Filtering Device for Parallax Validation in Environment Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing collision avoidance and cruise control systems face challenges in accurately determining valid parallax values due to similar patterns in images, leading to erroneous derivations and failure to effectively exclude invalid parallax values under varying environmental conditions.
Innovation Solution
A filtering device and environment recognition system that evaluates correlations between image blocks using an evaluation value deriving module, sets an evaluation range based on the highest correlation value, and determines the validity of parallax values using a difference value determining module, which considers the ratio or area of evaluation values within this range to exclude invalid parallax values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pattern matching is used to calculate parallax by comparing image blocks, then relative distance information can be acquired, but erroneous parallaxes are derived when similar patterns continue horizontally in the images
Solution Approach 1:
The patent applies preliminary action by evaluating multiple candidate parallaxes in advance before final selection. The system calculates correlation values for multiple candidate blocks, evaluates their validity using predetermined conditions, and selects the most reliable parallax from valid candidates, thereby preventing erroneous parallaxes from being used in distance calculations.
Solution Approach 2:
The patent implements feedback by using correlation values to evaluate the validity of candidate parallaxes. The system calculates correlation between reference blocks and comparison blocks, uses these correlation values to determine whether candidate parallaxes are valid based on predetermined conditions, and iteratively selects the best parallax candidate, ensuring reliable parallax derivation even in challenging imaging conditions.
2Adaptability or versatility
If fixed threshold values are used to evaluate parallax validity, then the evaluation process is simple, but the system cannot adapt to varying environmental conditions such as fogging or brightness changes
Solution Approach 1:
The patent applies dynamics by making the evaluation process adaptive to varying environmental conditions. Instead of using fixed threshold values, the system dynamically evaluates candidate parallaxes based on correlation values calculated from actual image data, allowing the validity criteria to adjust automatically to different imaging conditions such as fog, brightness variations, or pattern similarities.
Solution Approach 2:
The patent implements parameter changes by using correlation values as dynamic evaluation parameters rather than fixed thresholds. The system calculates correlation values between image blocks and uses these parameter values to determine parallax validity, allowing the evaluation criteria to change based on actual image characteristics and environmental conditions, thereby improving adaptability.
3Reliability
If average value difference matching is used to reduce noise effects, then low-frequency noise is suppressed, but evaluation values still greatly vary under different imaging conditions
Solution Approach 1:
The patent applies preliminary action by performing multiple evaluation steps before final parallax determination. The system first calculates correlation values, then evaluates candidate parallaxes against predetermined conditions, and finally selects valid candidates. This multi-stage preliminary evaluation ensures that noise-resistant matching is combined with adaptive validity checking, improving both reliability and measurement precision.
Solution Approach 2:
The patent implements feedback by using correlation values to continuously evaluate parallax validity. The system calculates correlation between image blocks, uses these values to determine whether candidate parallaxes are valid, and adjusts selection based on this feedback. This feedback mechanism maintains noise resistance while improving evaluation value consistency across different imaging conditions.
Data Source
AI summary
A filtering device includes an evaluation value deriving module that derives, for a pair of images having mutual relevance, multiple evaluation values indicative of correlations between any one of blocks (reference block) that is extracted from one of the images (reference image) and multiple blocks (comparison blocks) extracted from the other image (comparison image), an evaluation range setting module that sets an evaluation range of the evaluation values, the evaluation range having one of boundaries at the evaluation value with the highest correlation among the multiple evaluation values, and a difference value determining module that determines whether the evaluation value with the highest correlation is valid as a difference value based on the multiple evaluation values and the evaluation range.