Stereo Camera Convergence Control Using Line Reference Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current stereo camera systems face challenges with image alignment, leading to user dizziness and fatigue due to misalignment between camera convergence and user convergence, and existing calibration methods increase computational load and power consumption, making real-time calibration difficult.
Innovation Solution
A stereo camera system employing a line shifting method for convergence calibration, using a camera unit, filter unit, line memory, and convergence control unit to detect reference lines and calculate image control amounts, thereby aligning both-eyes images and generating an optimal synthesis image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frame memory is used to store and shift both-eyes images for convergence calibration, then convergence alignment can be achieved, but calculation complexity increases and real-time calibration becomes difficult
Solution Approach 1:
The patent divides the image data into individual line units and processes only the reference lines for convergence calibration, rather than processing entire frames. This segmentation approach reduces the volume of data requiring complex calculations while maintaining alignment accuracy.
Solution Approach 2:
The patent extracts only the essential reference line information from the full image data for convergence calibration purposes. By taking out only the necessary components (reference lines) rather than processing complete frames, the calculation complexity is significantly reduced while preserving the ability to achieve accurate convergence alignment.
2Measurement precision
If frame memory is used to store both-eyes images for convergence calibration, then convergence alignment can be achieved, but power consumption increases
Solution Approach 1:
The patent segments image processing to only the necessary reference lines, avoiding the need to load and process entire frames in memory. This reduces the active memory usage and associated power consumption while maintaining convergence calibration accuracy.
Solution Approach 2:
The patent extracts only the essential reference line data from complete image frames, reducing the amount of data that needs to be stored and processed. This extraction approach lowers memory bandwidth requirements and power consumption while preserving convergence alignment capability.
3Measurement precision
If frame memory is used to store both-eyes images for convergence calibration, then convergence alignment can be achieved, but system size increases
Solution Approach 1:
The patent segments the image data into line-level units, allowing convergence calibration to be performed on minimal data (reference lines only). This eliminates the need for large frame memory structures, reducing hardware footprint while maintaining alignment accuracy.
Solution Approach 2:
The patent extracts only the necessary reference line information from complete images, removing the requirement for large-capacity external frame memory. This extraction strategy significantly reduces hardware size while preserving the essential functionality for convergence calibration.
4Measurement precision
If all images are shifted using frame memory for convergence calibration, then convergence alignment can be achieved, but image implementation speed decreases
Solution Approach 1:
The patent segments the calibration process to operate on individual reference lines rather than complete frames. This segmentation enables faster processing by reducing the amount of data that needs to be manipulated, while still achieving accurate convergence alignment.
Solution Approach 2:
The patent extracts only the critical reference line data needed for convergence calibration, eliminating the need to process and shift entire image frames. This extraction approach significantly improves image implementation speed by reducing computational workload while maintaining alignment accuracy.
Data Source
AI summary
Disclosed herein are a stereo camera system and a method for controlling convergence, including: a camera unit photographing both-eyes images; a filter unit filtering signal values of pixels for each line for any one of the both-eyes images along a line direction to detect a reference line of any one image; a line memory unit storing data for the reference line and a reference line of the other one image corresponding to the reference line; and a convergence control unit calculating the image control amount so as to align convergences of the both-eyes images by performing a comparison operation on the data for the reference lines and generating an optimal synthesis image of the both-eyes images by applying the image control amount.