Dynamic Disparity Mapping for Road Anomaly Focused ADAS Vision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced driver assistance systems (ADAS) and autonomous driving systems face challenges in generating high-density disparity maps for accurate obstacle detection and road anomaly recognition without significantly increasing processing load on vehicle processors.
Innovation Solution
A system that dynamically generates disparity maps by prioritizing areas of interest based on historical data and real-time road anomaly information, adjusting bit-depth and image compression levels, and optimizing search ranges to conserve computing resources while maintaining high accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If higher density disparity maps are generated to provide higher accuracy, then measurement precision is improved, but processing load increases
Solution Approach 1:
The patent applies local quality by generating high-density disparity maps only for selected areas of interest (such as regions containing obstacles, road anomalies, or pedestrians) while using lower-density disparity maps for the remaining areas. This selective approach maintains high measurement precision where needed while reducing overall processing load on the vehicle processor.
Solution Approach 2:
The patent segments the road scene into multiple regions and processes each region differently. High-density disparity computation is applied only to critical segments (areas with detected objects or potential hazards), while other segments receive low-density processing. This segmentation strategy resolves the contradiction by localizing high-precision processing to only where necessary.
2Reliability
If higher density disparity maps are generated to improve obstacle detection accuracy, then reliability is improved, but use of energy increases
Solution Approach 1:
The system applies high computational effort and energy consumption only to specific local regions containing obstacles or road anomalies, rather than uniformly processing the entire field of view. This localized high-quality processing maintains reliable obstacle detection where needed while significantly reducing overall energy usage by using lower computational intensity for background areas.
Data Source
AI summary
In some examples, a system may receive, from at least one camera of a vehicle, at least one image including a road. The system may further receive vehicle location information including an indication of a location of the vehicle. In addition, the system may receive at least one of historical information from a historical database, or road anomaly information, where the road anomaly information is determined from at least one of a road anomaly database or real-time road anomaly detection. Based on the at least one image, the indication of the location of the vehicle, and the at least one of the historical information or the road anomaly information, the system may generate at least one of a disparity map or a disparity image.