Stereo Image Rectification via Feature Point Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current stereo image acquisition systems often result in distorted images of the left and right eyes, leading to misalignment of pixels and visual discomfort for the viewer due to frequent distortions in images captured from different viewpoints.
Innovation Solution
A method and device for image rectification that involves determining positions of feature point pairs in two images captured from different viewpoints, using error functions and optimized parameters to align the images, and optionally incorporating regularization terms to adjust rotation-related parameters, ensuring accurate alignment and splicing of images to form a stereogram.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If images are captured from different viewpoints using a multi-view image acquisition device, then the ability to capture stereo images is improved, but image distortion increases causing pixel misalignment
Solution Approach 1:
The patent applies preliminary action by performing image rectification before the images are used for stereo display. The system determines corresponding feature point pairs between left and right images and calculates transformation parameters to rectify the images in advance, ensuring pixel alignment is achieved prior to any rendering or display operations.
Solution Approach 2:
The patent implements feedback through an iterative optimization process. The system defines an error function based on feature point correspondence and transformation parameters, then iteratively adjusts these parameters to minimize the error function, ensuring that the rectification achieves optimal pixel alignment between the two images.
2Manufacturing precision
If feature point pairs are determined to align images, then pixel alignment is improved, but computational complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the image alignment process into distinct stages: first identifying candidate feature point pairs, then filtering these candidates to determine corresponding feature points, and finally using these correspondences to calculate transformation parameters. This segmentation makes the complex computation more manageable and efficient.
Solution Approach 2:
The patent extracts only the essential information needed for alignment by focusing on feature point pairs that have corresponding relationships between the two images. By extracting and processing only these critical correspondences rather than all image data, the system reduces computational complexity while maintaining alignment precision.
3Object-affected harmful factors
If image rectification is performed to ensure pixel alignment, then visual comfort is improved, but processing time increases
Solution Approach 1:
The patent applies partial action by performing rectification only on the necessary portions of the images - specifically the regions containing corresponding feature points and their transformation data. Rather than processing every pixel uniformly, the system focuses computational resources on the critical feature correspondence regions, reducing overall processing time while maintaining visual comfort.
Data Source
AI summary
The present invention provides a method and a device for image rectification, which is applied to the field of image processing. The method includes: receiving two images, wherein the two images are images of a target object captured at different viewpoints, and an image capturing device configured to capture the two images, includes a first image capturing device and a second image capturing device; determining positions of corresponding feature point pairs in the two images; and aligning the two images according to the positions of the feature point pairs and parameters of the first image capturing device and the second image capturing device. The above-mentioned method can rectify the images captured at different viewpoints, thereby ensuring the alignment of pixels, and avoiding visual discomfort of the viewer.