Multi-Projection Alignment via Overlap Region Vertex Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-projection systems fail to effectively adjust for positional misalignment between overlapping projection images due to trapezoidal distortion caused by tilted projection apparatuses, making it difficult to superimpose overlap regions accurately.
Innovation Solution
A projection type image display apparatus and method that includes a transformation unit for geometrical transformation of projection images and a setting unit to adjust parameters based on the position of overlap regions before and after transformation, allowing for precise alignment by changing the amount of transformation according to the position of the overlap region's vertex.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multi-projection is performed by combining multiple projection images, then the display area is expanded, but positional misalignment between overlap regions occurs due to trapezoidal distortion
Solution Approach 1:
The patent applies geometrical transformation parameters (translation, rotation, scaling) to the projection images to correct trapezoidal distortion and achieve precise alignment of overlap regions. The setting unit adjusts these parameters based on the positional relationship between markers in overlapping regions, transforming the distorted images into properly aligned projections that maintain both expanded display area and high alignment precision.
2Manufacturing precision
If geometrical transformation is applied to correct trapezoidal distortion, then alignment precision is improved, but the complexity of the adjustment process increases
Solution Approach 1:
The system performs self-adjustment by automatically detecting marker positions in overlap regions and computing the appropriate geometrical transformation parameters. The setting unit autonomously adjusts the transformation parameters based on the detected positional misalignment, eliminating the need for manual calibration and simplifying the adjustment process while maintaining high alignment precision.
Solution Approach 2:
The system uses feedback from marker position detection in overlap regions to iteratively refine the geometrical transformation parameters. By continuously monitoring the alignment of markers and adjusting the transformation based on the detected positional relationships, the system achieves precise alignment while keeping the adjustment process automated and manageable.
3Ease of operation
If simple marker superposition is used for adjustment, then the adjustment process is simplified, but trapezoidal distortion caused by tilted projection apparatuses is not accounted for
Solution Approach 1:
The patent enhances the simple marker superposition method by incorporating geometrical transformation parameters that account for trapezoidal distortion. Instead of merely superimposing markers, the system applies translation, rotation, and scaling transformations to the projection images based on the detected marker positions, thereby correcting the distortion while maintaining ease of operation through automated parameter adjustment.
Data Source
AI summary
An image display apparatus betters adjustment precision in the multi-projection in which two projection images are combined so that the images partially overlap each other. A display apparatus configuring an image projection system configured to project an image by combining on a screen a plurality of projection images so that the images partially overlap each other, includes a transformation unit configured to perform a geometrical transformation on a projection image, and a setting unit configured to set a parameter of the transformation by the transformation unit based on a relation between a position of an overlap region in a projection image obtained before the transformation by the transformation unit is performed, and a position of the overlap region in the projection image obtained after the transformation is performed, wherein an amount or value of the transformation is changed according to the position of a vertex of the overlap region.