Surround-View Image Artifact Reduction via Local Resolution Adjustment

Solution Overview

Problem

Surround-view systems for vehicles often generate artificial noise due to image artifacts when processing and projecting camera images, particularly in areas close to the vehicle, which affects image quality and requires unnecessary computational resources.

Innovation Solution

The method involves capturing actual camera images, spatially expanding them in a virtual space, identifying and selectively reducing the resolution of specific image areas with artifacts, and generating a virtual camera image using the appropriate resolution to minimize these artifacts, thereby maintaining image quality and conserving resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If the resolution of the entire actual camera image is reduced to minimize image artifacts, then image artifacts are reduced, but image quality in artifact-free areas is degraded

Engineering Contradiction:

Improveimage artifactsVSAvoidimage quality

Core Design Contradiction:

Object-affected harmful factorsVSManufacturing precision

Solution Approach 1:

The patent applies local quality by providing the preselected image area containing artifacts in a reduced resolution while maintaining full resolution for the remaining image areas. This allows resolution adjustment to be localized only where artifacts occur, preventing artifact degradation in artifact-free regions and optimizing the balance between artifact reduction and image quality preservation.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent segments the actual camera image into a preselected image area containing artifacts and remaining image areas without artifacts, applying different resolution processing to each segment. This segmentation enables selective resolution reduction only in the artifact-containing region while preserving quality in other areas.

Inventive Principle:

Principle #1Segmentation

2Power

If the resolution of the entire actual camera image is reduced to reduce computational load, then processing power requirements are reduced, but overall image quality is degraded

Engineering Contradiction:

Improveprocessing powerVSAvoidimage quality

Core Design Contradiction:

PowerVSManufacturing precision

Solution Approach 1:

The patent applies local quality by providing the preselected image area containing artifacts in a reduced resolution while maintaining full resolution for the remaining image areas. This allows resolution adjustment to be localized only where artifacts occur, preventing artifact degradation in artifact-free regions and optimizing the balance between artifact reduction and image quality preservation.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent applies partial action by reducing resolution only for the preselected image area containing artifacts rather than the entire image. This partial resolution reduction decreases computational load compared to processing the full image at reduced resolution, while maintaining necessary quality in artifact-free regions.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If actual images are used as textures of the projection surface, then the surround-view system can be implemented, but artificial noise is generated during image processing

Engineering Contradiction:

Improvesurround-view display capabilityVSAvoidartificial noise

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent applies local quality by providing the preselected image area containing artifacts in a reduced resolution while maintaining full resolution for the remaining image areas. This allows resolution adjustment to be localized only where artifacts occur, preventing artifact degradation in artifact-free regions and optimizing the balance between artifact reduction and image quality preservation.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent converts the harmful effect of using actual camera images as textures (which generates artifacts) into a benefit by selectively reducing resolution only in artifact-prone areas. This selective processing eliminates the harmful artifacts while preserving the useful surround-view display capability provided by using actual images as textures.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentEP3073446B1Method for representing the surroundings of a vehicle

Publication Date: 2019.11.20 ROBERT BOSCH GMBH

Figure 1
Figure 2
Figure 3~4

AI summary

The present invention relates to a method for displaying the vehicle environment of a vehicle (1), comprising the steps of capturing at least one actual camera image (4) by means of at least one camera (2a, 2b, 2c, 2d) arranged on the vehicle, wherein the actual camera image (4) represents the vehicle environment, spatially spanning the at least one actual camera image (4) in a virtual space (6), providing a preselected image area (3) of the actual camera image (4) in which image artifacts occur, in several resolutions, and generating a virtual camera image (7) from the perspective of a virtual camera (5) at a position in the virtual space (6), which represents the spanned actual camera image (4), wherein for the preselected image area (3) of the actual camera image, the resolution provided is used which reduces the optical presence of the image artifacts.