Blending Mask Width for Seamless Driver Assist Camera Views
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional surround view systems for vehicles face issues with optimal visibility of objects due to the projection of camera images, leading to objects appearing to disappear or be cut off at the overlap areas of camera views, causing confusion for the driver.
Innovation Solution
An image synthesizer that combines images from overlapping camera views using a blending mask with a non-zero minimal width, ensuring objects are visible by interpolating along an interpolation direction within a wide blending region, allowing for complete depiction of traffic objects without cutting them off.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional gradient blending is used to combine overlapping camera views, then smooth transition between views is achieved, but objects may disappear or be cut off at the blending area
Solution Approach 1:
The patent changes the blending parameter from a gradient approach (where blending percentage continuously varies) to a fixed minimum width approach (where the blending region has a defined minimum width). This ensures that the blending area is always sufficient to contain complete objects while maintaining smooth transitions. The blending mask is applied within this fixed minimum width region, preventing objects from being cut off while still achieving seamless view transitions.
2Area of stationary object
If narrow blending is used to reduce the blending area, then sharper transitions are achieved, but objects close to the vehicle may be cut off
Solution Approach 1:
The patent applies excessive action by ensuring the blending region has a minimum width that is wider than absolutely necessary for smooth transitions alone. This excessive width is deliberately chosen to guarantee that even objects close to the vehicle (which project larger in the image) will be fully contained within the blending region and not cut off. The minimum width is determined based on the largest expected object projection.
3Reliability
If wide blending is used to ensure objects are visible, then object completeness is improved, but the blending area becomes too large causing slow transitions
Solution Approach 1:
The patent optimizes the blending region width parameter to find the optimal balance point. The width is set to the minimum value that still guarantees complete object visibility (based on maximum object projection size), rather than using an excessively wide blending region. This minimized sufficient width ensures objects are fully visible while keeping the blending area as small as possible, thereby maintaining fast and smooth transitions.
4Adaptability or versatility
If the blending region width varies with distance from the vehicle, then objects at different distances are handled differently, but the blending complexity increases
Solution Approach 1:
The patent applies local quality by making the blending region width adaptive to local conditions (distance from vehicle and object size). In areas where objects are likely to be larger and closer (near the vehicle), the blending region is wider to ensure complete object containment. In areas where objects are smaller and farther away, the blending region can be narrower. This localized adaptation ensures optimal object visibility throughout the scene without requiring excessive width everywhere.
Data Source
AI summary
An image synthesizer for a vehicle driver assisting system, which includes a first camera to capture a first image from a first scene and a second camera to capture a second image from a second scene, the first and second scenes overlap in an overlapping region, which includes a blending region. The image synthesizer includes: a receiving module to receive the first and second images from the first and second cameras; and a combining module to combine the first and second images in the overlapping region by applying a blending mask. The blending mask is applied to the blending region and provides an interpolation along an interpolation direction between the first and second images depicted on opposite sides of the blending region, which includes a width greater than a minimal width along any path of light received by the first camera or of the second camera from the blending region.