Vehicle Surround View Projection Geometry for Distortion Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image systems for vehicles often distort objects in the vicinity due to fixed virtual projection surfaces, which can compromise safety and user experience, especially in perspective views and adverse conditions.
Innovation Solution
Adjust the geometry of the virtual projection surface based on object data from multiple sensors, including image capture devices and sensors operating in the non-visible spectrum, to reduce distortion and improve realism, using curved or parabolic surfaces that adapt to the vehicle's surroundings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a fixed geometry virtual projection surface is used, then the system is simple and computationally efficient, but objects in the vicinity of the vehicle appear distorted in the resulting image
Solution Approach 1:
The virtual projection surface geometry is made dynamic by adjusting geometrical parameters based on detected object properties (distance, size, shape). This allows the projection surface to adapt to different viewing scenarios, reducing object distortion while maintaining computational efficiency through parameter adjustment rather than complete geometric reconstruction
Solution Approach 2:
Geometrical parameters of the virtual projection surface (such as curvature, scale, and orientation) are modified based on object data from sensors. This parameter adjustment enables the system to maintain accurate object representation without requiring a completely complex irregular geometry, balancing image accuracy with computational efficiency
2Manufacturing precision
If complex or irregular geometries are used for the virtual projection surface, then distortion of objects is reduced, but computational resources increase and the system becomes more complex
Solution Approach 1:
The system uses dynamic adjustment of geometrical parameters based on real-time object detection data, allowing the virtual projection surface to adapt its complexity only when needed. This maintains computational efficiency by avoiding constant recalculation while improving object representation accuracy when objects are present
Solution Approach 2:
By modifying specific geometrical parameters of the virtual projection surface based on object properties, the system achieves better object representation without adopting fully complex irregular geometries. This selective parameter adjustment maintains a balance between accuracy and computational efficiency
3Manufacturing precision
If the virtual projection surface is constantly recalculated for a moving vehicle, then the representation remains accurate, but this may prove distracting to a user viewing the resulting image
Solution Approach 1:
The virtual projection surface is made dynamically adaptable to the moving vehicle's environment by adjusting geometrical parameters based on object detection, rather than constant recalculation. This provides accurate representation while maintaining visual stability for the user
Solution Approach 2:
The system uses feedback from object detection sensors to adjust the virtual projection surface parameters. This feedback mechanism ensures the representation remains accurate by adapting to detected objects while avoiding unnecessary constant recalculation that would distract the user
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Aspects of the present invention relate to a system, method and vehicle for generating a virtual projection surface. Geometrical parameters of the virtual projection surface are determined in dependence on properties of the objects surrounding a vehicle in order to reduce distortion of the objects in a subsequent visual representation of the vehicle surroundings. The system comprises: one or more image capture devices associated with the host vehicle and configured to capture image data indicative of an environment of the host vehicle; a plurality of sensors associated with the host vehicle and configured to capture object data indicative of the presence of at least one object in a vicinity of the host vehicle; and a processor communicatively coupled to the one or more image capture devices and the plurality of sensors to receive the captured image data and captured object data. The processor is configured to: aggregate the object data captured by each of the plurality of sensors; determine, in dependence on the aggregated object data, one or more geometrical parameters of a virtual projection surface; generate a virtual projection surface in dependence on the one or more geometrical parameters; determine, in dependence on the captured image data, an image texture; and map the image texture onto the generated virtual projection surface.