Projection Display Calibration Using User-Selected Feature Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing projection-type display systems face accuracy deterioration in image calibration due to the inclusion of unnecessary feature points, which can lead to calibration failures and reduced precision.
Innovation Solution
A calibration method that involves acquiring a first image, detecting and superimposing feature points, allowing users to select reference points by designating masking or effective regions, and then using these points to detect deviations in the relative position between the camera and projection area in a second image, thereby excluding unnecessary points and maintaining calibration accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic feature point detection is performed without user selection, then calibration speed is improved, but calibration accuracy deteriorates due to inclusion of unnecessary feature points
Solution Approach 1:
The system performs preliminary automatic detection of multiple candidate feature points, then presents them to the user for selection before final calibration processing. This preliminary action allows the system to prepare all potential candidates in advance while giving the user opportunity to eliminate unnecessary points, thus maintaining both speed and accuracy.
Solution Approach 2:
The user acts as an intermediary between automatic detection and final calibration. The user reviews detected feature points and selects appropriate ones, serving as a mediator that filters out unnecessary points while preserving the efficiency of automatic detection. This intermediary step resolves the contradiction by combining machine speed with human judgment.
2Measurement precision
If user selection of reference points is required, then calibration accuracy is improved, but operation complexity increases
Solution Approach 1:
The system provides self-service by automatically detecting and presenting candidate feature points to the user. Instead of requiring the user to manually search for and select all feature points, the system performs the detection work automatically and only requires user confirmation or selection from pre-detected candidates, thus improving accuracy while minimizing operational complexity.
Solution Approach 2:
The system performs preliminary detection of feature points and prepares a curated list of candidates before user selection. This preliminary action reduces the complexity of user operation by presenting only relevant options rather than requiring the user to search through all possible points, making the selection process simpler while maintaining accuracy.
3Productivity
If all detected feature points are used for calibration, then processing completeness is improved, but calibration accuracy deteriorates due to irrelevant points
Solution Approach 1:
The system extracts and separates necessary feature points from unnecessary ones through user selection. By taking out only the relevant feature points from the complete set of detected points, the system maintains processing completeness (all candidates are detected) while improving calibration accuracy (only relevant points are used).
Solution Approach 2:
The system performs excessive detection by detecting more feature points than will ultimately be used, then allows user selection to reduce to the appropriate subset. This partial use of detected points resolves the contradiction by ensuring no necessary point is missed (completeness) while eliminating irrelevant points (accuracy).
Data Source
AI summary
A calibration method includes: acquiring a first image captured by a camera at a first timing, the first image showing an area on which an image is projected by a projection-type display device; detecting a plurality of feature points in the acquired first image; superimposing the plurality of detected feature points on the first image and displaying the superimposed feature points on a monitor; receiving a user's operation of selecting a part of the plurality of detected feature points as a plurality of reference points; acquiring a second image captured by the camera at a second timing after the first timing, the second image showing the area; and detecting a deviation of a relative position between the camera and the area by using the plurality of selected reference points and the acquired second image.