Vehicle Surround View Reconstruction for Camera Shadow Areas
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems are impaired by objects or obstacles in the vehicle's vicinity, which create shadow areas and reduce the quality of the vehicle environment view, leading to distorted projections and impaired assistance functions.
Innovation Solution
A method and device that detect obscured image areas using vehicle cameras, replace visible obscured portions with data from other cameras, and extrapolate non-visible portions based on spatial, temporal, or frequency-selective extrapolation, while adaptively adjusting a 3D projection surface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If vehicle cameras are used to capture the vehicle's surroundings, then image data of the surroundings is obtained, but shadow areas are created by objects or obstacles that reduce the quality of the vehicle environment view
Solution Approach 1:
The patent converts the harmful shadow areas into beneficial information by detecting them and using extrapolation algorithms to reconstruct the obscured regions. The shadow areas, which initially represent lost information, become the target for computational recovery through analyzing surrounding visible areas and temporal sequences, thereby transforming the problem into a solution.
Solution Approach 2:
The patent introduces computational algorithms as intermediaries between the captured image data and the final environment view. These algorithms detect shadow areas, analyze visible surrounding regions, and generate extrapolated image data to fill the obscured areas, acting as a mediator that reconstructs information lost due to shadows.
2Adaptability or versatility
If objects or obstacles are present in the vehicle environment, then the immediate vicinity can be occupied, but the image data quality is impaired and projection surfaces become distorted
Solution Approach 1:
The patent applies dynamic adjustment by continuously monitoring the vehicle environment and adapting the projection surface in real-time. When objects or obstacles are detected, the system dynamically modifies the projection geometry and recalculates the environment view to compensate for distortions, ensuring accurate representation despite changing environmental conditions.
Solution Approach 2:
The patent changes key parameters of the projection system, including projection angles, surface geometry, and image mapping coordinates, to compensate for the presence of objects or obstacles. By adjusting these parameters dynamically, the system maintains accurate projection surfaces even when the environment is occupied by various objects.
3Ease of operation
If conventional driver assistance systems provide support for parking maneuvers, then assistance functions are available, but the functions are impaired when shadow areas are present
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously evaluates the quality of captured image data, detects shadow areas that would impair assistance functions, and triggers compensatory extrapolation processes. This feedback loop ensures that parking assistance and other driver support functions maintain high reliability by actively correcting for shadow-induced information loss.
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method and a device for producing a view of the surroundings (FUA) of a vehicle (F), said method comprising the following steps: supplying (S1) image data of surroundings of the vehicle (F) by means of vehicle cameras (2) provided on the body (3) of the vehicle (F); identifying (S2) image regions of the vehicle surroundings, which are covered by an object (O) in the vehicle surroundings, from the perspective of the vehicle cameras (2); replacing (S3) image sections of an image region (SF) covered by the object (O), which can be seen by at least one other vehicle camera (2), with image data supplied from the other vehicle camera (2); and replacing (S4) image sections of the image region (KSF) covered by the object (O), which cannot be seen by any of the other vehicle cameras (2), with extrapolated image data which is extrapolated on the basis of image data in the environment of the object (O), supplied by the other vehicle cameras (2).