Vision System Time-Sharing Disparity Block for Optical Flow
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 it impractical and costly for real-world applications.
Innovation Solution
The system employs time-sharing of the disparity calculation block between stereo disparity and optical flow calculations, using correlation search in the horizontal direction to estimate horizontal motion, and performs both calculations in a single dedicated hardware device like an FPGA, with adaptations to handle both positive and negative motion, and includes compensation for vehicle pitch and movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional 2D optical flow calculation is performed, then measurement precision of optical flow is improved, but processing resources and hardware complexity increase significantly
Solution Approach 1:
The patent extracts only the horizontal component of optical flow calculation from the full 2D optical flow computation. By focusing solely on horizontal motion estimation using correlation search in the horizontal direction, the system achieves sufficient accuracy for vehicle vision applications while dramatically reducing processing requirements. This selective extraction of the essential horizontal component resolves the contradiction between measurement precision and device complexity.
Solution Approach 2:
The patent segments the optical flow calculation into two independent parts: horizontal optical flow calculation (performed using the disparity calculation block) and vertical optical flow calculation (performed separately when needed). This segmentation allows the system to use efficient horizontal correlation search for most applications while optionally adding vertical component analysis only when required, thus balancing precision needs with processing resource constraints.
2Device complexity
If time-sharing of the disparity calculation block is implemented, then hardware resource usage is reduced, but calculation time may increase
Solution Approach 1:
The patent implements multi-functionality by enabling the disparity calculation block to serve dual purposes: traditional stereo disparity calculation and horizontal optical flow calculation. The same hardware resource block is time-shared between these two functions, with the disparity calculation block configured to perform correlation search in the horizontal direction for optical flow estimation. This universality significantly reduces hardware resource requirements while maintaining acceptable calculation performance through efficient time-multiplexing.
Solution Approach 2:
The system dynamically switches the disparity calculation block between stereo disparity mode and optical flow mode based on operational requirements. The block can be configured to perform disparity calculation when depth information is needed, and switch to optical flow calculation when motion estimation is required. This dynamic reconfiguration optimizes hardware utilization and resolves the time-resource trade-off by activating only the necessary calculation type at any given time.
Data Source
AI summary
A vision system (10) for a motor vehicle includes a stereo imaging apparatus (11) adapted to capture images from a surrounding of the motor vehicle, a disparity calculation block (17) adapted to calculate a stereo disparity of left/right images captured by the stereo imaging apparatus (11), and a processing device (14) adapted to perform image processing of images captured by the imaging apparatus (11). The vision system is adapted to perform time sharing of the disparity calculation block (17) between the stereo disparity calculation and calculation of a one-dimensional optical flow of captured images in a horizontal direction only.