Vision System Optical Flow Calculation via Time-Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vision systems for motor vehicles face challenges in calculating optical flow due to high processing and hardware resource demands, making real-time two-dimensional optical flow calculation impractical and costly.
Innovation Solution
The system calculates one-dimensional optical flow in the vertical direction by time-sharing the disparity calculation block, using the stereo disparity algorithm with minor adaptations, and combines it with horizontal optical flow to approximate two-dimensional optical flow, reducing processing resources and hardware costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If two-dimensional optical flow is calculated in one optimization pass, then measurement precision is improved, but device complexity and processing resources increase significantly
Solution Approach 1:
The patent segments the two-dimensional optical flow calculation into two separate one-dimensional calculations: first calculating horizontal optical flow, then calculating vertical optical flow. This segmentation reduces the search space from O(n²) to O(n), significantly lowering processing requirements while maintaining measurement precision through sequential refinement.
Solution Approach 2:
The patent performs preliminary horizontal optical flow calculation first, using the results as an initial estimate for the subsequent vertical optical flow calculation. This preliminary action provides a starting point that reduces the search space and computational burden of the second calculation pass.
2Measurement precision
If stereo disparity calculation block is dedicated solely to stereo disparity calculation, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent makes the stereo disparity calculation block multi-functional by enabling it to perform both stereo disparity calculation and one-dimensional optical flow calculation (both horizontal and vertical components). This universality eliminates the need for separate dedicated hardware blocks, reducing overall device complexity while maintaining calculation precision through algorithmic adaptations.
3Device complexity
If time sharing is implemented between stereo disparity calculation and optical flow calculation, then device complexity is reduced, but productivity decreases
Solution Approach 1:
The patent implements periodic time-sharing operation where the shared disparity calculation block alternates between stereo disparity calculation and optical flow calculation tasks. By carefully managing the timing and prioritization of these periodic operations, the system achieves real-time processing capability while using a single hardware block for multiple functions.
Solution Approach 2:
The patent ensures continuous useful action by overlapping computation and data preparation phases. While the disparity calculation block is busy with one task, other processing units prepare data for the next task, ensuring that the system maintains real-time processing throughput without idle periods that would reduce productivity.
Data Source
Figure 1
Figure 2~3
Figure 4~6
AI summary
A vision system (10) for a motor vehicle comprises an imaging apparatus (11) adapted to capture images from a surrounding of the motor vehicle, a correlation search engine (17) adapted to perform a horizontal correlation search between images captured by said imaging apparatus (11), and a processing device (14) adapted to perform image processing of images captured by said imaging apparatus (11). The vision system is adapted to perform time sharing of said of said correlation search engine (17) between said horizontal correlation search and calculation of a one-dimensional optical flow of captured images in a vertical direction.