Pattern Image Detection with Corrective Feedback for Projector Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing projector calibration methods face accuracy issues when reference points in a pattern image cannot be detected, leading to low precision in position detection.
Innovation Solution
A method involving a display apparatus to display a first pattern image, capture it with a camera to generate a first captured image, perform initial detection, apply corrections to increase detection success, and repeat corrections until successful detection of a second pattern image is achieved.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the pattern image detection is performed directly from the captured image without correction, then the detection process is simple and fast, but the detection accuracy is low when reference points cannot be detected
Solution Approach 1:
The patent applies preliminary action by performing correction processing on the captured image before pattern detection. Specifically, when reference points cannot be detected initially, the system corrects the captured image by adjusting brightness, contrast, or applying filtering operations to enhance the visibility of the pattern, then performs detection on the corrected image. This preliminary correction step ensures that the detection can proceed successfully with improved accuracy.
Solution Approach 2:
The patent implements feedback by using the detection result to determine whether correction processing is needed. The system first attempts detection on the original captured image, and based on whether reference points are successfully detected, it feedbacks to the correction module to apply appropriate corrections. This closed-loop feedback mechanism ensures correction is applied only when necessary, balancing accuracy improvement with processing efficiency.
2Reliability
If corrections are applied to the captured image to improve detection success rate, then the detection accuracy improves, but the processing time increases
Solution Approach 1:
The patent applies partial action by implementing correction processing only when detection fails on the original captured image. The system first attempts detection without correction, and only when reference points cannot be detected does it proceed to apply correction operations. This conditional approach avoids unnecessary processing time while ensuring detection success when needed.
Solution Approach 2:
The patent changes image parameters such as brightness, contrast, and noise levels through correction processing. By adjusting these parameters on the captured image, the system enhances the detectability of reference points and pattern features, thereby improving detection success rate without requiring complete re-capture of the image.
3Reliability
If multiple correction operations are performed iteratively to ensure detection success, then the detection reliability improves, but the computational complexity increases
Solution Approach 1:
The patent performs preliminary correction operations that are designed to be computationally efficient, such as brightness and contrast adjustment followed by simple filtering operations. These preliminary corrections are sufficient to enhance most patterns without requiring complex iterative processing, thus maintaining low computational complexity while improving detection reliability.
Solution Approach 2:
The system uses feedback from the detection result to control the number and type of correction operations performed. When detection succeeds on the first attempt, no corrections are applied. When detection fails, correction is applied once or twice with simple operations. This feedback-controlled approach ensures detection reliability while minimizing computational complexity by avoiding unnecessary multiple iterations.
Data Source
AI summary
A method for detecting a pattern image includes causing a projector to display a first pattern image, acquiring a first captured image that is an image, captured with a camera, of the first pattern image, performing detection of a second pattern image corresponding to the first pattern image from the first captured image, when the detection of the second pattern image from the first captured image fails, performing first correction that increases possibility of successful detection of the second pattern image on the first captured image to acquire a second captured image, and performs detection of the second pattern image from the second captured image.