Image Rectification via Epipolar and Content Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-view image acquisition systems frequently distort images of the left and right eyes, making it impossible to ensure pixel alignment, resulting in visual discomfort for the viewer.
Innovation Solution
A method and device for image rectification that performs epipolar rectification and alignment based on image contents, using feature point pairs and optimized parameters to align images captured from different viewpoints, and splices them to ensure proper alignment and reduce distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If epipolar rectification is performed on images captured from different viewpoints, then pixel alignment between left and right eye images is improved, but image distortion occurs
Solution Approach 1:
The patent applies preliminary action by performing epipolar rectification before final image processing. The rectification process pre-aligns the images from different viewpoints along the epipolar lines, establishing a geometric foundation that facilitates subsequent stereo matching and depth extraction while managing distortion through controlled transformation
Solution Approach 2:
The patent utilizes parameter changes by adjusting rectification parameters such as epipolar angle, baseline distance, and focal length to optimize the balance between pixel alignment and distortion. By varying these parameters, the system can achieve better stereo alignment while minimizing unwanted image deformation
2Measurement precision
If images are rectified based on feature point pairs and optimized parameters, then alignment accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies self-service by using feature points that are inherently present in the image content (such as corners, edges, and distinctive patterns) to drive the rectification process. The algorithm automatically detects and matches these feature points across views, eliminating the need for external calibration targets or manual intervention, thereby achieving high alignment accuracy through the images' own structural information
Solution Approach 2:
The patent implements feedback mechanisms by using optimized parameters derived from feature point matching to iteratively refine the rectification transformation. The system evaluates alignment quality based on matched feature points and adjusts rectification parameters accordingly, creating a closed-loop process that improves alignment accuracy while managing computational resources through adaptive refinement
Data Source
AI summary
The present invention provides a method and a device for image rectification, which are applied in the field of image processing. The method includes: receiving two images, the two images are images of a target object captured from different viewpoints; performing an epipolar rectification on the two images; rectifying the two images after the epipolar rectification based on the image contents; and splicing the two images after the rectification based on the image contents. The method can rectify the images captured from different viewpoints, thereby ensuring the pixel alignment, and avoiding visual discomfort to the viewer.