Projection Image Alignment Using Absolute Pixel Shift Direction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing technologies, such as JP-A-2021-061510, can detect the relative pixel shift between projection images but fail to determine the absolute direction of pixel shift and do not provide a method to correct the pixel shift effectively.

Innovation Solution

A method involving a camera to capture overlapping images from multiple projectors, calculate the relative pixel shift, and adjust the position of the projectors based on the shift direction to correct the pixel shift, using a processing apparatus to move images in opposite directions when the shift exceeds a threshold.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the camera captures images of overlapping projections to detect pixel shift, then the measurement precision of pixel shift is improved, but the system cannot determine the absolute pixel shift direction

Engineering Contradiction:

Improvepixel shift detection precisionVSAvoidabsolute shift direction information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The system performs iterative feedback correction by repeatedly capturing images, calculating pixel shifts, and adjusting projector positions. The correction amount is halved in each iteration until the pixel shift falls within a predetermined range, enabling convergence to the correct absolute alignment state.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system preliminarily determines relative pixel shift direction and magnitude before performing absolute alignment. By first establishing the relationship between projectors and then using iterative feedback, the system recovers absolute position information that was initially lost.

Inventive Principle:

Principle #10Preliminary action

2Area of stationary object

If multiple projectors project images that overlap on the projection surface, then the coverage area is improved, but the alignment precision between overlapping images deteriorates

Engineering Contradiction:

Improveprojection coverage areaVSAvoidimage alignment precision

Core Design Contradiction:

Area of stationary objectVSManufacturing precision

Solution Approach 1:

The system uses feedback control by continuously monitoring pixel shift through camera capture and automatically adjusting projector positions. The iterative correction process with decreasing correction amounts ensures that multiple projectors achieve precise alignment while maintaining their overlapping coverage areas.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

This principle is not applicable to this patent as the solution does not involve chemical oxidation processes.

Inventive Principle:

Principle #38Strong oxidants (Accelerated oxidation)

3Manufacturing precision

If the system adjusts projector positions to correct pixel shift, then the image alignment is improved, but the device complexity increases

Engineering Contradiction:

Improveimage alignment precisionVSAvoidsystem complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system performs self-alignment by automatically detecting pixel shifts through the camera and adjusting projector positions without external intervention. The automated feedback loop with iterative correction reduces the need for manual calibration and complex external alignment equipment.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The camera serves multiple functions: capturing projection images for pixel shift detection, providing feedback for correction, and verifying alignment results. This multi-functionality reduces the need for separate dedicated alignment devices, thereby managing system complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12621416B2Projection image adjustment method, information processing apparatus, and non-transitory computer-readable storage medium storing program

Publication Date: 2026.05.05 SEIKO EPSON CORP

AI summary

A projection image adjustment method includes acquiring first captured image by capturing image of first image projected at first position on projection surface and second image projected at second position on projection surface and at least partially overlapping with first image, acquiring first amount of shift between first image and second image based on first captured image, moving first image from first position to third position separate therefrom in first direction, acquiring second captured image by capturing image of first image projected at third position and second image projected at second position, acquiring second amount of shift between first image and second image based on second captured image, and moving second image from second position to a fourth position separate therefrom in a second direction that is the opposite direction of the first direction when the second amount of shift is smaller than the first amount of shift.