Binocular Optical Flow Rotation Axis Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for estimating a rotation axis and mass center of a spatial target based on binocular optical flows are limited by discretization errors and lack accuracy and stability, especially in reconstructing three-dimensional motion vectors, which affects the detection, tracking, and reconstruction of spatial targets in complex environments.
Innovation Solution
A method that extracts feature points from stereo image pairs, calculates binocular optical flows, removes ineffective areas, reconstructs effective three-dimensional motion vectors, and uses weighted averaging to estimate the rotation axis and mass center, iteratively refining the process to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional binocular optical flow reconstruction methods are used, then the process can be completed with standard algorithms, but the reconstruction accuracy is limited by discretization errors and areas ineffective for reconstructing three-dimensional motion vectors
Solution Approach 1:
The patent applies preliminary action by pre-identifying and removing areas ineffective for reconstruction before performing the main reconstruction process. The method calculates which regions of the binocular optical flow contain insufficient information for accurate 3D motion vector reconstruction, removes these problematic areas in advance, and then performs reconstruction only on the remaining effective areas. This prevents discretization errors from propagating through the reconstruction process and improves both accuracy and stability.
2Measurement precision
If feature points are extracted from stereo image pairs and binocular optical flows are calculated, then motion information can be obtained, but areas with large errors are reconstructed incorrectly affecting overall accuracy
Solution Approach 1:
The patent applies the extraction principle by separating and removing the harmful components (areas ineffective for reconstruction) from the binocular optical flow data before processing. The method identifies regions where discretization errors cause large reconstruction errors and extracts these problematic areas for removal, leaving only the effective areas for accurate trajectory reconstruction. This eliminates the source of errors rather than attempting to correct them later.
3Measurement precision
If iterative reconstruction is performed to improve accuracy, then trajectory reconstruction precision increases, but computational complexity and processing time increase
Solution Approach 1:
The patent reduces computational complexity by performing preliminary removal of ineffective areas before the iterative reconstruction process. By eliminating regions that would contribute errors and require multiple iterations to correct, the method enables faster convergence of the iterative algorithm. The preliminary filtering step reduces the volume of data requiring complex iterative processing while maintaining the benefit of improved accuracy through iteration.
Data Source
AI summary
A method for estimating a rotation axis and a mass center of a spatial target based on binocular optical flows. The method includes: extracting feature points from binocular image sequences sequentially and respectively, and calculating binocular optical flows formed thereby; removing areas ineffective for reconstructing a three-dimensional movement trajectory from the binocular optical flows of the feature points, whereby obtaining effective area-constrained binocular optical flows, and reconstructing a three-dimensional movement trajectory of a spatial target; and removing areas with comparatively large errors in reconstructing three-dimensional motion vectors from the optical flows by multiple iterations, estimating a rotation axis according to a three-dimensional motion vector sequence of each of the feature points obtained thereby, obtaining a spatial equation of an estimated rotation axis by weighted average of estimated results of the feature points, and obtaining spatial coordinates of a mass center of the target according to two estimated rotation axes.