. This preliminary segmentation of the image into relevant and irrelevant areas allows the system to reduce processing load by focusing only on areas containing potential markers, while maintaining reliable detection coverage.
3Measurement precision
If multiple determination steps are added for accurate marker detection, then detection accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the detection process into distinct functional modules: contour detection, vertex detection, area identification, and pattern matching. Each module performs a specific function, making the complex multi-step process more manageable and maintainable. This segmentation allows high detection accuracy through multiple determination steps while controlling device complexity through modular design.
4Productivity
If detection position information is stored and reused, then processing load for subsequent detections is reduced, but position deviation may accumulate
Solution Approach 1:
The patent uses feedback by comparing the detected marker position with the stored detection position information to identify and correct deviations. The system performs pattern matching based on both the current image and stored position information, then updates the stored information with corrected positions. This feedback mechanism reduces processing load through intelligent search area selection while preventing position deviation accumulation.
Data Source
AI summary
First, it is determined whether or not a contour or vertices have been detected from a certain image. Then, when it is determined that a contour or vertices have been detected from the certain image, a pattern matching process is performed on, in the image, an area surrounded by the contour or the vertices detected from the image. Then, when it is determined that a predetermined object or a predetermined design is displayed in the area, detection position information indicating a position related to the area is stored in a storage area. On the other hand, when it is determined that a contour or vertices have not been detected from the certain image, a pattern matching process is performed on the image on the basis of the detection position information stored in the storage area.