U-Disparity Map Drivable Region Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current road detection methods fail to reliably identify drivable regions, especially in environments without white lines or sparse disparity maps, leading to inefficiencies in driving assistance technologies.
Innovation Solution
A method and device that derive a disparity map, convert it into a U-disparity map, and detect drivable regions by removing parts above a predetermined height threshold, using a detection window to identify active points and fit road lines, allowing for accurate detection without relying on white lines or fences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If white lines on both sides of the road are used for detection, then the drivable region can be detected, but the method cannot be applied to roads where white lines do not exist
Solution Approach 1:
The patent extracts and removes parts of the disparity map that correspond to regions above the road (using height threshold filtering). This extraction isolates the relevant road surface information from irrelevant above-road regions, enabling reliable detection without dependence on white lines or fences.
Solution Approach 2:
The patent transforms the 2D disparity map into a 1D U-disparity map by projecting along the vertical dimension. This dimensional transformation consolidates depth information and enables robust drivable region detection even when traditional road markings are absent.
2Reliability
If distance and segmental slope are used for road segmentation, then drivable region can be detected, but the method cannot be applied to sparse disparity maps
Solution Approach 1:
The patent merges multiple detection cues (disparity information, height threshold filtering, U-disparity projection, and active point density analysis) into a unified detection framework. This combination enables reliable drivable region detection even when disparity map points are sparse, as multiple complementary signals reinforce each other.
Solution Approach 2:
The U-disparity map serves as an intermediary representation that consolidates depth information from the disparity map. This intermediate form enables more robust detection by aggregating scattered depth points into continuous drivable region boundaries, even when original disparity data is sparse.
3Adaptability or versatility
If stereovision system and U-disparity map classification are used, then drivable region can be detected, but anti-noise robustness is insufficient
Solution Approach 1:
The patent applies local quality analysis by evaluating active point density within local detection windows across the U-disparity map. By analyzing noise characteristics locally and adapting detection thresholds to local conditions, the system achieves high anti-noise robustness while maintaining detection accuracy.
Solution Approach 2:
The patent incorporates feedback mechanisms where detection results inform subsequent processing stages. The detected drivable region boundaries are refined iteratively by feedback from active point density analysis and height threshold validation, enabling the system to distinguish true signal from noise and achieve high robustness.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and a device are disclosed for detecting a drivable region of a road, the method comprising the steps of: deriving a disparity map from a gray-scale map including the road and detecting the road from the disparity map; removing a part with a height above the road greater than a predetermined height threshold from the disparity map so as to generate a sub-disparity map; converting the sub-disparity map into a U-disparity map; detecting the drivable region from the U-disparity map; and converting the drivable region detected from the U-disparity map into the drivable region within the gray-scale map.