Position Detection on Curved Surfaces Using Color-Coded Segmented Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing projector technologies face challenges in generating accurate calibration data when projecting images on complex surfaces, such as curved surfaces, due to the inability to detect all dots in a dot pattern, leading to inaccurate image alignment.
Innovation Solution
A method involving the display of pattern images with regions of specific colors on a surface, imaging these patterns, determining the colors of the regions in the images, and using this information to detect identification patterns and associate image and camera positions, even when not all patterns are detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a dot pattern is projected for position detection, then image alignment can be achieved, but accurate calibration data cannot be generated when all dots cannot be detected on curved surfaces
Solution Approach 1:
The identification information is divided into multiple pieces (first partial information and second partial information), with each piece assigned to different regions in the pattern image. This segmentation allows the system to detect position information from multiple independent regions, ensuring that even if some regions are undetected on curved surfaces, other regions can still provide sufficient calibration data.
Solution Approach 2:
Instead of requiring all dots in the dot pattern to be detected for accurate calibration, the invention uses multiple pieces of identification information distributed across different regions. Detection of a sufficient subset of these regions provides excessive information that enables accurate calibration even when some regions are undetected, thus achieving reliable calibration data without complete dot detection.
2Reliability
If multiple pieces of identification information are distributed across multiple regions, then position detection reliability improves, but the complexity of the pattern image increases
Solution Approach 1:
Different pieces of identification information are represented by different colors in the pattern image regions. This color-based encoding allows the system to convey multiple pieces of information through visual color differentiation, making the complex multi-region pattern more manageable and interpretable while maintaining reliability through color-based identification.
Solution Approach 2:
The invention varies parameters such as color, brightness, or other visual attributes across different regions to encode multiple pieces of identification information. By changing these parameters rather than increasing structural complexity, the system maintains pattern image simplicity while embedding multiple information elements that improve position detection reliability.
Data Source
AI summary
A position detection method includes projecting a first pattern image on a display surface, the first pattern image including a first region having a first color corresponding to a first portion of a first identification information indicating the first region and a second region having a second color corresponding to a first portion of a second identification information, obtaining a first captured image by capturing the display surface, projecting a second pattern image onto the display surface including the first region having a third color which corresponds to a second portion of the first identification information and the second region having a fourth color corresponding to a second portion of the second identification information, obtaining a second captured image by capturing the display surface, associating a first position in an image projected with a second position in a third captured image obtained by capturing the display surface.