Projector Image Corner Correction Using Feature Point Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing projection systems require time-consuming manual adjustments of projector position, orientation, and angle to correct image shape on a projection surface, especially when the desired shape differs from the initial display.
Innovation Solution
A method and program that utilize a projector to display a projection image with defined corners, detect feature points, and allow user-driven corrections through pointer and instruction images, followed by a projective transformation to maintain the desired image shape and position despite changes in projector setup.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual adjustment of projector position, orientation, and angle is performed to correct image shape, then image shape accuracy is improved, but adjustment time increases
Solution Approach 1:
The patent replaces manual mechanical adjustment of the projector with an automated image processing system. The processor detects feature points in the captured image, calculates transformation parameters, and automatically adjusts the projection image shape through software-based projective transformation, eliminating the need for time-consuming manual positioning and angle adjustment while maintaining high shape accuracy
Solution Approach 2:
The system performs self-correction by automatically detecting the display area boundaries through feature point detection, calculating the necessary transformation parameters, and applying the correction without user intervention. The processor autonomously completes the entire adjustment process from detection to correction, making the system self-sufficient and eliminating manual adjustment steps
2Ease of operation
If pointer image is displayed for user operation to adjust projection image shape, then ease of operation is improved, but device complexity increases
Solution Approach 1:
The patent introduces a pointer image as an intermediary visual aid between the user and the projection system. The pointer image displays the current shape adjustment state and allows users to provide input through simple operations, making the complex image transformation process intuitive and easy to control while the underlying complex calculations are handled automatically by the processor
3Stability of the object's composition
If projective transformation is applied to maintain positional relationship, then image position consistency is improved, but calculation complexity increases
Solution Approach 1:
The patent performs preliminary detection of feature points and calculation of transformation parameters before applying the projective transformation. By pre-calculating the transformation matrix based on detected corner positions and desired display area boundaries, the system ensures consistent positional relationships are maintained while the actual transformation is simply applied, separating the complex calculation phase from the execution phase
Data Source
AI summary
A projection method includes: displaying, by controlling a projection apparatus, a projection image having four corners inside a display area having four corners, the display area being an area on a projection surface; acquiring a first captured image that is a captured image of a range including the display area where the projection image is displayed; detecting, by performing image processing on the first captured image, a plurality of first feature points corresponding one-to-one to each of four corners of a first image indicating the display area in the first captured image; displaying, by controlling the projection apparatus, a pointer image for adjusting a shape of the projection image, the pointer image being located at a corner of the projection image; performing a first correction of correcting the shape of the projection image by changing a position of the corner of the projection image based on a user operation for moving the pointer image; and performing, based on the plurality of first feature points, a second correction for maintaining a positional relationship between the four corners of the display area and the four corners of the projection image determined by the first correction.